<%BANNER%>

Effects of fire on nutrient availability and limitation in Florida scrub ecosystems

Permanent Link: http://ufdc.ufl.edu/UFE0041437/00001

Material Information

Title: Effects of fire on nutrient availability and limitation in Florida scrub ecosystems
Physical Description: 1 online resource (257 p.)
Language: english
Creator: Schafer, Jennifer
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: archbold, flatwoods, nitrogen, phosphorus, quercus, serenoa
Botany -- Dissertations, Academic -- UF
Genre: Botany thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Effects of fire on nutrient availability and limitation in Florida scrub ecosystems Nitrogen (N) and phosphorus (P) are essential plant nutrients that limit productivity in most, if not all, terrestrial ecosystems. Fire, a common disturbance in many shrublands, may have different effects on the relative availability of N and P because N volatilization occurs at lower temperatures than P volatilization. I investigated the short- and long-term effects of fire on soil and plant nutrients in Florida scrub ecosystems. In addition, I tested the hypothesis that nutrient limitation of plant productivity of scrubby flatwoods changes with time after fire. In flatwoods ecosystems, fire caused a greater increase in phosphate (PO43-) than ammonium (NH4+), resulting in a decrease in the soil available N:P ratio shortly after fire. Similarly, foliar %P of resprouting species increased more than foliar %N, resulting in a decrease in foliar N:P ratios shortly after fire. In scrubby flatwoods, PO43-, but not total inorganic N, varied with time after fire, causing N:P ratios to be greatest at intermediate times after fire and lowest 13 years after fire. In surface soils, soil %C and %N, dissolved organic N, net N mineralization, and microbial N were all highest 13 years after fire, and measures of N availability were often highly correlated. In recently burned scrubby flatwoods, shrubs appear to invest more in aboveground productivity, and Quercus inopina responded to P and N + P addition, but Serenoa repens responded to N addition. At intermediate times after fire, shrubs appear to invest more in belowground than aboveground productivity and show co-limitation by N and P with a stronger P-limitation, while in long unburned sites, scrubby flatwoods shrubs appear to invest in both aboveground and belowground productivity and show co-limitation by N and P. Overall, my research suggests that: (1) increased soil nutrient availability and reallocation of nutrients from below- to aboveground can be important for plant nutrient status after fire; (2) species composition and fire frequency, as well as time after fire, are important in affecting soil nutrient availability; and (3) the effects of nutrient addition on biomass and growth depend on time after fire and species identity.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jennifer Schafer.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Mack, Michelle C.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041437:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041437/00001

Material Information

Title: Effects of fire on nutrient availability and limitation in Florida scrub ecosystems
Physical Description: 1 online resource (257 p.)
Language: english
Creator: Schafer, Jennifer
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: archbold, flatwoods, nitrogen, phosphorus, quercus, serenoa
Botany -- Dissertations, Academic -- UF
Genre: Botany thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Effects of fire on nutrient availability and limitation in Florida scrub ecosystems Nitrogen (N) and phosphorus (P) are essential plant nutrients that limit productivity in most, if not all, terrestrial ecosystems. Fire, a common disturbance in many shrublands, may have different effects on the relative availability of N and P because N volatilization occurs at lower temperatures than P volatilization. I investigated the short- and long-term effects of fire on soil and plant nutrients in Florida scrub ecosystems. In addition, I tested the hypothesis that nutrient limitation of plant productivity of scrubby flatwoods changes with time after fire. In flatwoods ecosystems, fire caused a greater increase in phosphate (PO43-) than ammonium (NH4+), resulting in a decrease in the soil available N:P ratio shortly after fire. Similarly, foliar %P of resprouting species increased more than foliar %N, resulting in a decrease in foliar N:P ratios shortly after fire. In scrubby flatwoods, PO43-, but not total inorganic N, varied with time after fire, causing N:P ratios to be greatest at intermediate times after fire and lowest 13 years after fire. In surface soils, soil %C and %N, dissolved organic N, net N mineralization, and microbial N were all highest 13 years after fire, and measures of N availability were often highly correlated. In recently burned scrubby flatwoods, shrubs appear to invest more in aboveground productivity, and Quercus inopina responded to P and N + P addition, but Serenoa repens responded to N addition. At intermediate times after fire, shrubs appear to invest more in belowground than aboveground productivity and show co-limitation by N and P with a stronger P-limitation, while in long unburned sites, scrubby flatwoods shrubs appear to invest in both aboveground and belowground productivity and show co-limitation by N and P. Overall, my research suggests that: (1) increased soil nutrient availability and reallocation of nutrients from below- to aboveground can be important for plant nutrient status after fire; (2) species composition and fire frequency, as well as time after fire, are important in affecting soil nutrient availability; and (3) the effects of nutrient addition on biomass and growth depend on time after fire and species identity.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jennifer Schafer.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Mack, Michelle C.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041437:00001


This item has the following downloads:


Full Text

PAGE 1

1 EFFECTS OF FIRE ON NUTRIENT AVAILABILITY AND LIMITATION IN FLORIDA SCRUB ECOSYSTEMS By JENNIFER LYNN SCHAFER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUI REMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Jennifer Lynn Schafer

PAGE 3

3 To my Mom, who has provided constant support, and my grandfather, Dr. George Schafer, who passed away before I decided to follow in his foot steps

PAGE 4

4 ACKNOWLEDGMENTS I thank my a dvisor Michelle Mack for her guidance and support. I thank my committee members Emilio Bruna, Nick Comerford, Doria Gordon, and Ted Schuur for their advice and comments I thank my office mates Silvia Alvarez Clare, Je nnie DeMarco, and Caitlin Hicks Pries for their scientific discussions and emotional support. I thank Julia Reiskind for her help in the lab and Grace Crummer for assistance processing samples on the elemental analyzer. Other members of the Mack and Sch uur labs, including Heather Alexander, Fay Belshe, Catherine Cardels, Martin Lavoi e, Hanna Lee, Jordan Mayor, Laura Schreeg, and Eddie Watkins, have provided valuable input on a variety of aspects of my research. Participants in the Plant and Ecosystem Ec ology Research Symposium (PEERS), including Kaoru Kitajima and Jack Putz, have also given valuable input. The office staff in the Botany and Biology departments at UF, particularly Tangelyn Mitchell, Karen Patterson, and Kim Williams, were very helpful. I n addition, I thank Terrell Bostic, James Lange, Natasha Johnson, Olivia Ellen Martin, Nicole Motzer Jennifer Petriella, Jennifer Tucker, Mirela Vasconcelos, and Olivia Vasquez for help with lab work at the University of Florida. I thank the Plant ecology lab at Archbold Biological Station, especially Eric Menges and Carl Weekley for making me always feel welcome in the lab and for providing logistical and personell support I thank Gretel Clarke, Sarah Haller, Marcia Rickey, and Stacy Smity, research ass istants in the Plant Ecology lab during my time there, who were a great help and have become great friends. I thank Kevin Main and all others who have conducted prescribed fires at Archbold, for without them, my research wou ld not have been possible I tha nk Roberta Pickert and Kye Ewing for making maps of the fire history in all burn units. I thank Hilary Swain, the director of Archbol Biological

PAGE 5

5 Station, for allowing me to conduct my research there. I thank Silvia Alvarez Clare, Jeremy Ash, Craig Beatty, Rachel Burnett, Kaitlynn Earnshaw, Megan Larson, Hanna Lee, Jennifer Navarra, Josh Picotte, Catherine Pociask, Harrison Price, Alan Rivero, Melinda Schafer, Morgan Sherwood Lauren Sullivan, Oona Takano, and A shley Williams for help in the field and lab at Archbold Biological Station. I thank Patrick Bohlen and Adam Peterson for help processing samples at the MacArthur Agro Ecology Research Center. I thank Verna Dunbar, Judy Maynard, and Louise for keeping me well fed during my summer field work. I thank my sister Melinda, my brothers Tim and Brian, and my father Richard for their support Most of all, I thank my mom, Charlotte, who collected soil samples with me i n the field, helped me process soil samples in the lab, washed many dishes, weighed plant samples, and did a lot of pipetting to help me complete my field and lab work. In addition, she is the one person who I know will read my entire dissertation. Thanks mom!

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 12 ABSTRACT ................................ ................................ ................................ ................... 15 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 17 2 SHORT TERM EFFECTS OF FIRE ON SOIL AND PLANT NUTRIENTS IN PALMETTO FLATWOODS ................................ ................................ ..................... 23 Introduction ................................ ................................ ................................ ............. 23 Methods ................................ ................................ ................................ .................. 25 Field Sampling and Lab Analyses ................................ ................................ .... 25 Statistical Analyses ................................ ................................ .......................... 30 Results ................................ ................................ ................................ .................... 31 Discussion ................................ ................................ ................................ .............. 32 3 VARIATION IN GROWTH RATIOS, ABOVEGROUND BIOMASS ALLOCATION, AND ALLOMETRIC RELATIONSHIPS OF RESPROUTING SHRUBS WITH TIME AFTER FIRE ................................ ................................ ....... 52 Introduction ................................ ................................ ................................ ............. 52 Methods ................................ ................................ ................................ .................. 55 Study Site and Species ................................ ................................ .................... 55 Statistical Analyses ................................ ................................ .......................... 57 Allometry of growth and biomass allocation ................................ ............... 57 Allometric equations to estimate biomass ................................ .................. 59 Results ................................ ................................ ................................ .................... 60 Allometry of Growth and Biomass Allocation ................................ .................... 60 Allometric Relationships to Estimate Biomass ................................ .................. 64 Discussion ................................ ................................ ................................ .............. 66 Allometry of Growth and Biomass Allocation ................................ .................... 66 Allometric Relationships to Estimate Bi omass ................................ .................. 73 4 SOIL NUTRIENT DYNAMICS ALONG A TIME SINCE FIRE CHRONOSEQUENCE IN SCRUBBY FLATWOODS ................................ ............. 99 Introduction ................................ ................................ ................................ ............. 99 Methods ................................ ................................ ................................ ................ 103

PAGE 7

7 Study Site ................................ ................................ ................................ ....... 103 Field and Lab Sampling ................................ ................................ .................. 103 Statistical Analyses ................................ ................................ ........................ 107 Results ................................ ................................ ................................ .................. 110 Species Composition ................................ ................................ ...................... 110 Root Biomass ................................ ................................ ................................ 111 Resin Exchangeable Nutrients ................................ ................................ ....... 111 Nitrogen Pools and Fluxes ................................ ................................ ............. 112 Bulk Soil Properties ................................ ................................ ........................ 113 Relationships Among Soil Variables ................................ ............................... 114 Discussion ................................ ................................ ................................ ............ 115 Effects of Time After Fire on Soil Characteristics and Nutrient Availability ..... 115 Variation in Soil Characteristics and Nutrie nt Availability with Soil Depth ....... 122 Effects of Abiotic Factors on Soil Nutrient Availability ................................ .... 123 Conclusion ................................ ................................ ................................ ...... 124 5 DISTURBANCE EFFECTS ON NUTRIENT LIMITATION OF PLANT PRODUCTIVITY IN SCRUBBY FLATWOODS: DOES FIRE SHIFT NITROGEN VERSUS PHOSPHORUS LIMITATION? ................................ .............................. 147 Introduction ................................ ................................ ................................ ........... 147 Methods ................................ ................................ ................................ ................ 150 Study Site ................................ ................................ ................................ ....... 150 Experimental Des ign ................................ ................................ ...................... 150 Soil Nutrients ................................ ................................ ................................ .. 152 Aboveground Biomass and Growth ................................ ................................ 154 Li tterfall ................................ ................................ ................................ ........... 157 Foliar Nutrients ................................ ................................ ............................... 158 Root Productivity ................................ ................................ ............................ 159 Results ................................ ................................ ................................ .................. 160 Soil Nutrients ................................ ................................ ................................ .. 160 Aboveground Cover, Biomass, and Growth ................................ ................... 161 Litterfall ................................ ................................ ................................ ........... 164 Foliar Nutrients ................................ ................................ ............................... 164 Root Productivity ................................ ................................ ............................ 165 Disc ussion ................................ ................................ ................................ ............ 165 Soil Nutrients ................................ ................................ ................................ .. 165 Aboveground Biomass and Growth ................................ ................................ 1 69 Litterfall ................................ ................................ ................................ ........... 176 Foliar Nutrients ................................ ................................ ............................... 177 Root Productivity ................................ ................................ ............................ 178 Concl usions ................................ ................................ ................................ .... 179 6 CONCLUSION ................................ ................................ ................................ ...... 204 APPENDIX

PAGE 8

8 A SHORT TERM EFFECTS OF FIRE ON SOIL AND PLANT NUTRIENTS IN SCRUBBY FLATWOODS ................................ ................................ ..................... 209 B SEASONAL VARIATION IN RESIN EXCHANGEABLE NITROGEN AND PHOSPHORUS IN SCRUBBY FLATWOODS ................................ ...................... 216 C SCRUBBY FLATWOODS SOIL ANALYSIS ................................ ......................... 226 D DIFFERENCES IN ABOVEGROUND BIOMASS AND STEM TURNOVER WITH TIME AFTER FIRE IN SCRUBBY FLATWOODS ................................ ....... 229 LIST OF REFERENCES ................................ ................................ ............................. 233 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 257

PAGE 9

9 LIST OF TABLES Table page 2 1 Results of repeated measures analysis o f variance for soil variables, and means (+ se) of soil variables pre fire and 0.125, 20, 62, 129, and 494 days (d) post fire.. ................................ ................................ ................................ ....... 45 2 2 Results of repeated measures analysis of variance for f 15 N, 15 15 N, foliar %P, and foliar N:P ratios ................................ ................................ ................................ ................... 46 3 1 Results of two way ANOVA analyses with time since fire, species, and thei r interaction as main effects ................................ ................................ .................. 77 3 2 Results of Kruskall Wallis tests analyzing differences in height:diameter ratios ................................ ................................ ................................ ................... 78 3 3 Results of regressions analyses comparing stem height to stem diameter for shrub species at ......... 78 3 4 Results of Kruskal Wallis tests analyzing differences in biomass ratios among times since fire for each species. ................................ ............................ 79 3 5 Results of Kruskall Wallis tests analyzing differences in biomass and biomass ratios among species within each time since fire (TSF). ...................... 79 3 6 Results of Kruskall Wallis tests and one way ANOVAs (indicated by *) analyzing differences in growth measures for Q. geminata and Q. inopina ........ 80 3 7 Results of Kruskall Wallis and t tests (indicated by *) analyzing difference in growth measures between Q. geminata and Q. inopina ................................ ..... 80 3 8 Results of regressions analyses for shrub species at each time since fire. ........ 81 3 9 Results of regressions analyses for palmettos S. etonia (left side; ln total 1 independent variable) and S. repens (right side; sqrt total 1 independent variable) at each time since fire. ...................... 82 3 10 Allometric equa tions predicting total stem biomass for shrub species overall and at each time since fire ................................ ................................ .................. 83 3 11 Allometric equations predicting leaf biomass for shrub species over all and at each time sinc e fire ................................ ................................ ............................. 84 3 12 Allometric equations predicting total leaf biomass for palmetto species overall and at each time since fire ................................ ................................ .................. 85

PAGE 10

10 3 13 Allometric equations predicting leaf lamina biomass for palmetto species overall and at each time since fire ................................ ................................ ...... 86 4 1 Description of study sites. ................................ ................................ ................. 125 4 2 Species composition (in 2006) of the twelve plots where soil samples were collected (in 2005). ................................ ................................ ........................... 126 4 3 Results of nested analyses of variance of soil varia bles at each depth. ........... 128 4 4 Results of repeated measures analyses of differences with depth for sites one, four, and six years after fire. ................................ ................................ ..... 130 4 5 Results of repeated measures analyses of differences with depth for sites eight, ten, and thirteen years after fire. ................................ ............................. 132 4 6 Results of multiple regressions, with root biomass (g m 2 ) as the dependent variable and dissolved inorganic N (DIN), dissolved organic N (DON), and percent N, for each soil depth. ................................ ................................ .......... 134 4 7 Partial correlations between soil variabl es at each depth (controlling for time since fire). ................................ ................................ ................................ ......... 135 4 8 Results of multiple regressions, with dissolved ino rganic N ( g N g soil 1) as the dependent variable and microbial biomass N (MBN), dissolved organic N 1 day 1 ) (N min), and soil C:N as independent variables ................................ ................................ ...................... 136 4 9 1 day 1 ) as the dependent variable and microbial biomass N (MBN), dissolved organic N (DON), soil C:N, and soil %N as independent variables ............................... 137 5 1 Mean (median) percent of total shrub cover of foc al shrubs in all plots pre fertilization at different times after fire ................................ ............................... 182 5 2 Mean (se) concentrations of macronutrients and micronutrients in scrubby flatwoods soils. ................................ ................................ ................................ 183 5 3 Results of one way ANOVAs (df = 2,6) and Kruskall Wallis tests (df = 2) analyzing differences in soil nutrients with time after fire ................................ .. 184 5 4 Results of general linear univariate models with treatment as a fixed effect and block as a random effect for soil nutrients. ................................ ................ 185 5 5 Results of general linear univariate models with trea tment as fixed effect and block as random effect testing for treatment effects on percent change in biomass and growth. ................................ ................................ ........................ 187

PAGE 11

11 5 6 Results of Kruskal Wallis tests analyzing the effects of treat ment on Quercus inopina stem survival and turnover over the first year of fertilization. ............... 188 5 7 Results of general univari ate models with treatment as a fixed effect, block as a random effect, and a covariate. ................................ ................................ 189 5 8 Effects of treatment on percent change in foliar and litter nutrients. ................. 190 5 9 Results of one way ANOVAs analyzing pre fertilization foliar nutrients of Quercus inopina and Sere noa repens ................................ ............................. 191 A 1 Mean (+ se) of soil variables measured in the scrubby flatwoods site before fire and after fire with results of repeated measures analyses. ......................... 212 A 2 Results of repeated measures analyses of variance for foliar %N, %P, and N:P ratios. ................................ ................................ ................................ ......... 213 B 1 Results of repeated measures analyses of resin exchangeable nutrients. ....... 222 B 2 Partial correlation coefficients (controlling for the effects of time since fire) for resin exchangeable total inorganic N and PO 4 3 ................................ ............... 223 C 1 Mean (se) concentrations of elements in scrubby flatwoods soils at different times after fire and soil depths. ................................ ................................ ......... 227

PAGE 12

12 LIST OF FIGURES Figure page 2 1 Mean (+ SE) soil extractable NH 4 + and NO 3 (A), soil extractable PO 4 3 (B), and soil inorganic N:P ratios (C) in palmet to flatwoods ................................ ...... 47 2 2 Mean (+ SE) foliar %N (A), foliar %P (B), and foliar N:P ratios (C) for Serenoa repens (n = 5) in palmetto flatwoods ................................ .................... 48 2 3 15 N for Serenoa repens (n = 5) (A), Quercus geminata (n = 4) (B), and three ericaceous species ( Lyonia fruticosa Lyonia lucida and Vaccinium myrsinities ; n = 4) (C) ................................ ...................... 49 2 4 Relationship between total extractable inorganic N and foliar %N ...................... 50 2 5 Relationship between soil extractable PO43 (natural log transformed) and foliar %P of Serenoa repens ................................ ................................ ............... 51 3 1 Mean (+ se) diameter (top panel), height (middle panel), and total stem biomass (bottom panel) of shrub species at each time since fire. ...................... 87 3 2 Grouped boxplots of height (cm):diameter (mm) ratios of shrub species at each time since fire. ................................ ................................ ............................ 88 3 3 Relationship between height and diameter (both natural log transformed) for all shrub species at each time s ince fire. ................................ ............................ 89 3 4 Grouped boxplots of leaf:shoot biomass ratios of shrub species at each time since fire. ................................ ................................ ................................ ............ 90 3 5 Grouped boxplots of new:old shoot biomass ratios of oak species at each time since fir e. ................................ ................................ ................................ .... 91 3 6 Grouped boxplots of the number of new apical shoot growth increments, mean length of new apical shoot growth increments per stem, number of leaves per cm of new shoot growth, and the ratio of height to the number of new apical shoot growth increments ................................ ................................ ... 92 3 7 Mean (+ se) total length of new apical shoot growth (top panel) and total number of new leaves (bottom panel) for Q. geminata and Q. inop ina ............... 93 3 8 Mean (+ se) heigh t, area, height(cm):area(cm 2 ) ratio, and number of leaves of palmettos at each time since fire. ................................ ................................ ... 94 3 9 Mean (+ se) total abo veground biomass of S. etonia (left panel) and S. repens (right panel) at each time since fire. ................................ ........................ 95

PAGE 13

13 3 10 Grouped boxplots of petiole:lamina biomass ratios of palmettos at each time since fire. ................................ ................................ ................................ ............ 96 3 11 Relationships between height and total stem biomass (left panel s; both natural log transformed) and between diameter and total stem biomass (right panels; both natural log transformed) ................................ ................................ 97 3 12 Relationships between length, width, height, and number of leaves versus total biomass for S. etonia (left panels) and S. repens (right panels) ................. 98 4 1 Mean (+ se) root biomass (g m 2 ) at each sampling depth for each time since fire. ................................ ................................ ................................ .......... 138 4 2 Relationship between root biomass and soil percent N ................................ .... 139 4 3 Mean (+ se) resin extractable NH 4 + NO 3 total inorganic N (A), PO 4 3 (B), and N:P ratio s (C) over one year. ................................ ................................ ..... 140 4 4 Relationship between pH of surface soils (0 5 cm) and resin exchangeable PO 4 3 ................................ ................................ ................................ ................ 141 4 5 Mean (+ se) soil %C (A), soil %N (B), soil C:N (C), K 2 SO 4 extractable dissolved inorganic N (DIN) (D), dissolved organic N (DON) (E), ratio of DIN 15 N (H), and soil pH (I) ..... 142 4 6 Mean (+ se) net N mineralization rates ( A) and nitrification rates (B) with H 2 O addition at each sampling depth for each time since fire. ................................ 143 4 7 Mean (+ or se) net N mineralization rates (left panels) and nitrification rates (right panels) at ambient field conditions and with H 2 O addition ....................... 144 4 8 Mean ( se) C pools (A) and N pools (B) with depth for each time since fire. ... 145 4 9 Relationship between microbial biomass N and dissolved inorganic N (DON) for each sampling depth. ................................ ................................ .................. 146 5 1 Mean monthly precipitation and total monthly precipitation during the study period at Archbold Biological Station. ................................ ............................... 191 5 2 Mean (+ se) resin exchangeable inorganic N (A), phosphorus (B), and N:P ratios (C) in control plots over one year. ................................ ........................... 192 5 3 Mean (+ se) soil extractable inorganic N in surface (0 10 cm) and deep (10 20 cm) soils in control and treatment plots ................................ ....................... 193 5 4 Mean (+ se) soil extractable P in surface (0 10 cm) and deep (10 20 cm) soils in control and treatment plots ................................ ................................ ... 193

PAGE 14

14 5 5 Mean (+ se) percent change in Serenoa repens biomas s during the first and second years of fertilization ................................ ................................ .............. 194 5 6 Mean (+ se) percent change in Quercus inopina plant biomass during the first and second years of fertilization ................................ ................................ 195 5 7 Mean (+ se) percent change in Quercus inopina stem height during the first and second years of fertilization ................................ ................................ ....... 196 5 8 Mean (+ se) percent change in Quercus i nopina stem basal diameter during the first and second years of fertilization ................................ .......................... 197 5 9 Mean (+ se) length of Quercus inopina total apical shoot growth increments during the first and second years of fertilization ................................ ............... 198 5 10 Mean (+ se) Quercus inopina leaf litterfall from March to May ......................... 199 5 11 Mean (+ se) foliar %N (A), foliar %P (B), and foliar N:P ratios o f Quercus inopina and Serenoa repens ................................ ................................ ............. 200 5 12 Mean (+ se) percent change i n foliar %N of Q. inopina and S. repens over the first year of fertilization ................................ ................................ ................ 201 5 13 Mean (+ se) percent change in foliar %P (A,B) and N:P ratios (C,D) of Q. inopina and S. repens over the first year of fertilization. ................................ ... 202 5 14 Mean (+ se) root productivity during the first year of fertilization. ..................... 203 A 1 Mean (+ se) foliar %N (A), foliar %P (B), and foliar N:P ratios (C) of S. repens in scrubby flatwoods (n = 5). ................................ ................................ ............. 214 A 2 Mean (+ se) foliar %N of plants in scrubby flatwoods pre fire and over time post fire. ................................ ................................ ................................ ........... 215 B 1 Mean monthly rainfall (1932 2005) and total monthly rainfall during my study period (June 2005 June 2006) at Archbold Biological Station .......................... 223 B 2 Mean (+ se) resin extractable NH 4 + (A), NO 3 (B), total inorganic N (C), PO 4 3 (D), and N:P (E) during each sampling period. ................................ ................. 224 B 3 Relationship between resin extractable NH 4 + (A), NO 3 (B), total inorganic N (C), PO 4 3 (D), and N:P (E) and rainfall during each 3 month sampling period. 225 D 1 Boxplots of percent survival (A) and percent change in stem number (B) of Quercus inopina plants 6 weeks, 8 years and 20 years since fire. .................. 232

PAGE 15

15 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF FIRE ON NUTRIENT AVAILABILITY AND LIMITATION IN FLORIDA SCRUB ECOSYSTEMS By Jennifer Lynn Schafer M ay 2010 Chair: Michelle C. Mack Major: Botany Nitrogen (N) and phosphorus (P) are essential plant nutrients that limit productivity in most, if not all, terrestrial ecosystems. F ire a common disturbance in many shrublands, may have different effects on the relative availability of N and P because N volatilization occurs at lower tem peratures than P volatilization I investigated the short and long term effects of fire on soil and plant nutrients in Florida scrub ecosystems. In addition, I tested the hyp othesis that nutrient limitation of plant productivity of scrubby flatwoods changes with time after fire. In flatwoods ecosystems, f ire caused a greater increase in phosphate ( PO 4 3 ) than ammonium ( NH 4 + ) resulting in a decrease in the soil available N:P r atio shortly after fire. Similarly, foliar %P of resprouting species increased more than foliar %N, resulting in a decrease in foliar N:P ratios shortly after fire. In scrubby flatwoods PO 4 3 but not total ino rganic N, varied with time after fire, causin g N:P ratios to be grea test at intermediate times after fire and lowest 13 years after fire. In surface soils, soil %C and %N, dissolved organic N, net N mineralization, and microbial N were all highest 13 years after fire and m easures of N availability w ere often highly correlated. In recently burned scrubby flatwood s shrubs appear to invest more in aboveground productivity, and Quercus inopina responded to

PAGE 16

16 P and N + P addition, but Serenoa repens responded to N addition. At intermediate times after fire shrubs appear to invest more in belowground than aboveground productivity and show co limitation by N and P with a stronger P limitation, while in long unburned sites, scrubby flatwoods shrubs appear to invest in both aboveground and belowground producti vity and show co limitation by N and P. Overall, my research suggest s that: (1) increased soil nutrient availability and reallocation of nutrients from below to aboveground can be important for plant nutrient status a fter fire ; (2) s pecies composition and fire freque ncy, as well as time after fire are impo rtant in affecting soil nutrient availability; and (3) t he effects of nutrient addition on biomass and growth depend on time after fire and species identity

PAGE 17

17 CHAPTER 1 INTRODUCTION Fire is a natural disturbance that maintains the structure and composition of many shrub dominated ecosystems (Little 1979, Abrahamson et al. 1984, Christensen 1985, Keeley and Keeley 1988, Moreno and Oechel 1994, Bradstock et al. 2001). Fire can have profound impacts on nu trient cycling and availability because f ire consumes plant biomass, litter, and soil organic matter, converting organic nutrients to inorganic forms (Certini 2005), which can be lost to the atmosphere or returned to the ecosystem in ash. Although fire oft en has no detectable effects on total soil nitrogen (N) (Christensen and Muller 1975, Jensen et al. 2001, Wan et al. 2001, Britton et al. 2008, Boerner et al. 2009) or phosphorus (P) pools (Kauffman et al. 1993) numerous studies have measured increases in concentrations of soil ammonium (NH 4 + ), nitrate (NO 3 ), and/or phosphate (PO 4 3 ) after fire (e.g., Lewis 1974, Wilbur and Christensen 1983, Stock and Lewis 1986, Schmidt and Stewart 1997, Giardina et al. 2000, Grogan et al. 2000, Wan et al. 2001, Smithwic k et al. 2005 a Turner et al. 2007). F ire has different effects on the relative availability of N and P due to fundamental differences in their biogeochemistry. Since N is more readily volatilized than P, relatively more N than P is lost from an ecosystem during fire (Raison et al. 1985a). In highly weathered soils, most P is in organic matter. Fire rapidly mineralizes the P in these pools, often resulting in enhanced P availability after fire. Nitrogen, however, may be relatively less available than P aft er fire due to greater combustion losses, and N inputs in unpolluted ecosystems are largely dependent upon biological N fixation, which accumulates N over the inter fire cycle. During fire, nutrients can be lost from an ecosystem to the atmosphere through volatilization (non particulate forms) or

PAGE 18

18 transport of ash (particulate forms). Numerous studies have measured the effect of fire on nutrient loss to the atmosphere by calculating the difference between the pre fire nutrient content of fuel (i.e. understor y plants and/or litter) and the post fire nutrient content of ash. For example, in a low intensity fire in a Mediterranean forest, 77% of N and 35% of P was lost (Gillon and Rapp 1989); during fires in Australian forests and woodlands, 92 94% of N and 41 5 3% of P was lost (Cook 1994); during slash and burn agriculture in Amazonia, 93 98% of N and 27 47% of P was lost (Mackensen et al. 1996); and fires in Brazilian savannas caused 95% of N and 51% of P to be lost (Pivello and Coutinho 1992). Regardless of th e type of ecosystem or fire, approximately twice as much N as P is lost to the atmosphere during fire due to differences in volatilization temperatures and forms of nutrient loss. N volatilization occurs at temperatures as low as 200C (White et al. 1973), whereas P is volatilized at temperatures above 774C (Raison et al. 1985a). The majority of N in combusted fuel is lost in non particulate forms, while P is lost in both non particulate and particulate forms. Thus, ash on the soil surface contains high co ncentrations of P and low concentrations of N (Debano and Conrad 1978, Raison et al. 1985b). Nitrogen and phosphorus limit plant growth in most, if not all, terrestrial ecosystems (Vitousek and Howarth 1991).Because fire has the potential to alter the re lative availability of N versus P both immediately following fire and over inter fire cycles, a fundamental question about nutrient limitation is whether fire causes shifts in N versus P limitation. This is important because nutrient limitation of plant pr oductivity is a fundamental control over the structure and function of ecosystems and has consequences for biomass accumulation (Bret Harte et al. 2004), nutrient retention and

PAGE 19

19 loss (Hedin et al. 1995), biodiversity (Wassen et al. 2005) and composition (Ch apin et al. 1987), and the ecosystem values and services provided to humans (Daily et al. 2000). The different fates of N and P in consumed fuel affect the total nutrient budget of an ecosystem. I nputs of N through symbiotic N fixation and rainfall are not high enough, over the short term, to replace the amount of N volatilized in fire (Carter and Foster 2004, Cook 1994). Over time, P in ash becomes relatively less available as it is immobilized by plants and microbes or fixed via geochemical reactions. Nit rogen availability, in contrast, tends to increase as inputs accumulate. Because plant growth is limited by any nutrient present in the soil below an optimum supply (Chapin et al. 2002), fire mediated differences in nutrient supply suggest that nutrient li mitation may change with time since fire, with recently burned sites being N limited and long unburned sites being P limited, particularly in old, highly weathered soils. Research on the effects of fire on nutrient limitation, however, is scarce. Pines an d oaks in a fire adapted Mediterranean forest were P limited in a site 5 years post fire (Sardans et al. 2004), only one of four species was N limited in a lodgepole pine forest 3 5 years after fire (Romme et al. 2009), and N limits productivity of trees i n secondary Amazonian forests, possibly due to the residual effects of fire on N availability (Davidson et al. 2004). In a study of aquatic systems, phytoplankton biomass was limited by P or co limited by N and P in lakes within unburned catchments, while phytoplankton biomass was limited by N in lakes within burned catchments, likely due to increased loading of P relative to N post fire (McEachern et al. 2002). Although the differential effects of fire on the fate of N and P suggest that the nutrient most limiting to

PAGE 20

20 plant production may change with time since fire, research has not directly addressed this question. Florida scrub provides a novel and interesting ecosystem in which to investigate the effects of fire on nutrient availability and limitation. F ire dependent Florida scrub ecosystem s occur on infertile, sandy, well drained soils Scrub ecosystems are found along both the Atlantic and Gulf coasts of Florida and along ridges in central Florida (Myers 1990). The broad definition of scrub includes man y vegetation types including sand pine scrub, oak scrub, rosemary scrub, coastal scrub, and scrubby flatwoods (Myers 1990) Shrubby oaks, palmettos, and ericaceous shrubs occur in all of these scrub types, but the dominant species varies and can be sand pi ne ( Pinus clausa ) or Florida rosemary ( Ceratolia ericoides ) Flatwoods occur at slightly lower elevations than scrubby flatwoods (Abrahamson et al. 1984), are fire adapted, have less well drained soils, and are often dominated by palmettos. Scrubby flatwoo ds are often an ecotone between flatwoods and other scrub vegetation types (Myers 1990). Previous research on nutrient availability in scrub ecosystems has focused on differences among plant communities ( Abrahamson et al. 1984, Kalisz and Stone 1984) or t he effects of elevated CO 2 (e.g. Hungate et al. 1999, Johnson et al. 2001, Johnson et al. 2003, McKinley et al. 2009 ). Few studies have focused on changes in soil nutrients with time after fire (Schmalzer and Hinkle 1992, Schmalzer and Hinkle 1996). Furthe rmore, the majority of these studies have been conduted in sites where the dominant oak species is Quercus myrtifolia (Schmalzer and Hinkle 1992). In many scrub areas along the central Florida Lake Wales Ridge, however, Quercus inopina is the dominant oak species.

PAGE 21

21 The dominant shrubby species in Florida scrub ecosystems resprout after fire. Belowground biomass may comprise up to 88% of total biomass of resprouting shrubs in Florida scrub (Saha et al. in review), and in a coastal scrub oak ecosystem, root biomass to one meter depth is 8000 g m 2 (Brown et al. 2007). Belowground reserves (McPherson and Williams 1998, Paula and Ojeda 2009), pre fire plant size (Bonfil et al. 2004, Konstantinidis et al. 2006), and fire intensity (Moreno and Oechel 1991, Llor et and Lpez Soria, 1993) affect resprouting ability (Moreno and Oechel 1991) and biomass of resprouts (Lloret and Lpez Soria, 1993, Cruz et al. 2002). The growth and biomass allocation of resprouts may depend on pre fire plant status and fire characteris tics and may vary among species because the differential growth of and resource allocation to aboveground plant parts depends on species specific constraints (Niklas 1995a). These differences may influence how nutrient availability affects growth of domina nt shrubs. U nderstand ing the role of fire in nutrient cycling may provide insight into the factors that contribute to the maintenance of ecosystem structure and function of Florida scrub. While many studies have investigated the demography of endangered p lants (e.g. Quintana Ascencio et al. 2003, Menges and Quintana Ascencio 2004) or the effects of fire on plant community composition (e.g. Schmalzer 2003, Weekley and Menges 2003), little is known about nutrient availability, plant productivity, or nutrient limitation in central Florida scrub despite evidence for low fertility. Large areas of land along the central Florida Lake Wales Ridge have been converted to agriculture, pastureland, and urban areas, leading to increased nutrient inputs and reduced fire frequency. Because Florida scrub ecosystems have been shaped over time by fire and

PAGE 22

22 low nutrient availability, human induced changes complicate restoration efforts. The goal of my dissertation research is to understand the effects of fire on nutrient availa bility and limitation in central Florida scrub ecosystems.

PAGE 23

23 CHAPTER 2 SHORT TERM EFFECTS OF FIRE ON SOIL AND PLANT NU TRIENTS IN PALMETTO FLATWOODS Introduction Fire, a natural disturbance in many shrubland ecosystems (Little 1979; Abrahamson et al. 1984; Christensen 1985; Keeley and Keeley 1988; Moreno and Oechel 1994; Bradstock et al. 2001), has profound impacts on nutrient cycling and availability. Fire consumes plant biomass, litter, and soil organic matter, converting organic nutrients into inorganic forms (Certini 2005) that may be lost to the atmosphere or returned to the ecosystem in ash. Although fire often has no detectable effects on total soil nitrogen (N) (Christensen and Muller 1975; Jensen et al. 2001; Wan et al. 2001; Britton et al. 2008; Bo erner et al. 2009) or phosphorus (P) pools (Kauffman et al. 1993) numerous studies have measured increases in concentrations of soil ammonium (NH 4 + ), nitrate (NO 3 ), and/or phosphate (PO 4 3 ) after fire (e.g., Lewis 1974; Wilbur and Christensen 1983; Stock and Lewis 1986; Schmidt and Stewart 1997; Giardina et al. 2000; Grogan et al. 2000; Wan et al. 2001; Smithwick et al. 2005 b ; Turner et al. 2007). F ire can have different effects on the relative availability of N and P because N volatilization occurs at t emperatures as low as 200C (White et al. 1973), whereas P is volatilized at temperatures above 774C (Raison et al. 1985a). Regardless of ecosystem type or fire intensity, approximately twice as much N as P is lost to the atmosphere during fire (Gillon an d Rapp 1989; Pivello and Coutinho 1992; Cook 1994; Mackensen et al. 1996). Thus, ash on the soil surface contains high concentrations of P and low concentrations of N ( Debano and Conrad 1978; Raison et al. 1985b), suggesting that fire affects both the abso lute and relative availability of soil N and P.

PAGE 24

24 Plant species in pyrogenic habitats have evolved a variety of mechanisms that allow them to persist and recover after fire (Sousa 1984; Christensen 1985). While some species are killed by fire and recolonize via seedling recruitment, other species are resilient and resprout after burning (Keeley 1977; Keeley and Zedler 1978; Menges and Kohfeldt 1995; Weekley and Menges 2003). Plant species that recruit from seed after fire rely on nutrients made available by fire; whereas, plant species that resprout after fire may similarly utilize nutrients made available by fire or reallocate nutrients from below to aboveground tissues (El Omari et al. 2003). The effect of fire induced changes in soil nutrient availability on plant nutrition, however, remains unclear. Several studies have found increases in foliar N and P after fire (Gilliam 1988; Franco Vizcano and Sosa Ramirez 1997), while others have found no effect of fire on foliar nutrients (Bennett et al. 2002; Ferr an et al. 2005). Understanding the effects of fire on foliar nutrient concentrations is important because variation in foliar N:P ratios with time since fire may indicate changes in plant nutrient status and nutrient limitation, as foliar N:P ratios have b een used to indicate N limitation, P limitation, or co limitation by N and P (Koerselman and Meuleman 1996; Gsewell 2004). 15 N values have been used as indicators of ecosystem nitrogen cycling (Martinelli et al. 1999). Fire consumes surface soils layers and volatilizes N, which can leave post fire soils enriched in 15 15 N tends to increase with depth ( Nadelhoffer et al. 1996; Frank and Evans 1997) 15 N signatures are related to plant N sources, my corrhizal status, rooting depth, N assimilation, and within plant N reallocation (Hgberg 1997; Evans 2001). Thus, taken

PAGE 25

25 15 N values may provide insight into integrated fire effects on plant and soil N dynamics and the causes of in creased foliar N concentrations after fire. I examined the effects of fire on plant and soil nutrient dynamics in flatwoods ecosystems of the Lake Wales Ridge in central peninsular Florida, where fire has historically maintained shrub dominated habitats (Abrahamson et al. 1984; Menges 1999). Although N and P are essential plant nutrients that limit plant growth in most, if not all, terrestrial ecosystems (Vitousek and Howarth 1991), few studies have investigated the effects of fire on both soil and plant N, P, and N:P ratios. Understanding nutrient dynamics in flatwoods ecosystems is important because nutrient availability is low and fires occur relatively frequently. I assessed the short term effects of fire on soil and plant nutrients and 15 N isotopic s ignatures. I hypothesized that soil extractable N and P would increase immediately post fire, but that the ratio of soil extractable N:P would decrease immediately post fire due to the differential effects of fire on N and P. Furthermore, I hypothesized th at N and P concentrations of resprouting plants would increase after fire. I investigated 15 N isotopic signatures to differentiate among mechanisms that can cause increased foliar N concentrations. M ethods Field Sampling and Lab Analyses This study was co nducted at Archbold Biological Station (ABS) in Highlands County, Florida, USA (2710'50"N, 8121'0" W), near the southern tip of the Lake Wales Ridge. The Lake Wales Ridge supports fire adapted Florida scrub ecosystems characterized by deep sandy soils de rived from paleo dunes (Abrahamson et al. 1984), and high endemism, with many endangered and threatened species (Menges 1999). Archbold Biological Station typically has warm wet summers and cool dry winters

PAGE 26

26 (Abrahamson et al. 1984). Mean annual precipitati on is 136.5 cm (ABS weather records, 1932 2004), and mean annual temperature is 22.3C (ABS weather records, 1952 2004). ABS includes a 5,193 acre preserve, which is divided into burn units that have been managed with prescribed fires for over 35 years. AB S comprises a mosaic of plant communities including seasonal ponds, flatwoods, scrubby flatwoods, oak hickory scrub, and sand pine scrub. My research focused on the palmetto flatwoods plant community. Palmetto flatwoods are dominated by saw palmetto ( Sere noa repens (W. Bartram) Small), a repent shrub that reaches heights of 1 2 m, and scattered shrubs with occasional to dense slash pines ( Pinus elliottii Engelm.). Palmetto flatwoods often occur as a distinct zone around seasonal ponds on entisols, inceptis ols, and spodosols that are poorly drained and can have standing water during times of high rainfall (Abrahamson et al. 1984). Flatwoods typically burn every 2 9 years (Main and Menges 1997). Palmettos and other dominant shrubs resprout after fire, while s lash pines survive by resisting fire and recruit from seed after fire (Menges and Kohfeldt 1995). Fires are intense and leave few areas unburned due to the high flammability of palmettos and pine duff (Abrahamson et al. 1984). Maximum sustained fire temper atures in flatwoods range from 373C to 688C, while absolute maximum temperatures have been measured as high as 796C (E. Menges, unpublished data). On 4 August 2006, I randomly selected five sampling locations within the palmetto flatwoods vegetation as sociation in a 19 acre burn unit that had previously burned in 2003, 1996, 1993, and 1972. At all sampling locations, which were separated by at least 5 m, I marked a soil sampling site and the nearest individual of five common

PAGE 27

27 flatwoods species (when pres ent within 1 m of the soil sampling location). My focal species, all of which resprout after fire, were the palmetto S. repens the shrubby oak Quercus geminata Small, and the ericaceous shrubs Lyonia fruticosa (Michx.) G.S. Torr., Lyonia lucida (Lam.) K. Koch, and Vaccinium myrsinities Lam. (Wunderlin and Hansen 2003). On 4 August 2006, several hours before ignition of a prescribed fire, I collected five soil samples (0 15 cm depth, 8 cm diameter core), one at each sampling location, and thirteen foliar sa mples, two to four at each sampling location depending on the species present. Serenoa repens was present at all sampling locations, while Q. geminata was present at four of the five sampling locations. Eleven of the thirteen plants sampled were completel y consumed by the fire. The first post fire soil samples (n=5) were collected on the afternoon of 4 August 2006, within three hours after the fire had burned through the unit. Subsequent post fire soil samples (n=5) were collected on 24 August, 5 October, and 11 December 2006 and 11 December 2007. I collected post fire foliar samples (n=13) on 6 October 2006, 12 December 2006, and 10 December 2007. At all sampling times, I collected the newest leaves from the upper portion of shrub stems. To collect foliar samples of the palmetto S. repens I clipped a small portion of the newest leaves (1 to 3 depending on total leaf number), thereby permanently marking the leaves. Thus, at all sampling times post fire, I collected a portion of only the new leaves that had been produced after the previous sampling event. Within 24 hours of collection, I passed soil samples through a 2 mm sieve and sub sampled for determination of gravimetric soil moisture, pH, total percentages of N and C, inorganic P concentration, inorgan ic N concentration, N mineralization rates, and

PAGE 28

28 15 N. Gravimetric moisture content was determined on samples dried at 105C for 48 hrs. For soil pH, 10 g of air dried soil was added to 10 mL of deionized water, shaken for 30 sec, allowed to stand for 10 min (Thomas 1996), then pH was determined with an electronic pH meter (Thermo Orion 250A+, Orion Research, Inc., Boston, Massachusetts, USA). A subsample of soil was dried at 60C for 48 hrs, ground to a fine powder on a spex mill (8000D dual mixer/mill Spex Certiprep Inc., Metuchen, New Jersey) at the MacArthur Agro Ecology Research Center (MAERC), and analyzed for percentages of N, C, and 15 N natural abundance at the University of Florida on an elemental analyzer (ECS 4010, Costech Analytical, Valenci a, California, USA) coupled with an isotope ratio mass spectrometer (Delta Plus XL, ThermoFinnigan, Brenen, Germany). Abundances of 15 N 2 as the standard. To measure inorganic P concentrations, 30 mL of 0.05 M hydrochloric acid (HCl) and 0.0125 M hydrogen sulfate (H 2 SO 4 ) was added to 15 g of field moist soil, shaken for 5 min, then filtered through Whatman #42 filter paper. I stored filtered samples in a refrigerator for up to three weeks before analy sis for phosphate (PO43 ) concentrations Tek Instruments, Inc., Winooski, Vermont, USA) using the malachite green method To measure inor ganic N concentrations, 50 mL of 0.5 M potassium sulfate (K 2 SO 4 ) was added to 10 g of field moist soil, shaken for 30 seconds, and allowed to stand overnight. I filtered solutions through Whatman #42 filter paper that was pre leached with 0.5 M K 2 SO 4 Filt ered samples were frozen then taken to the University of

PAGE 29

29 Florida where ammonium ( NH 4 + ) and nitrate (NO 3 ) concentrations were determined colorimetrically on a segmented flow autoanalyzer (Astoria Pacific, Inc., Clackamas, Oregon, USA). For N mineralization rates, 10 g of field moist soil was contained in a specimen cup and stored in the dark at room temperature (~ 24C). After one week, 50 mL of 0.5 M K 2 SO 4 was added to the soil, shaken for 30 sec, and allowed to stand overnight. I filtered, stored, and ana lyzed solutions as described above. Net rates of N ( NH 4 + + NO 3 ) g soil 1 of initial and one week extractions. Leaf samples were dried at 60C for 48 hours and ground on a spex mill (8000D dual mix er/mill, Spex Certiprep Inc., Metuchen, New Jersey) at the MAERC. All foliar samples were analyzed for percentages of N and C and 15 N natural abundance at the University of Florida on an elemental analyzer (ECS 4010, Costech Analytical, Valencia, Californi a, USA) coupled with an isotope ratio mass spectrometer (Delta Plus XL, ThermoFinnigan, Brenen, Germany). Abundances of 15 N were measu notation with atmospheric N 2 as the standard. I determined foliar phosphorus for all samples of Serenoa repens Subsamples of 0.05 to 0.5 grams were weighed into crucibles, ashed in a muffle furnace at 500C for 5 hours, extracted with 6 M HCl, then brought to volume so that the solution was 0.6 M. Extracts were stored in the refrigerator for several days then analyzed colorimetrically on a spectrophotometer microplate reader (PowerWave XS Microplate Reader, Bio Tek Instruments, Inc., Wi nooski, Vermont, USA) at the University of Florida using the ascorbic acid molybdenum blue method (Murphy and Riley 1962). Standard NIST peach leaves were used to determine the efficiency of the digestion.

PAGE 30

30 Statistical Analyses To examine changes in soil v ariables over time after fire, I used a one way mixed analysis of variance model with repeated measures with time as the within subjects factor (SAS 9.1; Littell et al. 2006). Differences in soil variables among times were determined with post hoc pairwise comparisons with Bonferroni confidence interval adjustments. Soil NH 4 + concentrations, total inorganic N, PO 4 3 concentrations, N:P ratios, soil 15 N, and soil %C were natural log transformed before analyses. Soil %N was square root transformed before analysis. Soil NO 3 concentrations could not be transformed to fit normality because of many zeros. To examine changes in foliar nutrients (%N, %P, 15 N over time after fire, I used one way repeated measures analysis of variance with time as the within subjects factor (SPSS 11.5; Field 2009). Differences in foliar nutrients and 15 N among times were determined with p ost hoc pairwise comparisons with Bonferroni confidence interval adjustments. In addition, I calculated the absolute 15 15 N (Chang and Handley 2000; Schuur and Matson 2001) for each plant at each soil sampling location and analyzed differences over time after fire with a one way repeated measures analysis of variance. Foliar nutrient variables were analyzed separately for each species/family ( Serenoa repens Quercus geminata and Ericaceae (ericaceous shrubs include Lyo nia lucida Lyonia fruticosa and Vaccinium myrsinities )). I used linear regression to assess the relationship between soil and foliar nutrients (Sigma Plot 11.0). I correlated the foliar %N of each individual at each site with total soil extractable inor ganic N at each site. Analyses were conducted separ ately for each species/family. I correlated foliar %P of S. repens with the natural log of soil

PAGE 31

31 extractable PO 4 3 Because the first post fire foliar sample collection corresponded with the third post fire soil sample collection, data from only four time points (pre fire and 62/63, 129/130, and 493/494 days post fire) were used in the regression analyses. Results Three hours post fire, soil NH 4 + concentrations were 5.5 times higher than pre fire values, an d NH 4 + remained higher through at least 20 days after fire (F 5,20 = 6.16, p = 0.001; Figure 2 1 ). Three hours post fire, PO 4 3 concentrations were 30 times higher than pre fire values, and 62 days after fire, PO 4 3 concentrations were 21 times higher than pre fire values (F 5,20 = 15.45, p < 0.001; Figure 2 1 ). Soil extractable N:P ratios decreased by 40% immediately after fire, from 8.4 to 5.1, due to the larger increase in PO 4 3 relative to NH 4 + but extractable N:P ratios were only significantly differen t between 62 days and 129 and 494 days after fire (F 5,20 = 5.85, p = 0.001, Figure 2 1 ). Soil pH increased over time after fire, and by 494 days, was significantly higher than pre fire values (Table 2 1). There were no differences in soil %N, %C, C:N rati os, NO 3 concentrations, or 15 N over time after fire (Table 2 1). Foliar %N and %P of Serenoa repens increased after fire (Figure 2 2). Foliar N:P ratios decreased after fire because of the larger increase in %P (1.39 times pre fire values) than %N (1.15 times pre fire valu es). Foliar %N, %P, and N:P ratios of S. repens were similar to pre fire va lues by 494 days post fire (Figure 2 2). Foliar %N of Quercus geminata and ericaceous species increased shortly after fire then decreased to pre fire va lues by 494 days post fire (F igure 2 3). Foliar 15 N of S. repens decreased significantly over time after fire, while foliar 15 N of Q. geminata increased then decreased after fire (Figure 2 3), although this change was only marginally significant (Table 2 15 N of ericaceous species

PAGE 32

32 did not change with time since fire (Table 2 2). The absolute difference between foliar 15 15 N did not vary over time after fire for any plant species/family (Table 2 2). Over the entire study period, the mean (+ se) difference between foliar and soil 15 N was 3.06 (+ 0.29) for S. repens 5.14 (+ 0.27) for Q. geminata and 5.83 (+ 0.15) for ericaceous shrubs. Total extractable inorganic N was positively correlated with foliar %N of Q. geminata (F 1,14 = 3.77, p = 0.073, R 2 = 0.21) and ericaceous shrub s (F 1,14 = 5.49, p = 0.034, R 2 = 0.28). Total extractable inorganic N was not correlated with foliar %N of S. repens (F 1,18 = 0.85, p = 0.368, R 2 = 0.04; Figure 2 4). Foliar %P of S. repens was positively correlated with soil extractable PO 4 3 (F 1,18 = 10. 57, p = 0.004, R 2 = 0.37; Figure 2 5). Discussion Fire caused a short term increase in soil extractable nutrients in the palmetto flatwoods ecosystem investigated in my study. While soil N concentrations remained elevated above pre fire levels for at le ast one month after fire, soil P, in contrast, remained elevated above pre fire levels for at l east two months after fire (Figure 2 1). Thus, flatwoods shrubs, which resprouted within a month after fire, experienced a sustained increase in P availability, but only a short pulse of N availability. Regardless, both foliar %N and %P increased over the short term after fire. The relative magnitude of soil P increase was greater than that of soil N, leading to a decrease in the soil extractable N: P ratio shortly after fire (Figure 2 1). Similarly, for the palmetto Serenoa repens the relative increase in foliar %P was greater than the increase in foliar %N, causing a decrease in the foliar N: P ratio shortly after fire (Figure 2 15 N did not

PAGE 33

33 vary with tim e since fire (Table 2 1), and only S. repens showed significant variation in 15 N with time after fire (Table 2 2). Soil ammonium ( NH 4 + ) concentrations increased immediately after fire and decreased to pre fire levels wi thin two months after fire (F igure 2 1). High concentrations of NH 4 + and NO 3 in burned sites may be related to high N mineralization and nitrification rates (DeLuca et al. 2002; DeLuca and Sala 2006); however, Turner et al. (2007) found that NH 4 + increased d uring the first year after severe stand replacing fire in pine forests, while net N mineralization rates were negative. A lthough n et N mineralization rates were affected by fire in my study (Table 2 1), they were negative throughout the study, indicating that post fire increases in inorganic N availability are not due to increased mineralization, but rather due to microbial or ash derived N. Increased soil temperatures associated with fire (Ewel et al. 1981) kill soil microbes, indicated by a decrease in microbial C and N after fire (Prieto Fernndez et al. 1998), which causes the release of N from ruptured microbial cells (Dunn et al. 1985; Serrasolsas and Khanna 1995). In addition, ash can contain high concentrations of N (Ewel et al. 1981; Kauffman et al. 1993), which can cause an increase in soil N after fire. Similarly to NH 4 + soil extractable phosphate ( PO 4 3 ) increased immediately after fire; however, in contrast to NH 4 + PO 4 3 decreased to pre fire levels wit hin four months after fire (Figure 2 1). High concentrations of PO 4 3 post fire are related to high concentrations of P in ash (Raison et al. 1985b; Kauffman et al. 1993). Loss of high nutrient ash can occur by wind (Giardina et al. 2000) or water (Ewel et al. 1981), suggesting that post fire weather contributes to variat ion in PO 4 3 concentrations. In my study, the first rain event occurred two days after fire, and 25.9 cm of rain fell between

PAGE 34

34 the first and second post fire sampling dates (ABS weather records), which likely limited loss of wind blown ash and could have co ntributed to high concentrations of soil PO 4 3 after fire (Tomkins et al. 1991). Soil extractable PO 4 3 increased more than total inorganic N after fire, causing a decrease in s oil extractable N:P ratios (Figure 2 1). This result is consistent with the fi ndings that more N than P is volatilized by fire (Gillon and Rapp 1989; Pivello and Coutinho 1992; Cook 1994; Mackensen et al. 1996) and that ash has higher concentrations of P than N (Raison et al. 1985b; Marcos et al. 2009). In a wetland with acidic, san dy soils, the soil extractable N:P ratio after fire was similar to my flatwoods site, but the pre fire soil extractable N:P ratio was greater than in my site, so the magnitude of the decline was greater (Wilbur and Christensen 1983). While post fire soil e xtractable N:P ratios may be similar across sites, differences in soil properties such as organic matter quantity or differences in fire temperature may affect the magnitude of fire induced changes in soil extractable N:P ratios. The post fire pulse of P O 4 3 persisted twice as long as the post fire pulse of NH 4 + (Figure 2 1). Although fire can kill soil microbes, the effects of fire on soil temperature decrease with depth (Ewel et al. 1981; Giardina et al. 2000; Jensen et al. 2001), so growth of microbes below the soil surface may be stimulated by post fire increases in nutrient availability (Singh et al. 1991) or root exudates (Blagodatskaya et al. 2009) from damaged roots (Scott Denton et al. 2006). Considering that soil microbial biomass N:P ratios aver age 7:1 at the global scale (Cleveland and Liptzin 2007), increased microbial growth would cause a faster decrease in soil N than soil P. Alternatively, sandy soils have lower sorption capacity than clayey soils (Villani et al. 1998), and high

PAGE 35

35 concentratio ns of potassium (K + ), calcium (Ca 2+ ) (Ewel et al. 1981; Kauffman et al. 1993), and chloride (Cl ) (Khanna and Raison 1986) in ash may affect the mobility of NH 4 + and PO 4 3 after fire. Leaching of Cl may be accompanied by leaching of NH 4 + (Khanna and Raiso n 1986), and K + can compete with NH 4 + for surface exchange sites (Chappell and Evangelou 2000); both of these interactions may contribute to high leaching losses of NH 4 + after fire. Phosphate ( PO 4 3 ) can form minerals with Ca 2+ and limited leaching of Ca 2 + after fire (Khanna and Raison 1986) suggests that leaching losses of PO 4 3 may be low after fire. Regardless of the mechanism that leads to a shorter pulse of NH 4 + than PO 4 3 plants experience a greater period of elevated P; however, if microbial uptake rather than leaching, reduces extractable NH 4 + N is retained in the ecosystem, rather than lost from the ecosystem, which may prevent or slow N limitation of primary productivity. In contrast to the effects of fire on inorganic nutrients, fire had no ef fect on total soil N, C, or C:N ratios (Table 2 1). Debano and Conrad (1978) reported decreases in total N in the top 2 cm of soil after fire, which was associated with high soil surface temperatures and a loss of soil organic matter; however, any change i n soil N or C in the top 0 2 cm of soil would likely be small relative to the total amount of N and C in the 0 15 cm of soil collected in my study. Other studies have found no effect of fire on total soil N (Christensen and Muller 1975; Jensen et al. 2001; Wan et al. 2001; Britton et al. 2008; Boerner et al. 2009), C, or C:N ratios (Boerner et al. 2009). Although fire often has limited effects on bulk soil properties, soil organic matter content and fire severity may mediate fire effects on total soil N and C.

PAGE 36

36 Soil pH increased over time after fire in my study (Table 2 1). The presence of ash may increase soil pH (Grogan et al. 2000; Bada and Mart 2003; Molina et al. 2007) due to the high pH of ash (Jensen et al. 2001; Goforth et al. 2005; Molina et al. 20 07; Marcos et al. 2009) and the high concentration of cations, such as Ca 2+ and K + in ash (Raison et al. 1985b; Arocena and Opio 2003). The majority of aboveground biomass in the flatwoods was consumed by fire, leaving large amounts of ash on the soil sur face. Soil pH increases with % base saturation (Magdoff and Bartlett 1985), and leaching of ash covered soils increases soil pH (Molina et al. 2007), so integration of cation rich ash through the top 15 cm of soil after rain events likely contributed to th e increase in soil pH over time. Soils are not well buffered between pH 4 and 7 (Magdoff and Bartlett 1985; James and Riha 1986), so even a small increase in sorption of Ca 2+ and K + could have caused an increase in soil pH (Skyllberg et al. 2001). In addit ion, microbial biomass N, total microbial respiration, and total phospholipid fatty acids are lower in soils at pH 4.17 than at pH 4.65 (Aciego Pietri and Brookes 2009), suggesting that the increase in pH over time after fire in my study, from 4.09 to 4.41 could have significant effects on the microbial community. Foliar %N and %P of the dominant flatwoods species, Serenoa repens increased shortly after fire, and were similar to pre fire values within 4 months after fire (Figure 2 2). The increase in fol iar %N occurred after soil extractable N was similar to pre fire levels; whereas, the increase in foliar %P persisted over the same time scale as the increase in soil extractable P. Foliar %N of Quercus geminata and ericaceous shrubs also tended to be high er shortl y after fire than pre fire (Figure 2 3). Several hypotheses could explain the increase in foliar %N and %P after fire. First, plants may

PAGE 37

37 be increasing foliar nutrients post fire due to increased availability of N and P. Increases in foliar nutrien ts in my study tended to mirror changes in extractable N and P. Foliar %N of Q. geminata and ericaceous shrubs was positively correlated with soil extractable N (Fig ure 2 4), and foliar %P of S. repens was positively correlated with soil extractable P. Sim ilarly, other studies have found that foliar nutrients of resprouting species are correlated with soils nutrients after fire (Gilliam 1988; Franco Vizcano and Sosa Ramirez 1997). Second, N and P stored in belowground tissues of resprouting plants may be r etranslocated to aboveground tissues. For example, to support new shoot growth, the resprouting shrub Quercus ilex first remobilizes N from belowground reserves then uses available N resources (El Omari et al. 2003). Decreases in retranslocation over time are suggested to occur as resource supply and root biomass increase (Salifu and Timmer 2001); however, for species that resprout after fire, an extensive root system already exists. Percent nonsoluble sugars in belowground structures of Q. geminata and V. myrsinities increase and decrease, respectively, with time since fire (Olano et al. 2006), suggesting that these species vary in their capacity to resprout after fire and that time since fire may affect the ability of shrubs to reallocate nutrients to abo veground tissues. In my study, foliar %N of S. repens was not correlated with soil extractable N, suggesting that retranslocation of N from below to aboveground may be more important for S. repens than for other species. Third, higher foliar nutrient conce ntrations after fire could be related to leaf age, increased leaf:shoot ratios, or the concentration of nutrients in a smaller amount of aboveground biomass post fire. New leaves tend to have higher N concentrations than old leaves (Hikosaka et al. 1994; A nten et al. 1998, Han et al. 2008). Although I have not investigated the effects of age

PAGE 38

38 on foliar %N of flatwoods species, I collected the newest leaves at each sampling time to minimize the effects of ontogeny on foliar nutrient concentrations. A decline in foliar N concentrations with leaf age can result from dilution of N (Han et al. 2008), but when plants are grow at high soil NO 3 concentrations, there is less difference in foliar N content with leaf age (Hikosaka et al. 1994), suggesting that an incre ase in extractable N and P after fire may contribute to more similar nutrient concentrations among leaves of different ages. In a study of savanna grasses, Van de Vijver et al. (1999) determined that higher foliar N concentrations after fire were due to hi gher leaf:stem ratios after fire, higher N concentrations in young rather than old leaves, and the distribution of N over the lower amount of post fire biomass; however, higher foliar P concentrations after fire were not easily explained. In my study, the number of palmetto leaves and the size of resprouting shrubs increased over the sampling period, so dilution of nutrients through more biomass could have occurred. Van de Vijver et al. (1999), however, did not find an effect of fire on soil nutrient availa bility, so their results do not rule out the possibility that higher foliar nutrient concentrations could be related to higher soil nutrient concentrations post fire when they occur. In my study, the relative increase in foliar %P was greater than the rel ative increase in foliar %N, so foliar N:P ratios of Serenoa repens decreased 20% from 15.8 pre fire to 13.2 two months post fire (Figure 2 2). Over the same time period, soil N:P ratios decreased 83% from 8.4 pre fire to 1.4 two months post fire, suggesti ng that changes in plant nutrition post fire are related to changes in soil extractable nutrients; however, since foliar N:P ratios did not decrease as much as soil extractable N:P ratios, reallocation of nutrients, particularly N, to aboveground tissue, i ncreased leaf:shoot

PAGE 39

39 ratios, and concentration of nutrients in a smaller amount of aboveground biomass post fire also likely contribute to the post fire increase in foliar %N and %P. In addition, the foliar N:P ratios of S. repens suggests that flatwoods sp ecies are co limited by N and P, because across habitats, N limitation occurs at foliar N:P ratios of 6.7 to 16 and P limitation occurs at foliar N:P ratios of 12.5 to 26.3 (Tessier and Raynal 2003). 15 N (Table 2 1). Saito et al. (2007) found that soils had to be burned at 400C for at least 5 minutes to cause a significant enrichment of soil 15 N, suggesting that high, sustained fire temperatures cause a greater loss of 14 N compared to 15 N. In addition, if fire consumes surface soils, volatilization of N may cause soils to become enriched in 15 N (Hgberg 1997). Temperatures recorded throughout a flatwoods fire usually exceeded 400C for only one or two minutes (E. Menges, un published data). Thus, low sustained fire temperatures in flatwoods concomitant with low soil organic matter could explain the lack of an effect of fire on soil 15 N. Flatwoods plants were depleted in 15 N compared to the soil, which is common in ecosystems with mycorrhizal species (Michelsen et al. 1998; Schmidt and Stewart 2003); however, there was no change in 15 N and 15 N over time after fire (Table 2 15 N signatures of flatwoods species; Quercus geminata tended to be more enriched in 15 N 62 and 130 days post fire compared to pre fire and 494 days post fire, while Serenoa repens became more depleted in 15 N over time (Figure 2 3). Grogan et al. (2000) found that all species in a pine forest were enriched in 15 N after fire, but this corresponded with 15 N af ter fire (Grogan et al. 2000). Considering that I found no 15 15 N signatures post fire could be caused

PAGE 40

40 by: (1) use of a different N source or change in discrimination of the same N source, (2) a change in the soil depth at which nutrient uptake occurs, (3) increased or decreased dependence on mycorrhizae for nutrient acquisition, and/or (4) within plant reallocation of N (Hgberg 1997, Evans 2001). Nitrogen sources (e.g. NH 4 + versus NO 3 ) vary in their isotop ic signatures (Dawson et al. 2002), which can affect foliar 15 N signatures (Evans 2001); however, extractable NO 3 was low throughout my study, suggesting that use of NO 3 as a N source did not change with time after fire. Discrimination against 15 N during N uptake can occur at high concentrations of NO 3 and NH 4 + (Kolb and Evans 2003); however, even the increased concentrations of NH 4 + after fire were likely not high enough to cause discrimination against 15 N. In addition, the lack of change in the difference in 15 N over time suggests that neither a change in N sources nor greater discrimination occurred in my study. S oil 15 N values tend to increase with depth (Nadelhoffer et al. 1996), and can 50 cm (Frank and Evans 1997). A shift in N uptake from surface (0 15 cm) to deeper (> 15 cm) roots could explain the increase in foliar 15 N of Q. geminata immediately after fire, while a shift in uptake from deeper to surface roots could cause the foliar 15 N of S. repens to become more depleted over time. In a coastal Florida scrub oak e cosystem, where CO 2 enrichment caused a decrease in soil extractable inorganic N, both Q. geminata and S. repens took up N from the water table, but S. repens showed a greater use of deep soil N (McKinley et al. 2009). Thus, S. repens may shift uptake of N from deep soil to surface soil in response to an increase in extractable N in surface soils after fire.

PAGE 41

41 Fire causes an increase in temperature (Ewel et al. 1981; Giardina et al. 2000; Jensen et al. 2001) and a decrease in moisture (Tomkins et al. 1991) o f surface soils, which likely affects nutrient uptake by surface roots. Little is known about the root distribution of Q. geminata and S. repens in palmetto flatwoods, but root biomass has been investigated in other ecosystems where these species occur. In a coastal Florida scrub oak ecosystem, where Q. geminata and S. repens comprise approximately 20% of the plant community, slightly less than half of roots < 0.25 mm in diameter occur in the top 10 cm of soil (Brown et al. 2007). In scrubby flatwoods, a le ss mesic shrubland ecosystem often occurring at slightly higher elevations than palmetto flatwoods, approximately 85% of palmetto roots and 60% of oak roots in the top 50 cm of soil are < 2 mm in diameter (Saha et al. in review). Thus, it seems unlikely th at Q. geminata and S. repens would differ in fire related root damage, suggesting that a shift in uptake from surface to deeper roots, or vice versa, could occur in response to changes in availability of or competition for soil nutrients. Quercus geminata has associations with ectomycorrhizae (Langely et al. 2002), ericaceous species have associations with ericoid mycorrhizae (Pearson and Read 1973), and S. repens has associations with arbuscular mycorrhizae (Fisher and Jayachandran 1999); fractionation dur ing the transfer of N from mycorrhizal fungi to a host plant results in plant tissue depleted in 15 N relative to the N source (Evans 2001; Hobbie and Colpaert 2003). Foliar 15 N of Q. geminata could be more enriched shortly 15 N of S. repens of could become more depleted over time if nutrient uptake through AM 15 N and the differen 15 N did not change over

PAGE 42

42 time for any species, however, suggesting that dependence on mycorrhizae for N uptake did not change after fire. In addition, Anderson and Menges (1997) and Eom et al. (1999) found that fire had no effect on colonization of roots by arbuscular 15 N signatures were more similar among plants with the same mycorrhizal status than among plants with the same post fire response (e.g. resprouter vs. seeder species), so mycorrhizal status may 15 N after fire. F oliar 15 N signatures decrease over time after leaf initiation (Bergersen et al. 1988), suggesting that reallocation of N within a plant can affect foliar 15 N signatures (Evans 2001). Although I sampled the newest leaves, as leaf number increased, I was 15 N over time after fire. In addition, leaves may be enriched in 15 N compared to roots (Evans et al. 1996), and the foliar 15 N of S. repens could have decreased over time due to reallocation of N depleted in 15 N from belowground to aboveground tissues. I hypothesize that changes in foliar 15 N signatures af ter fire are influenced by N reallocation and leaf age and that a change in N uptake from roots at different levels in the soil may also contribute to variation in foliar 15 N One limitation of my experimental design is that I did not measure soil or plan t nutrients in an unburned control site over the same time period that I measured soil and plant nutrients after fire in my flatwoods site; however, fire effects are the most likely explanation for my results for several reasons. First, across a scrubby fl atwoods time since fire chronosequence, resin exchangeable NH 4 + and PO 4 3 was 2.7 and 1.5 times higher, respectively, during September through December compared to June through

PAGE 43

43 September (J. Schafer, unpublished manuscript). In my flatwoods site, extractab le NH 4 + and PO 4 3 were 5.5 and 30 times higher, respectively, three hours after fire than before fire. This change is much greater, and occurred over a much shorter time period, than seasonal variation in nutrient availability; thus, it is unlikely that th e increases in extractable nutrients measured in this study are due to seasonal variation. Second, in the same palmettos flatwoods site used in this study, soil pH did not vary between September and November 2009 (J. Schafer, unpublished data). In addition in oak and saw palmetto scrub, an ecosystem similar to flatwoods, Schmalzer and Hinkle (1991) found that i n the first year after fire, soil pH was greater in December (12 months after fire) than in June (6 months after fire); whereas, in the second year after fire, soil pH was greater in June (18 months after fire) than in December (24 months after fire). Thus, changes in soil pH are likely due to fire effects rather than seasonal variation. Third, foliar %N and %P of oaks, ericaceous shrubs, and palmetto s is higher six weeks after fire than before fire or one year after fire in scrubby flatwoods sites burned in March and July (J. Schafer, unpublished data), suggesting that the pattern of increased foliar nutrients after fire is consistent across sites and does not depend on burn season. I did not measure foliar nutrient concentrations of clipped plants over the same time scale that I measured foliar nutrient concentrations of burned plants, but burning and clipping can have similar effects on plant nutrien t concentrations (Van de Vijver et al. 1999). In my study, fire caused a short term increase in soil extractable NH 4 + and PO 4 3 in a pa lmetto flatwoods ecosystem (Figure 2 1); PO 4 3 remained elevated above pre fire levels twice as long as NH 4 + possibly d ue to differences in microbial uptake and

PAGE 44

44 mobility of NH 4 + and PO 4 3 Both foliar %N and %P of resprouting plants increased over the short term after fire (Table 2 2). The relative increase in soil extractable P and foliar P was greater than that of soil e xtractable N and foliar N, leading to a decrease in the soil extractable N:P ratio (Figure 2 1) and the foliar N:P ratio of the palmetto Serenoa repens (Figure 2 2) shortly after fire. The relationships between soil and foliar nutrients coupled with measur ements of soil and foliar 15 N suggest that both an increase in soil extractable nutrients and reallocation of nutrients from belowground to aboveground tissue contribute to the increase in foliar %N and %P shortly after fire. Previous research in Florida scrub ecosystems has found limited effects of fire on soil nutrient availability (Abrahamson 1984, Schmalzer and Hinkle 1991). I found that a pulse of nutrients is detectable if soils are sampled soon enough after fire. Furthermore, my results suggest that even a short term increas e in soil extractable nutrients can be important for plant nutrient status, especially in ecosystems with low nutrient availability.

PAGE 45

45 Table 2 1. Results of repeated measures analysis of variance for soil variables, and means (+ se) of soil variab les pre fire and 0.125, 20, 62, 129, and 494 days (d) post 0.05 determined by post hoc pairwise comparisons with Bonferroni confidence interval adjustments. Variable F 5,20 P Pre Fire 0. 125 d post fire 20 d post fire 62 d post fire 129 d post fire 494 d post fire pH 3.27 0.026 4.09 + 0.05 a 4.23 + 0.08 ab 4.38 + 0.14 ab 4.34 + 0.11 ab 4.39 + 0.12 ab 4.41 + 0.08 b % N a 0.16 0.975 0.092 + 0.016 0.102 + 0.007 0.094 + 0.009 0.096 + 0.007 0. 099 + 0.024 0.110 + 0.034 % C b 0.18 0.968 2.85 + 0.54 3.30 + 0.39 2.98 + 0.42 2.99 + 0.27 3.04 + 0.90 3.23 + 0.91 C:N 0.54 0.747 30.79 + 0.64 32.08 + 1.58 31.38 + 1.41 31.09 + 0.73 29.67 + 1.31 29.95 + 0.85 Total Inorganic N b 1 ) 6.01 0.001 0.377 + 0.182 a 2.135 + 0.345 b 2.183 + 0.678 bc 0.962 + 0.096 ac 1.104 + 0.309 ac 1.077 + 0.304 ac N mineralization 1 day 1 ) 12.55 <0.001 0.009 + 0.004 a 0.144 + 0.025 b 0.110 + 0.081 ab 0.062 + 0.017 ab 0.060 + 0.007 b 0.049 + 0.040 ab 15 N b 0.49 0.777 2.77 + 0.16 2.47 + 0.32 2.77 + 0.32 3.03 + 0.30 2.51 + 0.52 2.50 + 0.59 a analysis performed on square root transformed data b analysis performed on natural log transformed data

PAGE 46

46 Table 2 2. Resu 15 N, the absolute difference in foliar 15 15 N, foliar %P, and foliar N:P ratios of Serenoa repens Quercus geminata and three ericaceous shrubs ( Lyonia fruticosa Lyonia luci da and Vaccinium myrsinities ). %N 15 N 15 N 15 N %P N:P Species/ Family F df n,d p F df n,d p F df n,d p F df n,d P F df n,d p S. repens 5.71 3,12 0.012 13.27 3,12 < 0.001 1.38 3,12 0.296 31.07 3,12 < 0.001 12.48 3,12 0.001 Q. geminata 5.63 3, 9 0.019 3.55 3,9 0.061 0.87 3,9 0.492 Ericaceae 7.44 3,9 0.008 1.31 3,9 0.329 0.24 3,9 0.863

PAGE 47

47 Figure 2 1 Mean ( + SE) soil extractable NH 4 + and NO 3 (A), soil extractable PO 4 3 (B), and soil inorgan ic N:P ratios (C) in palmetto flatwoods pre fire and 0.125, 20, 62, 129, and 494 days post fire. Different letters represent significantly variance with post hoc pairwise comparisons w ith Bonferroni confidence interval adjustments

PAGE 48

48 Fig ure 2 2. Mean ( + SE) foliar %N (A), foliar %P (B), and foliar N:P ratios (C) for Serenoa repens (n = 5) in palmetto flatwoods pre fire and 63, 130, and 493 days pos t 0.05 determined by repeated measures analysis of variance with post hoc pairwise comparisons with Bonferroni confidence interval adjustments

PAGE 49

49 Fi g ure 2 3 Mean ( + 15 N for Serenoa repens (n = 5) (A), Quercus geminata (n = 4) (B), and three ericaceous species ( Lyonia fruticosa Lyonia lucida and Vaccinium myrsinities ; n = 4) (C) pre fire and 63, 130, and 493 days post fire in pal metto flatwoods. Uppercase letters represent significant differences in foliar %N and lowercase letters represent significant differences 15 variance with post hoc pairwise comparisons wit h Bonferroni confidence interval adjustments. There were no post hoc differences in foliar %N or foliar 15 N among times after fire for Q. geminata or ericaceous species

PAGE 50

50 Figure 2 4 Relationship between total extrac table inorganic N and foliar %N of Serenoa repens (p = 0.368, R 2 = 0.04; foliar %N = 1.29 + (0.05 inorganic N)), Quercus geminata (p = 0.073, R 2 = 0.21; foliar %N = 1.09 + (0.16 inorganic N)), and ericaceous shrubs (p = 0.034, R 2 = 0.28; foliar %N = 0. 82 + (0.22 inorganic N)).

PAGE 51

51 Figure 2 5 Relationship between soil extractable PO43 (natural log transformed) and foliar %P of Serenoa repens (p = 0.004, R 2 = 0.37; foliar %P = 0.11 + (0.01 ln PO43 ))

PAGE 52

52 CHAPTER 3 VARIATION IN GROWTH RATIOS, ABOVEGROUND BIOMASS ALLOCATION, AND ALLOMETRIC RELATIONS HIPS OF RESPROUTING SHRUBS WITH TIME AFT ER FIRE Introduction Plant allometry theory predicts that leaf and stem biomass scale in relation to each other and in relation to s tem h eight and diameter (Niklas 1994; West et al. 1999; Enquist and Niklas 2001; Niklas and Enquist 2001; Enquist and Niklas 2002; Niklas and Enquist 2002 a ). While allometric relationships hold across many species, the relationships have been developed fro m species that grow from seed. Stems of resprouting shrubs, particularly those in fire adapted habitats, may differ from the predicted allometric relationships. While trees build one stem during their lifetime, shrub stems consumed by fire are rebuilt from belowground resources. Root biomass comprises approximately 25% of total plant biomass in angiosperm species (Niklas and Enquist 2002 b ); whereas, belowground biomass may comprise up to 88% of total biomass of respro uting shrubs (Saha et al. in review ). Th us, the constraints on the allometry of resprouted shrub stems may differ from the constraints on stems of other plant growth forms. Belowground reserv es (McPherson and Williams 1998; Paula and Ojeda 2009), pre fire plant size (Bonfil et al. 2004; Konstan tinidis et al. 2006), and fire in tensity (Moreno and Oechel 1991; Lloret and Lpez Soria, 1993) affect resprouting ability (Moreno and Oechel 1991) and biomass of resprout s (Lloret and Lpez Soria, 1993; Cruz et al. 2002). Thus, the allometric relationship s of resprouts may depend on pre fire plant status and fire characteristics in addition to the effects of metabolic rate on size dependent changes in plant architecture (West et al. 1999; Enquist and Niklas 2002),

PAGE 53

53 and may vary among species because the dif ferential growth of and resource allocation to aboveground plant parts depends on species specific constraints (Niklas 1995a). As stems grow taller, they must increase their mechanical strength through diameter growth (King 1986), but there is a tradeoff b etween height and diameter growth (Tilman 1988). Diameter should increase faster than height as a stem grows ( Niklas 1994 ); however, height growth per unit diameter is expected to differ between crowded and uncrowded individuals, and the magnitude and dire ction of change depends on cohort age and light environment (Henry and Aarssen 1999). In resprouting species, new stems occur in clumps (Silva et al. 2009), often around dead charred stems. The number of stems per clump is higher in more recently burned si tes (Davies and Myerscough 1991), suggesting that intra specific crowding of resprouts is high after fire. The effects of a clumped distribution may cause resprouted stems to deviate from allometric growth models. In addition, the differential growth of an d resource allocation to aboveground plant parts is affected by light environment (Coomes and Grubb 1998) and comp etition (Weiner and Thomas 1992; Kozovits et al. 2005), which vary with time since fire. Patterns of post fire recovery of aboveground bioma ss differ among ecosystems. In forests, aboveground biomass may increase logarithmically (Vargas et al. 2008) or linearly (White et al. 2004) with time since fire. In shrublands dominated by species that recruit from seed after fire, aboveground biomass ma y increase linearly (Johnson et al. 1986) or exponentially (Cleary et al. 2008) for at least 40 years after fire. In shrublands dominated by resprouting species, however, aboveground biomass fluctuates, but does not increase, after four years since fire (J ohnson et al. 1986; Sah et al. 2004),

PAGE 54

54 suggesting that acquired resources are allocated to belowground biomass. Growth rates of leaves, stems, and roots are expected to scale isometri cally (Niklas and Enquist 2002a; Niklas and Enquist 2002b); however, if th is does not hold for resprouting species, then scaling relationships between leaf and stem biomass, and between stem height, diameter, and biomass, may not adhere to allometric model predictions. Measurements of aboveground biomass may require harvesting of all plant material. This method, however, does not allow for repeated measurements of the same individuals over time, and is often not feasible in ecosystems with woody species. Thus, it is necessary to use non destructive measurements and allometric re lationships to estimate recovery of biomass after fire (e.g. Toma et al. 2005); however, the variability of sites from which species are sampled to develop allometric equations can affect estimates of biomass. For example, regression models predicting biom ass of aspen tress were significantly different among sites varying in resource availability (Koerper and Richardson 1980), and allometric equations developed from black spruce trees from sites varying in soil moisture and geographic region produced differ ent estimates of black spruce biomass (Mack et al. 2008). Despite the potential site specific error in biomass estimates, different allometric equations to estimate biomass of plants at different stages post fire are seldom used. In addition, research on a llometric relationships of species in fire adapted habitats has focused on dominant trees (e.g. Turner et al. 2004; Mack et al. 2008) or understory shrubs (Sah et al. 2004), rather than dominant resprouting shrubs, and many allometric equations for estimat ing aboveground biomass of shrubs use measurements of crown area or crown volume

PAGE 55

55 (e.g. Murray and Jacobson 1982; Vora 1988; Huenneke et al. 2001; Sah et al. 2004) instead of measurements of stem diameter and height. I examined the effects of time since fire on allometric relationships of shrub species in scrubby flatwoods ecosystems of the Lake Wales Ridge in central peninsular Florida. Scrubby flatwoods are fire adapted ecosystems, and the dominant species are shrubs that resprout aft er fire (Abrahamso n et al. 1984; Menges and Kohfeldt 1995). I tested the hypotheses that: (1) plant size and biomass increase logarithmically with time since fire; (2) height growth per unit diameter decreases with time since fire; (3) leaf:shoot and new:old shoot biomass r atios decrease with time since fire; and (4) plant size and biomass allocation ratios vary among dominant scrubby flatwood species. Furthermore, I developed allometric relationships to estimate aboveground biomass of the dominant shrubs at different times since fire. Methods Study Site and S pecies This study was conducted at Archbold Biological Station (ABS) in Highlands County, Florida, USA (2710'50"N, 8121'0" W). Archbold Biological Station typically has warm wet summers and cool dry winters (Abrahamson et al. 1984). Mean annual precipitation is 136.5 cm (ABS weather records, 1932 2004), and mean annual temperature is 22.3C (ABS weather records, 1952 2004). Archbold Biological Station includes a 5,193 acre scrub preserve, which is divided into burn unit s that have been managed with prescribed fires for over 35 years. My research focused on the scrub oak ( Quercus inopina Ashe) phase of scrubby flatwood communities (Abrahamson et al. 1984), which occur on sandy soils that have no horizon development, lit tle organic matter, and low ion exchange capacity (Brown et

PAGE 56

56 al. 1990). Scrubby flatwoods experience fire return intervals of 8 16 years (Menges 2007), and the dominant shrubs resprout after fire. Shrub height averages 1 2 m, but varies with resource availa bility and time since fire, and herbaceous species are sparse (Abrahamson et al. 1984). I focused on the most abundant species in scrubby flatwoods: the shrubby oaks Quercus chapmanii Sarg., Quercus geminata Small, and Q. inopina the ericaceous shrubs Ly onia fruticosa (Michx.) G. S. Torr. and Lyonia lucida (Lam.) K. Koch, and the palmettos Sabal etonia Swingle ex Nash (Arecaceae), and Serenoa repens (W. Bartram) Small (Arecaceae) ( nomenclature follows Wunderlin and Hanson 2003 ). The oak species vary in ab undance and phylogeny; Q. inopina has the highest percent cover (37%) and is a red oak, Q. chapmanii is intermediate in percent cover (22%) and is a white oak, and Q. geminata has the lowest percent cover (4%) and is also a white oak (Abrahamson et al. 198 4 ). Percent cover of L. lucida and L. fruticosa is 3% and 2%, respectively (Abrahamson et al. 1984). The oaks and ericaceous shrubs are clonal, multi stemmed species. Sabal etonia and S. repens are low growing palms with large fan shaped leaves, and percen t cover of S. repens (22%) is greater than percent cover of S. etonia (8%) (Abrahamson et al. 1984). Sabal etonia leaves are longer in length and width, and thus, are larger in area than S. repens leaves, while S. repens has more leaves than S. etonia (Abr ahamson 2007). During summer 2007, I measured and harvested aboveground stems of L. fruticosa L. lyonia Q. chapmanii Q. geminata and Q. inopina from sites 6 weeks and 8 and 20 years since fire. During summer 2008, I measured and harvested aboveground stems of the same species from the same sites, which were then 1, 9, and 21 years

PAGE 57

57 since fire. I measured the basal diameter (to the nearest 0.01 mm using digital calipers) and height (to the nearest cm) of each stem, harvested each stem, and separated stem s into leaves and shoots. In 2008, I separated all oak stems into new and old growth, and for Q. geminata and Q. inopina I counted the number of new apical shoot growth increments, measured the length of each new apical shoot growth increment, and counted leaves on the new shoots. All samples were dried at 60C for 48 hrs then weighed. During spring and summer 2009, I measured and harvested leaves of S. etonia and S. repens from sites 6 weeks and 1, 2, 8, 10, and 22 years since fire. I measured the maximu m crown diameter (length), minimum crown diameter (width), and height and counted the number of leaves of each individual. Leaves were harvested and separated into leaf lamina and petiole and dried at 60 70C for 48 hrs then weighed. Statistical Analyses A llometry of g rowth and b iomass a llocation Total aboveground biomass per stem, stem diameter, and stem height were natural log transformed then analyzed using a two way ANOVA with time since fire, species, and their interaction as main effects. Differences among species and times since fire were determined with post hoc Tukey HSD tests (JMP 8.0). Kruskall Wallis tests were used to analyze differences in leaf:shoot biomass ratios and height:diameter ratios of shrub species because these variables could not b e transformed to fit normality (SPSS 11.5). Significant differences among species and times since fire were determined using Bonferroni adjusted signi ficance values (Sokal and Rohlf 1995). I used an ANCOVA model, with diameter as the dependent variable, ti me since fire as a fixed factor, and height as the covariate, to test for homogeneity of regression slopes to

PAGE 58

58 determine if the relationship between stem height and stem diameter changes with time since fire (SPSS 11.5). Kruskall Wallis tests were used to analyze new:old shoot biomass ratios of oak species, and the number of new apical shoot growth increments, the mean length of new apical shoot growth increments per stem, the number of leaves per cm of new shoot growth, and the ratio of height to the numbe r of new shoot growth increments (a measure of height vs. branching) of Q. geminata and Q. inopina because these variables could not be transformed to fit normality (SPSS 11.5). Significant differences among times since fire and among species were determin ed using Bonferroni adjusted signi ficance values (Sokal and Rohlf 1995). Total length of new apical shoot growth and total number of leaves were natural log transformed before analysis using one way ANOVAs and post hoc Tukey tests to assess differences amo ng times since fire. Differences between species for each time since fire were analyzed with t tests (SPSS 11.5). Total aboveground biomass of palmetto species was analyzed using one way ANOVAs; S. etonia data was natural log transformed and S. repens dat a was square root transformed before analyses. A Kruskall Wallis test was used to analyze differences between palmetto species in aboveground biomass at each time since fire. Kruskall Wallis tests were also used to analyze differences in petiole:leaf lamin a biomass ratios and biomass per leaf of palmetto species because these variables could not be transformed to fit normality (SPSS 11.5); significant differences among times since fire and among species were determined using Bonferroni adjusted signi ficance values (Sokal and Rohlf 1995). Height, area, height:area ratios, and leaf number of

PAGE 59

59 palmettos were analyzed using two way ANOVAs with time since fire, species, and their interaction as main effects. Area was square root transformed and height:area ratios were natural log transformed before analyses. One extreme outlier ( S. etonia 8 10 years since fire) was removed from the analysis of height:area ratios because its removal made the data normally distributed. This S. etonia individual had only one leaf tha t was tall but narrow, so the height:area ratio was 0.69, which was 9.7 times greater than the next highest height:area ratio and over five standard deviations away from the mean. Differences among species and times since fire were determined with post hoc Tukey HSD tests (JMP 8.0). Allomet ric e quations to e stimate b iomass I used separate ANCOVA models for each shrub species, with total stem biomass as the dependent variable, time since fire as a fixed factor, and either height or diameter as the covariate to determine if the relationship between total stem biomass and the covariates changed with time since fire. Stem diameter, stem height, and aboveground biomass were natural log transformed before analyses. I used separate ANCOVA models for each palmetto s pecies, with total biomass as the dependent variable, time since fire as a fixed factor, and either length, width, height, or number of leaves as the covariate to determine if the relationship between total biomass and the covariates changes with time sinc e fire (SPSS 11.5). Aboveground biomass of S. etonia was natural log transformed before analyses, and aboveground biomass of S. repens was square root transformed before analyses. To develop allometric equations to estimate aboveground biomass, I used mul tiple linear regressions with biomass as the dependent variable and all field measures (i.e. stem diameter and stem height for shrubs; length, width, height, and number of leaves

PAGE 60

60 for palmettos) as independent variables (Sigma Plot 11.0), because in simple linear regressions, all independent variables were significant predictors of biomass. In some multiple regressions, only a subset of field measures signif icantly predicted biomass, so I conducted further multiple linear regressions or simple linear regress ions to predict biomass. This procedure was carried out to estimate both total aboveground biomass (i.e. leaves + shoot for shrubs and leaf lamina + petiole for palmettos) and total leaf biomass (leaf lamina for palmettos). Allometric equations were develo ped for shrub stems and palmetto individuals separately for each time since fire and together across all times since fire. For S. etonia in sites 22 years since fire, all field measures could not be included in the multiple linear regressions because only five individuals were sampled; thus, I used height and number of leaves with either length or width in multiple regressions. Results Allometry of Growth and Biomass A llocation Height and basal diameter of all shrub species increased from six weeks to one y ear after fire and from one year to eight nine years since fire (Table 3 1); however, from eight nine to 20 21 years after fire, height and diameter remained similar or decreased (Figure 3 1). Basal diameter did not vary among species one year or eight nin e years after fire, but 20 21 years after fire, L. lucida had the smallest basal diameters, which were, on average, at least 2 mm smaller than all other species (Figure 3 1). Six weeks after fire, Q. chapmanii and Q. inopina had the tallest resprouts, whic h were 1.3 to 2.4 times taller than resprouts of other species (Figure 3 1). L. fruticosa stems reached greater maximum heights than other species in longer unburned sites; eight to nine years after fire, L. fruticosa stems were, on average, 16 26 cm talle r than

PAGE 61

61 stems of other species. Twenty to twenty one years after fire, L. lucida stems were, on average, 10 cm shorter than all other species (Figure 3 1). Height:diameter ratios of all species were lowest six weeks after fire (Table 3 2). Height:diamet er ratios of Lyonia species were 2 3 times greater eight nine years since fire than six weeks after fire (Figure 3 2). Height:diameter ratios of Q. chapmanii and Q. geminata increased with time since fire and were 1.3 times higher 20 21 years since fire th an six weeks after fire, while height:diameter ratios of Q. inopina were highest one year after fire (Figure 3 2). Quercus species had higher height:diameter ratios than Lyonia species six weeks after fire; whereas, eight nine years after fire Lyonia speci es had higher height:diameter ratios than Quercus species (Table 3 2). The slope of the relationship between stem height and stem diameter changed with time since fire for L. fruticosa (F = 10.62, df = 3, p < 0.001), L. lucida (F = 6.59, df = 3, p < 0.001) Q. chapmanii (F = 11.62, df = 3, p < 0.001), Q. geminata (F = 10.49, df = 3, p < 0.001), and Q. inopina (F = 39.71, df = 3, p < 0.001). For all species, the slope of the height vs. diameter relationship was lowest six weeks after fire, and slopes were tw ice as steep eight nine years after fire compared to six weeks after fire (Figure 3 3). Twenty to twenty one years after fire, slopes were similar to or less than slopes eight nine years after fire (Table 3 3). Total aboveground biomass per stem was 6 t imes higher one year after fire than six weeks after fire and 4.6 times higher eight nine years after fire than one year after fire, but decreased slightly from eight nine to 20 21 years since fire (Figure 3 1). Biomass per stem of Quercus species tended t o be higher than biomass per stem of Lyonia species, with L. lucida having the lowest biomass per stem at all times since fire,

PAGE 62

62 but eight nine years since fire, L. fruticosa had the highest mean biomass per stem. Q. chapmanii and Q. inopina had greater bio mass per stem than Q. geminata (Figure 3 1). Leaf:shoot biomass ratios of Quercus species decreased with time since fire (Table 3 4) and were 4 5 times higher six weeks after fire than 20 21 years after fire (Figure 3 4). Leaf:shoot biomass ratios of Lyon ia species were highest one year after fire (Figure 3 4). At all times since fire, leaf:shoot biomass ratios varied among species (Table 3 5); Q. inopina had the highest leaf:shoot biomass ratios six weeks after fire, whereas L. lucida had the highest leaf :shoot biomass ratios 20 21 years after fire (Figure 3 4). New:old shoot biomass ratios of Quercus species were 1.7 to 4.3 times higher one year after fire than nine and 21 years after fire (Figure 3 5). One year after fire, Q. geminata new:old shoot biom ass ratios were 1.2 and 1.7 times lower than Q. chapmanii and Q.inopina new:old shoot biomass ratios, respectively (Table 3 5). The number of new apical shoot growth increments and the ratio of height to the number of new growth increment ratios of Q. gem inata did not vary with time since fire (Table 3 6). Q. geminata new apical shoot growth increments generally were longer one year after fire than nine and 21 years after fire (Figure 3 6). Total length of new apical shoot growth of Q. geminata did not var y with time since fire (Table 3 6). Q. inopina had three to six times more new apical shoot growth increments nine and 21 years since fire than one year after fire, and Q. inopina new apical shoot growth increments were twice as long one year after fire co mpared to nine and 21 years since fire (Figure 3 6). The total length of new apical shoot growth of Q. inopina did not vary with time since fire

PAGE 63

63 (Figure 3 7). The ratios of height to the number of new growth increments of Q. inopina were twice as high one year after fire than nine and 21 years after fire (Figure 3 6). One year after fire, Q. geminata had more new apical shoot growth increments than Q. inopina while Q .inopina had longer new apical shoot growth increments that Q. geminata (Table 3 7), so t he total length of new apical shoot growth did not differ between Q. geminata and Q. inopina (Figure 3 7) Q. inopina also had a higher ratio of height to the number of new growth increments than Q. geminata one year after fire (Table 3 7). The number of n ew growth increments, the total length of new apical shoot growth, and the ratio of height to the number of new growth increments did not differ between Q. inopina and Q. geminata nine or 21 years after fire (Table 3 7). The number of new leaves produced b y Q. geminata did not vary with time since fire (Table 3 6), but Q. inopina produced fewer new leaves one year after fire than nine and 21 years since fire (Figure 3 7). Q.geminata had more new leaves than Q.inopina one year after fire, but not nine or 21 years after fire (Table 3 7). The number of leaves per centimeter of new shoot growth of Q. inopina increased with time since fire (Figure 3 6), while the number of leaves per centimeter of new shoot growth of Q. geminata did not change with time since fir e (Table 3 6). Q. geminata produced more leaves per centimeter of new shoot growth than Q. inopina at all times since fire (Table 3 7). Height of palmettos did not vary with time since fire, and height and area of palmettos did not differ between species (Table 3 1). Palmettos covered twice as much area 22 years since fire than six weeks since fire (Figure 3 8). The height:area ratio of palmettos was at least 2.4 times higher six weeks after fire than other times since fire. Overall, the height:area ratio of S. repens was 1.3 times greater than the height:area

PAGE 64

64 ratio of S. etonia (Figure 3 8). Palmettos had more leaves in longer unburned sites than in sites six weeks since fire, and S. repens had twice as many leaves as S. etonia at all times except six wee ks after fire (Figure 3 8). Aboveground biomass of S. etonia tended to increase, but not significantly, with time since fire (F = 1.68, df = 3, p = 0.192), while aboveground biomass of S. repens was higher one two and 22 years since fire than six weeks aft er fire (F = 4.29, df = 3, p = 0.008) (Figure 3 9). Aboveground biomass of S. repens was greater than biomass of S. etonia one two years after fire (Table 3 5). Petiole:lamina biomass ratios of palmetto species increased with time since fire (Table 3 4) an d were approximately three times greater 22 years since fire than six weeks since fire (Figure 3 10). Biomass per leaf of palmetto species did not change with time since fire (Table 3 4). Allometric Relationships to Estimate B iomass The slope of the rela tionship between height and total stem biomass varied with time since fire for L. lucida (F = 10.47, df = 3, p < 0.001), Q. chapmanii (F = 7.08, df = 3, p < 0.001), Q. geminata (F = 6.82, df = 3, p < 0.001), and Q. inopina (F = 25.35, df = 3, p < 0.001), b ut not for L. fruticosa (F = 0.93, df = 3, p = 0.427) (Figure 3 11). Slopes tended to be steeper for longer unburned (8 21 years) than more recently burned (6 wks 1 year) sites (Table 3 8). On the contrary, the slope of the relationship between diameter and total aboveground biomass varied with time since fire for L. lucida (F = 3.55, df = 3, p = 0.016), but not for L. fruticosa (F = 1.93, df = 3, p = 0.128), Q. chapmanii (F = 0.15, df = 3, p = 0.928), Q. geminata (F = 0.79, df = 3, p = 0.502), or Q. inop ina (F = 0.69, df = 3, p = 0.560) (Figure 3 11). For S. etonia the slope of the relationships between total biomass and length (F = 1.21, df = 3, p = 0.326), total biomass and width (F = 0.97, df = 3, p = 0.421), total

PAGE 65

65 biomass and height (F = 0.31, df = 3, p = 0.816), and total biomass and number of leaves (F = 0.34, df = 3, p = 0.795) did not change with time since fire (Figure 3 12). For S. repens the slope of the relationships between total biomass and length (F = 1.64, df = 3, p = 0.191), total bioma ss and width (F = 1.78, df = 3, p = 0.161), and total biomass and number of leaves (F = 0.59, df = 3, p = 0.624) did not change with time since fire. The slope of the relationship between total biomass and height changed slightly with time since fire (F = 2.65, df = 3, p = 0.058); the slope for six week old individuals was lower than for other times since fire (Table 3 9). In all but one case, both stem diameter and stem height were significant predictors of total biomass of shrub stems (Table 3 10); howe ver, in several cases, either stem diameter or stem height was not a significant predictor of leaf biomass of shrub stems (Table 3 11). Overall, total stem biomass was better predicted by stem diameter and height than leaf biomass of shrub stems. The R 2 of allometric equations for all stems combined across times since fire was similar to or higher than the R 2 of equations for stems from specific times since fire for both total biomass and leaf biomass of shrub stems (Tables 3 10 and 3 11). For Quercus speci es, the R 2 of equations predicting total biomass tended to increase with time since fire (Table 3 10). Palmetto total leaf biomass was predicted slightly better than leaf lamina biomass. Only maximum crown diameter and height were significant predictors of total leaf biomass and leaf lamina biomass of S. etonia ; whereas, maximum and minimum crown diameter, height, and number of leaves were all significant predictors of total leaf biomass and leaf lamina biomass of S. repens (Tables 3 12 and 3 13). In seve ral cases,

PAGE 66

66 none of the variables in the multiple regressions significantly predicted biomass, but the predictive equations had a high R 2 Discussion Allometry of Growth and Biomass A llocation Growth ratios and aboveground biomass allocation changed with time since fire for all dominant scrubby flatwood species, with the greatest changes occurring from six weeks to eight ten years after fire. The patterns of change in growth ratios and aboveground biomass allocation with time since fire, however, were not consistent species. Belowground starch and non structural carbohydrate reserv es (McPherson and Williams 1998; Paula and Ojeda 2009) and pre fire plant size (Bonfil et al. 2004; Konstantinidis et al. 2006) are important in determining post fire resprouting ability, and thus size and biomass of resprouts. Six weeks after fire, oak ( Quercus species) resprouts tended to be taller, larger in diameter, and have greater biomass than ericaceous shrub ( Lyonia species) resprouts. Total nonstructural carbohydrate conc entrations of Florida scrub species may increase or decrease with time since fire (Olano et al. 2006); however, all resprouts were collected from sites with the same fire history. Rhizome resprouting potential is similar for Q. chapmanii and Q. geminata (C avender Bares et al. 2004), but oak species have greater belowground biomass than other scrub species, including Lyonia species ( Saha et al. in review ). Based on heights of shrub stems in longer unburned sites (Figure 3 3) and measurements of dead charred stems in the sites where resprouts were collected (J. Schafer, unpublished data), L. lucida was shorter than other species pre fire. Differences in both belowground

PAGE 67

67 storage and pre fire size likely enable oaks to produce larger resprouts than Lyonia specie s. All species were larger in diameter, taller, and had greater biomass per stem one year after fire compared to six weeks after fire, but oaks and ericaceous shrubs did not differ from each other. Eight to nine years after fire, there was no difference i n diameter, height, or aboveground biomass per stem among shrub species. Stems of all species had larger diameters, and all species except Q. chapmanii were taller, eight nine years compared to one year after fire; however, only L. fruticosa and Q. inopina had significantly more aboveground biomass eight nine than one year after fire. Overall, stem diameter, height, and biomass per stem decreased from eight nine years to 20 21 years since fire. This trend was driven more by Lyonia species than by Quercus s pecies (Figure 3 1). Several mechanisms could explain the lack of change or decrease in stem diameter, height, and biomass per stem from eight to 21 years since fire. First, there is a negative relationship between abundance and total plant mass (Allen et al. 2008); however, percent cover of shrubs overall and the number of stems within clumps does not vary between eight and 20 years since fire (J. Schafer, unpublished data). Second, as stem diameter increases, resources are allocated to the root system to increase stability (Drexhage et al. 1999). Stem diameter did not increase from eight nine to 20 21 years since fire, and root productivity does not differ between sites 8 and 20 years since fi re (Chapter 5 ). I hypothesize that shrub stems experience die b ack due to water (Saha et al. 2008) or nutrient limitation. Although the longer unburned sites have received similar amounts of precipitation over the last eight nine years, the 20 21 years since fire sites were located at the southern end of Archbold

PAGE 68

68 Biol ogical Station, where scrubby flatwoods occur 2m higher in elevation than further North, where the eight nine years since fire sites were located (Abrahamson et al. 1984). Thus, the 20 21 years since fire sites are located further above the water table, su ggesting that during times of drought, species such as Q. chapmanii and Q. geminata which take up water from 40 200 cm (Saha et al. 2008), may experience greater water stress. Percent dieback (Au and Tardif 2007) and the ratio of biomass loss to the gross production of aboveground biomass (Kawamura and Takeda 2008) increase with stem age; however, the mean and maximum life span of Q. inopina stems is 4 years and 9 years, respectively, and the rate of stem die off is lower in recently burned than long unbur ned sites (Johnson and Abrahamson 2002), so it is unlikely that stem age alone causes the observed patterns in stem size. Previous research found that the height of Q. inopina stems did not vary over a nine year period in a long unburned site (Johnson and Abrahamson 2002). Aboveground biomass of Q. inopina did not vary from 2 34 years after fire in Florida rosemary scrub (Johnson et al. 1986); whereas, aboveground biomass of oak shoots increased, while aboveground biomass of other shrub species, including Lyonia species, did not vary, from 3 25 years since fire in scrubby flatwoods (Saha et al. in review ). Biomass of shrubs in Florida Keys pine forests did not increase from 12 to 30 years after fire (Sah et al. 2004). The height:diameter ratio of all specie s increased from six weeks to one year after fire, and the relative increase was greater for ericaceous shrubs than for oaks; but overall, height:diameter ratios were similar one, eight nine, and 20 21 years since fire. Crowded individuals have larger hei ght:diameter ratios than trees growing in the open

PAGE 69

69 (Holbrook and Putz 1989), suggesting that shrub stems experience crowding within the first year after fire. In addition, the slopes of the relationship between stem height and diameter tended to become ste eper with time since fire (Figure 3 3). The relationship between stem diameter and height may be dependent on light conditions (Coomes and Grubb 1998), with height growth per unit diameter being greater in crowded trees, causing a shallower slope in the re lationship between height and diameter (Henry and Aarssen 1999). This suggests that resprouts are more crowded than shrub stems in sites with longer times since fire, and since resprouts tend to occur in clumps, the increase in crowding is likely caused by intra specific rather than inter specific competition. Henry and Aarssen (1999) predict that height increases faster than diameter in young stems and diameter increases faster than height in older stems. This may have contributed to the observed height di ameter relationships, as stems in the recently burned sites were less than one year old and stems in the other sites could have been one to many years old. Furthermore, stems of taller species can be disproportionately more slender than stems of shorter sp ecies (Niklas 1995 b ), so species specific differences may also affect height diameter relationships. Six weeks after fire, oaks allocated a greater proportion of resprout biomass to leaves than did ericaceous shrubs. The leaf:shoot biomass ratio decrease d in oak species from six weeks to one year since fire; whereas, the leaf:shoot biomass ratio increased in ericaceous shrubs and tended to be higher than in oaks (Figure 3 4). For all species, the leaf:shoot biomass ratio was lower eight nine and 20 21 yea rs after fire than one year after fire, which may be related to a decrease in the proportion of shoot mass in leaf laminae with increasing plant height (Anten and Hirose 1998). Species

PAGE 70

70 from other habitats have shown similar allocation patterns, with younge r stems having a greater proportion of biomass in leaf tissue than older stems (Cleary et al. 2008; Zewdie et al. 2009). This pattern of allocation to photosynthetic versus structural tissue may be related to the effect of removal of aboveground stems by d isturbance on shoot biomass (Scogings and Mopipi 2008) or to changes in allocation with an increase in stem biomass (Johnson et al 1986; Aikawa and Hori 2006). In addition, both shading (Vil 1997) and aboveground competition (Kozovits et al. 2005) have b een shown to cause a decrease in leaf:shoot biomass ratios. Resprouts of the shrub species I studied often occur in clumps, and shortly after fire, oaks tend to have a greater number of resprouts within a clump than ericaceous shrubs (J. Schafer, unpublish ed data). This suggests that oaks experience greater self shading an d intra specific competition tha n ericaceous shrubs within the first year after fire, which could explain the difference in patterns of change in leaf:shoot biomass ratios between oaks and ericaceous shrubs shortly after fire. Although plants from dense understorys allocate more biomass to supporting tissue than leaf tissue (den Dubbelden and Knops 1993), my data suggest that greater shading and aboveground competition may lead to an increa se in leaf:shoot biomass ratios; in sites 20 21 years since fire, L. lucida had the shortest stems and the highest leaf:shoot biomass ratios. This strategy may be beneficial in shrub dominated ecosystems where shrubs do not create a closed canopy. New:old shoot biomass ratios of oak species were higher one year post fire compared to longer unburned sites, which is due to greater shoot elongation in resprouts than in ma ture stems (DeSouza et al. 1986; Clemente et al. 2005) and a decrease in new:total shoot length with age (Aikawa and Hori 2006). Species

PAGE 71

71 differences in new:old shoot biomass only occurred one year after fire, with Q. geminata having lower new:old shoot biomass ratios than Q. chapmanii and Q. inopina (Table 3 5). These differences may be relate d to genetic constraints in the ability of live oaks and red and white oaks to produce new apical shoot growth. Q. geminata had slight or no differences in several measures of new growth with time since fire (Table 3 6), suggesting that allometry of growt h of Q. geminata is constrained. On the contrary, only total length of new apical shoot growth of Q. inopina did not differ with time since fire (Table 3 6), suggesting that the allometry of growth of Q. inopina is more plastic. Within the constraints of p lant architecture, the response of plants to environmental differences varies with ontogeny (Farnsworth and Ellison 1996). Plasticity in allometry tends to be strongest in early development (Bosner and Aarseen 2001), as was the case for Q. inopina with yo ung resprouts differing from older stems. One year after fire, compared to nine and 21 years since fire, Q. inopina produced fewer, but longer, new apical shoot growth increments, fewer leaves, and fewer leaves per unit of new shoot growth. In other specie s, non crowded plants produce more, longer branches than crowded plants (Weiner et al. 1990), and old plants produce more leaves than young plants (Oate and Munn Bosch 2008). The number of new apical shoot growth increments are indicative of the number o f shoot endings; thus, the variation in the ratio of height to the number of new shoot growth increments (Figure 3 6) suggests that Q. inopina invests more in increasing crown area one year after fire and invests more in height growth nine and 21 years aft er fire. Similarly, the ratio of height to number of shoot endings of Acer saccharum tended to decrease with seedling and sapling age, especially in open habitats (Bosner and Aarssen 1994). Plasticity in

PAGE 72

72 branching may also be related resource availability (Salemaa and Sievanen 2002), which changes with time since fire in scrubby flatwoo ds (Chapter 4 ). Variation in aboveground allocation patterns may be important in determining patterns of species abundances, as the use of resources to sequester space above ground may be a quantitative parameter of competitiveness (Kozovits et al. 2005). For example, six weeks after fire, Q. inopina the most abundant species in the scrubby flatwoods communities studied (Abrahamson et al. 1984), had the highest leaf:shoot bio mass ratio. One year after fire, Q. inopina invests in branching rather than height. By 20 21 years after fire, Q. inopina had the highest biomass per stem. The ability of Q. inopina to sequester space aboveground after fire, may contribute to the ability of Q. inopina to maintain dominance in scrubby flatwood communities over many fire cycles. Aboveground biomass of S. etonia increased, though not significantly, with time since fire, while biomass of S. repens was higher one two and 22 years since fire tha n six weeks since fire (Figure 3 9). Abrahamson (2007) found that S. etonia had fewer leaves and lower biomass than S. repens in scrubby flatwoods. Similarly, I found that S. etonia had fewer leaves than S. repens ; however, I found that biomass of S. repen s was greater than biomass of S. etonia only in sites one to two years since fire. The petiole:lamina biomass ratio of S. etonia and S. repens was highest twenty two years since fire, suggesting that palmettos increase allocation to leaf support relative t o leaf photosynthetic tissue. The height:area ratio of palmettos tends to be lower in longer unburned sites, suggesting that the increase in allocation to petiole biomass occurs to increase width rather than height of leaves, which is important in preventi ng self shading. Previous research has found that palm species increase investment to petiole

PAGE 73

73 biomass (Takenaka et al. 2001) and length to reduce self shading (Kimura and Simbolon 2002). Furthermore, S. repens may experience greater self shading since havi ng more leaves increases self shading (Takenaka et al. 2001). The height of shrubs in scrubby flatwoods increases with time since fire (Figure 3 1) and shrubs can reach greater maximum heights than palmettos, but on average, palmettos are taller than shrub s. Thus, palmettos may experience more within plant competition for light than competition from other species. Abrahamson (2007) found that total leaf canopy area of S. etonia and S. repens increased with percent overstory. In my study, leaf canopy area wa s similar from one to 22 years since fire, suggesting that percent overstory does not vary significantly with time since fire. Allometric Relationships to Estimate B iomass The relationship between stem diameter and stem biomass changed with time since fir e for only one shrub species; whereas, the relationship between stem height and stem biomass changed with time since fire for all but one species (Figure 3 11). This suggests that biomass scales with diameter more consistently than biomass scales with heig ht. Previous research has found that allometric relationships of wetland species were size dependent (Maio et al. 2008), and allometric equations of shrubs differ between study years (Lufafa et al. 2009), suggesting that variation in precipitation or nutri ent availability may affect allometric relationships. Cleary et al. (2008) found that allometric equations estimating aboveground biomass of sagebrush did not differ across a fire chronosequence. Sagebrush, however, recruits from seed after fire, whereas a ll of the shrubs in my study resprout after fire. Furthermore, the sagebrush chronosequence ranged from 3 60 years since fire, and I found the biggest changes in allometric relationships from six weeks to eight nine years since fire. Contrary to the oak an d

PAGE 74

74 ericaceous shrubs studied, the relationships between biomass and length, width, height, and number of leaves of palmettos did not change with time since fire, suggesting that variation in allometry is dependent on or constrained by growth form. My result s suggest that changes in allometric relationships with time since fire should be considered when estimating variation in stand biomass with time since fire, particularly for resprouting species. Competition affects allometric relationships (Weiner and Tho mas 1992); the relationship between height and mass differed for plants growing with competition and in isolation (Anten and Hirose 1998). Variation in the height diameter relationships of the shrubs studied suggest that crowding, and thus competition, var ies with time since fire. This likely contributes to the variation in height biomass relationships with time since fire. For oaks and ericaceous shrubs, both height and diameter were significant pred ictors of total stem biomass. M i a o et al. (2008) found t hat plant height had a relatively greater influence on plant biomass than basal area. Seiler et al. (2009) developed allometric relationships for Q. chapmanii and Q. geminata using only stem diameter; however, I found that multiple regressions with stem he ight and stem diameter resulted in better predictive power than simple regression with only height or diameter (Table 3 8 vs. Table 3 10). In addition, stem basal diameter and height were better predictors of total stem biomass than total leaf biomass (Tab le 3 8 vs. Table 3 9), which is similar to results of other studies (Robertson and Ostertag 2009). Measures such as maximum and minimum crown area are likely more indicative of leaf biomass than stem diameter. Although the oaks and ericaceous shrubs studie d are clonal, and

PAGE 75

75 each genetic individual may have one or many stems, I measured allometric relationships of individual stems because it is difficult to distinguis h individuals without excavation Considering that local position on the lignotuber can affec t stem growth (Riba 1998), not distinguishing genets from ramets may have affected allometric relationships. Abrahamson (1995) developed allometric relationships to determine total (aboveground and belowground stem) biomass of S. etonia and S. repens bas ed on minimum crown width and the number of number of leaves. Using a larger sample size overall, I found that length (maximum crown diameter), width (minimum crown diameter), height, and number of green leaves were all significant predictors of abovegroun d leaf biomass depending on time since fire and the species of palmetto. Aboveground measures of palmettos were equally good at predicting total leaf biomass and leaf lamina biomass, likely because petioles make up a small proportion of total leaf biomass. Overall, growth, allocation of resources to aboveground tissues, and allometric equations tended to differ with time since fire, but the majority of differences occurred between recently burned and intermediate and longer unburned sites. Resprouting abili ty of scrubby flatwood species appears to be important in determining growth and allometry immediately after fire. The fire return interval for scrubby flatwoods in 8 16 years (Menges 2007), and there was little or no difference in plant size, biomass, or allometric relationships from 8 22 years since fire within a species; however, there were differences among species within these times since fire. Thus, as time since fire increases, species specific constraints in growth and allometry appear to become mor e

PAGE 76

76 important. My results also suggest that caution should be taken in using allometric equations developed for plants from longer unburned sites to estimate biomass of plants in recently burned sites.

PAGE 77

77 Table 3 1. Results of two way ANOVA analyses with tim e since fire, species, and their interaction as main effects with dependent variables for both shrubs and palmettos. Oaks and Ericaceous Shrubs F df P Diameter (mm)* TSF 8.91 3 <0.001 Species 213.87 4 <0.001 TSF Species 2.20 12 0.010 Height (c m)* TSF 10.48 3 <0.001 Species 246.27 4 <0.001 TSF Species 3.39 12 <0.001 Stem Biomass (g)* TSF 226.88 3 <0.001 Species 34.37 4 <0.001 TSF Species 4.29 12 <0.001 Palmettos F df P Height (cm) TSF 1.83 3 0.147 Species 0.01 1 0.923 TSF Species 0.82 3 0.484 Area (cm 2 )* TSF 5.73 3 0.001 Species 1.97 1 0.164 TSF Species 0.77 3 0.511 Height:Area^ TSF 12.54 3 <0.001 Species 5.24 1 0.024 TSF*Species 1.56 3 0.217 # of Leaves TSF 8.81 3 <0.001 Species 29.23 1 <0.001 TSF Species 2.24 3 0.089 natural log transformed before analyses ^ square root transformed before analyses

PAGE 78

78 Table 3 2. Results of Kruskall Wallis tests analyzing differences in height:diameter ratios among times since fire for each shrub spe cies (left) and among species for each time since fire (right). Height:Diameter Height:Diameter Species 2 df p TSF 2 df p L. fruticosa 39.09 3 <0.001 6 wks 96.42 4 <0.001 L. lucida 97.53 3 <0.001 1 yr 33.21 4 <0.001 Q. chapmanii 11.74 3 0.008 8 9 yrs 26.01 4 <0.001 Q. geminata 15.13 3 0.002 20 21 yrs 11.48 4 0.022 Q. inopina 43.65 3 <0.001 Table 3 3. Results of regressions analyses comparing stem height to stem diameter for R 2 Adj R 2 p L. fruticosa 6 wks 0.256 0.202 0.305 0.290 <0.001 1 yr 0.395 0. 485 0.742 0.733 <0.001 8 9 yrs 1.125 0.739 0.827 0.818 <0.001 20 21 yrs 0.535 0.602 0.564 0.541 <0.001 L. lucida 6 wks 0.194 0.263 0.193 0.181 <0.001 1 yr 0.580 0.525 0.640 0.631 <0.001 8 9 yrs 2.047 0.932 0.796 0.785 <0.001 20 21 yrs 1.46 9 0.792 0.452 0.433 <0.001 Q. chapmanii 6 wks 0.174 0.374 0.408 0.399 <0.001 1 yr 0.991 0.674 0.672 0.662 <0.001 8 9 yrs 1.883 0.958 0.703 0.687 <0.001 20 21 yrs 1.789 0.912 0.767 0.756 <0.001 Q. geminata 6 wks 0.086 0.325 0.548 0.540 <0.001 1 yr 1.015 0.734 0.761 0.753 <0.001 8 9 yrs 0.454 0.609 0.436 0.413 <0.001 20 21 yrs 1.144 0.776 0.758 0.748 <0.001 Q. inopina 6 wks 0.068 0.252 0.298 0.293 <0.001 1 yr 0.811 0.587 0.720 0.717 <0.001 8 9 yrs 1.205 0.799 0.843 0.839 <0.001 20 21 yrs 1.090 0.758 0.857 0.854 <0.001

PAGE 79

79 Table 3 4. Results of Kruskal Wallis tests analyzing differences in biomass ratios among times since fire for each species. Leaf:Shoot New:Old Shoot Species 2 df P 2 df P L. fruticosa 39.84 3 <0.001 L. lucida 14.28 3 0.003 Q. chapmanii 93.63 3 <0.001 18.27 2 <0.001 Q. geminat a 77.54 3 <0.001 16.88 2 <0.001 Q. inopina 209.41 3 <0.001 21.39 2 <0.001 Petiole:Lamina Biomass/Leaf Species 2 df P 2 df p S. etonia 16.83 3 0.001 3.29 3 0.349 S. repens 20.15 3 <0.001 3.81 3 0.282 Table 3 5. Results of Kruskall Wallis tests analyzing differences in biomass and biomass ratios among species within each time since fire (TSF) Numbers in parentheses indicate degrees of freedom for each test. Oaks and Ericaceous Shrubs Palmettos Leaf:Shoot Biomass (4) New:Old Shoot Biomass* (2) Aboveground Biomass (1) Petiole:Leaf Biomass (1) TSF 2 p 2 p TSF 2 p 2 P 6 wks 178.12 <0.001 na na 6 wks 0.15 0.699 2.40 0.121 1 yr 43.04 <0.001 8.01 0.018 1 2 yrs 4.46 0.035 10.77 0.001 8 9 yrs 10.56 0.032 1.59 0.451 8 10 yrs 0.01 0.933 1.61 0.204 20 21 yrs 32.22 <0.001 3.41 0.182 22 yrs 0.20 0.655 0.94 0.333 oaks only; time since fire is 1, 9, and 21 years

PAGE 80

80 Table 3 6. Results of Kruskall Wallis tests and one way ANOVAs (indicated by *) analyzing differences in growth measures for Q. geminata and Q. inopina among times since fire. Degrees of fre edom = 2 for all statistical tests. Q. geminata Q. inopina Variable 2 F p 2 F p # of new growth increments 2.93 0.231 23.21 <0.001 Mean length (cm) of new growth increments per stem 6.72 0.035 29.39 <0.001 Total length of new growth (cm)* 0.02 0.977 1.02 0.365 # of new leaves* 0.76 0.762 5.89 0.004 # of leaves per cm of new shoot growth 5.04 0.081 23.25 <0.001 Height (cm) / # of new shoot growth increments 1.59 0.451 27.66 <0.001 Table 3 7. Results of Kruskall Wallis and t tests (indicated by *) analyzing difference in growth measures between Q. geminata and Q. inopina at each time since fire. Time since fire 1 year 9 years 21 years Variable 2 t df P 2 t df p 2 t df p # of new growth increments 11.74 1 0.001 0.24 1 0.623 0.00 1 0.985 Mean length (cm) of new growth increments per stem 33.10 1 <0.001 2.39 1 0.122 4.32 1 0.038 Total length of new growth (cm)* 0.83 107 0.405 0.77 30 0. 466 1.03 32 0.311 # of new leaves* 2.41 107 0.018 0.39 30 0.700 0.29 32 0.775 # of leaves per cm of new shoot growth 42.89 1 <0.001 4.78 1 0.029 9.14 1 0.002 Height (cm) / # of new shoot growth increments 30.91 1 <0.001 2.39 1 0.122 0.06 1 0.806

PAGE 81

81 Table 3 8. Results of regressions analyses for shrub species at each time since fire. 1 ln height 1 ln diameter (right side). Height Diameter 1 R 2 Adj R 2 p 1 R 2 Adj R 2 p L. fruticosa 0.427 0.128 6 wks 5.152 1.712 0.888 0.885 <0.001 3.982 3.398 0.469 0. 458 <0.001 1 yr 4.241 1.701 0.870 0.866 <0.001 2.301 3.000 0.858 0.853 <0.001 8 9 yrs 5.260 1.999 0.897 0.892 <0.001 1.878 2.520 0.939 0.936 <0.001 20 21 yrs 3.559 1.540 0.584 0.562 <0.001 2.005 2.442 0.945 0.942 <0.001 L. lucida <0.001 0 .016 6 wks 4.812 1.571 0.862 0.859 <0.001 3.147 1.692 0.332 0.322 <0.001 1 yr 4.506 1.678 0.736 0.729 <0.001 2.122 2.680 0.809 0.804 <0.001 8 9 yrs 7.475 2.444 0.873 0.866 <0.001 1.741 2.374 0.898 0.892 <0.001 20 21 yrs 5.571 1.971 0.602 0.589 < 0.001 1.186 1.863 0.746 0.738 <0.001 Q. chapmanii <0.001 0.928 6 wks 5.329 1.946 0.848 0.846 <0.001 2.397 2.757 0.581 0.575 <0.001 1 yr 6.369 2.272 0.889 0.885 <0.001 2.190 2.705 0.851 0.846 <0.001 8 9 yrs 7.495 2.640 0.785 0.773 <0.001 1.920 2.536 0.945 0.942 <0.001 20 21 yrs 8.250 2.774 0.923 0.920 <0.001 2.186 2.654 0.916 0.912 <0.001 Q. geminata <0.001 0.502 6 wks 4.076 1.501 0.723 0.718 <0.001 2.773 3.157 0.618 0.611 <0.001 1 yr 5.566 2.261 0.898 0.895 <0.001 1.892 2.628 0.859 0.855 <0.001 8 9 yrs 5.715 2.224 0.697 0.685 <0.001 2.412 2.701 0.874 0.869 <0.001 20 21 yrs 6.575 2.421 0.785 0.777 <0.001 2.624 2.879 0.883 0.878 <0.001 Q. inopina <0.001 0.560 6 wks 3.867 1.465 0.738 0.736 <0.001 2.071 2.72 3 0.542 0.539 <0.001 1 yr 5.017 1.826 0.864 0.862 <0.001 1.910 2.599 0.836 0.834 <0.001 8 9 yrs 6.694 2.457 0.923 0.921 <0.001 2.521 2.817 0.918 0.916 <0.001 20 21 yrs 6.116 2.321 0.922 0.920 <0.001 2.435 2.862 0.940 0.938 <0.001

PAGE 82

82 Table 3 9. Res ults of regressions analyses for palmettos S. etonia 1 independent variable) and S. repens 1 independent variable) at each time since fire. S. etonia S. repens 1 R 2 Adj R 2 p 1 R 2 Adj R 2 p Length 6 wks 1.414 0.0308 0.792 0.740 0.018 2.410 0.154 0.711 0.663 0.009 1 2 yrs 1.931 0.0261 0.853 0.837 <0.001 0.603 0.167 0.834 0.823 <0.001 8 10 yrs 1.386 0.0307 0.945 0.940 <0.001 2.391 0.168 0.846 0.836 <0.001 22 yrs 0.552 0.0458 0.953 0.937 0.004 6.099 0.220 0.832 0.822 <0.001 Width 6 wks 1.392 0.0476 0.839 0.798 0.010 0.195 0.211 0.913 0.898 <0.001 1 2 yrs 2.333 0.0290 0.836 0.818 <0.001 2.715 0.224 0.861 0.852 <0.001 8 10 yrs 1.773 0.0325 0. 905 0.895 <0.001 0.201 0.170 0.856 0.847 <0.001 22 yrs 1.189 0.0444 0.808 0.744 0.038 4.139 0.227 0.852 0.842 <0.001 Height 6 wks 0.798 0.0457 0.871 0.838 0.007 1.275 0.141 0.875 0.854 <0.001 1 2 yrs 0.547 0.0606 0.870 0.855 <0.001 8.632 0.323 0.881 0.873 <0.001 8 10 yrs 0.506 0.0625 0.694 0.663 <0.001 3.637 0.254 0.850 0.841 <0.001 22 yrs 1.468 0.0521 0.787 0.716 0.045 5.221 0.280 0.806 0.794 <0.001 # of leaves 6 wks 2.007 1.143 0.409 0.261 0.171 0.263 3.227 0.636 0.575 0. 018 1 2 yrs 1.525 0.958 0.702 0.669 0.001 0.680 2.130 0.743 0.727 <0.001 8 10 yrs 1.536 0.846 0.764 0.740 <0.001 0.624 1.874 0.852 0.842 <0.001 22 yrs 2.850 0.652 0.983 0.977 <0.001 2.667 2.128 0.933 0.929 <0.001

PAGE 83

83 Table 3 10. Allometric equa tions predicting total stem biomass for shrub species overall and at each time since fire; ln 1 2 ln height; n.s. indicates that the corresponding independent variable was not significant in the multiple regression m odel. Species N 1 2 MSE R 2 Adj R 2 p L. fruticosa 119 4.904 1.424 1.260 0.372 0.938 0.937 <0.001 6 wks 48 5.454 1.176 1.474 0.168 0.927 0.923 <0.001 1 yr 29 3.632 1.541 0.954 0.185 0.929 0.923 <0.001 8 9 yrs 21 3.436 1.622 0.802 0.097 0.964 0.96 0 <0.001 20 21 yrs 21 2.055 2.442 n.s. 0.241 0.945 0.942 <0.001 L. lucida 164 5.176 0.868 1.531 0.205 0.944 0.943 <0.001 6 wks 69 4.926 0.590 1.416 0.125 0.897 0.894 <0.001 1 yr 43 3.483 1.764 0.752 0.209 0.862 0.855 <0.001 8 9 yrs 20 4.607 1.402 1.138 0.098 0.936 0.929 <0.001 20 21 yrs 32 3.593 1.347 0.985 0.157 0.816 0.803 <0.001 Q. chapmanii 144 4.873 1.351 1.327 0.125 0.959 0.959 <0.001 6 wks 66 5.144 1.063 1.549 0.138 0.899 0.896 <0.001 1 yr 33 5.045 1.337 1.371 0.068 0.957 0.954 <0.0 01 8 9 yrs 21 3.699 2.015 0.710 0.100 0.962 0.958 <0.001 20 21 yrs 24 5.792 1.374 1.522 0.093 0.980 0.978 <0.001 Q. geminata 146 3.947 1.629 0.985 0.266 0.925 0.923 <0.001 6 wks 60 3.956 1.392 1.048 0.395 0.777 0.769 <0.001 1 yr 33 4.360 1.189 1. 388 0.091 0.940 0.936 <0.001 8 9 yrs 27 4.822 1.966 1.027 0.087 0.958 0.955 <0.001 20 21 yrs 26 4.144 2.126 0.771 0.223 0.902 0.893 <0.001 Q. inopina 327 3.891 1.443 1.037 0.168 0.945 0.945 <0.001 6 wks 149 3.964 1.409 1.111 0.173 0.840 0.838 <0.00 1 1 yr 87 3.982 1.277 1.077 0.160 0.920 0.918 <0.001 8 9 yrs 43 4.984 1.419 1.323 0.122 0.960 0.958 <0.001 20 21 yrs 48 4.304 1.663 1.060 0.104 0.967 0.966 <0.001

PAGE 84

84 Table 3 11. Allometric equations predicting leaf biomass for shrub species overa ll and at each time since fire; ln leaf 1 2 ln height; n.s. indicates that the corresponding independent variable was not significant in the multiple regression model. Species 1 2 MSE R 2 Adj R 2 p L. fruticosa 5.421 0.745 1.485 0.669 0.877 0.875 <0.001 6 wks 6.465 1.103 1.722 0.369 0.880 0.874 <0.001 1 yr 3.741 1.408 0.931 0.261 0.892 0.884 <0.001 8 9 yrs 1.279 1.635 n.s. 0.187 0.844 0.836 <0.001 20 21 yrs 2.345 2.099 n.s. 0.432 0.877 0.870 <0.001 L. lucida 5.590 0.521 1.643 0.406 0.981 0.889 <0.0 01 6 wks 5.427 n.s. 1.668 0.229 0.834 0.832 <0.001 1 yr 3.631 1.692 0.718 0.345 0.776 0.765 <0.001 8 9 yrs 2.441 2.370 n.s. 0.689 0.654 0.635 <0.001 20 21 yrs 4.390 0.835 1.186 0.313 0.625 0.599 <0.001 Q. chapmanii 4.784 0.934 1.280 0.311 0.871 0 .869 <0.001 6 wks 5.456 0.906 1.609 0.206 0.855 0.850 <0.001 1 yr 5.816 1.100 1.518 0.152 0.907 0.900 <0.001 8 9 yrs 1.920 2.031 n.s. 0.251 0.858 0.851 <0.001 20 21 yrs 6.320 0.997 1.557 0.465 0.885 0.874 <0.001 Q. geminata 4.176 1.019 1.097 0.58 3 0.802 0.800 <0.001 6 wks 4.475 1.070 1.235 0.581 0.718 0.708 <0.001 1 yr 4.714 0.791 1.514 0.177 0.872 0.863 <0.001 8 9 yrs 5.134 1.498 1.045 0.299 0.828 0.814 <0.001 20 21 yrs 3.250 2.659 n.s. 0.852 0.658 0.644 <0.001 Q. inopina 3.816 0.805 1. 077 0.381 0.828 0.827 <0.001 6 wks 4.154 1.385 1.124 0.207 0.815 0.812 <0.001 1 yr 4.235 1.041 1.064 0.401 0.793 0.788 <0.001 8 9 yrs 6.427 n.s. 2.080 0.630 0.756 0.750 <0.001 20 21 yrs 4.398 0.903 1.194 0.262 0.881 0.876 <0.001

PAGE 85

85 Table 3 12. Al lometric equations predicting total leaf biomass for palmetto species overall and at each time since fire; for S. etonia ln total biomass = 1 2 3 4 number of leaves; for S. repens sqrt 1 2 3 4 number of leaves; n.s. indicates that the corresponding independent variable was not significant in th e multiple regression model; n.a. indicates that the corresponding independent variable was not included in the multiple regression model; equations with all predictors and only significant predictors are both shown if the difference in R 2 is greater than 0.01. N 1 2 3 4 MSE R 2 AdjR 2 P S. etonia 34 1.083 1.110 0.0154 0.0248 0.00639 n.s. 0.0157 0.0137 0.155 n.s. 0.147 0.168 0.936 0.922 0.928 0.917 <0.001 <0.001 6 wks** 6 0.618 0.0202 0.0138 0.00459 0.414 0.010 0.998 0.989 0.070 1 2 yrs** 11 0.812 0.00486 0 .0116 0.0435 0.149 0.214 0.910 0.849 0.003 8 10 yrs 12 1.386 0.0307 n.s. n.s. n.s. 0.188 0.945 0.940 <0.001 22 yrs^ 5 0.302 0.552 0.0320 0.0458 n.a. n.s. 0.0181 n.s. 0.0397 n.s. 0.0013 0.059 1 0.953 0.999 0.937 0.024 0.004 S. repens 62 5.088 0.058 n. s. 0.103 0.901 2.378 0.960 0.958 <0.001 6 wks 8 2.823 1.275 0.0153 n.s. 0.00781 n.s. 0.0989 0.141 1.654 n.s. 0.576 1.369 0.974 0.875 0.939 0.854 0.011 <0.001 1 2 yrs 18 7.541 0.075 n.s. 0.167 0.428 1.147 0.978 0.973 <0.001 8 10 yrs 18 3.311 n.s. 0. 0652 0.0874 0.773 1.412 0.972 0.966 <0.001 22 yrs 18 5.404 n.s. n.s. 0.106 1.524 2.410 0.974 0.970 <0.001 ** no field measures were significant predictors of biomass in multiple regression with all predictors ^ only three of four predictive factor inclu ded

PAGE 86

86 Table 3 13. Allometric equations predicting leaf lamina biomass for palmetto species overall and at each time since fire; for S. etonia ln lamina biomass = 1 2 3 4 number of leaves; for S. repens 1 2 3 4 number of leaves; n.s. indicates that the corresponding independent variable was not significant in t he multiple regression model; n.a. indicates that the corresponding independent variable was not included in the multiple regression model; equations with all predictors and only significant predictors are both shown if the difference in R 2 is greater than 0.01. 1 2 3 4 MSE R 2 AdjR 2 p S. etonia 1.114 0.0239 n.s. 0.0131 n.s. 0.164 0.918 0.913 <0.001 6 wks 0.657 1.420 0.0186 0.0298 0.0147 n.s. 0.00539 n.s. 0.360 n.s. 0.0044 0.239 0.999 0.789 0.995 0.736 0.047 0.018 1 2 yrs** 0.854 0.00322 0.0107 0.0404 0.1 35 0.213 0.905 0.841 0.003 8 10 yrs 1.330 0.0298 n.s. n.s. n.s. 0.195 0.940 0.934 <0.001 22 yrs^ 0.436 0.032 n.a. 0.0191 n.s. 0.0003 1 1 <0.001 S. repens 3.819 0.0527 n.s. 0.0861 0.746 2.049 0.953 0.951 <0.001 6 wks 2.443 1.064 0.0127 n.s. 0.00121 n.s. 0.0887 0.132 1.513 n.s. 0.513 1.253 0.973 0.869 0.937 0.847 0.011 <0.001 1 2 yrs 5.922 0.060 n.s. 0.151 0.367 1.090 0.971 0.965 <0.001 8 10 yrs 2.322 n.s. 0.0574 0.0662 0.720 1.017 0.973 0.967 <0.001 22 yrs 4.067 n.s. n.s. 0.0819 1.341 2.137 0.9 68 0.963 <0.001 ** no field measures were significant predictors of biomass in multiple regression with all predictors ^ only three of four predictive factor included

PAGE 87

87 Figure 3 1. Mean ( + se) diameter (top panel), height (middle panel), and total stem biomass (bottom panel) of shrub species at each time since fire. Different letters represent significantly different means.

PAGE 88

88 Figure 3 2. Grouped boxplots of height (cm):diameter (mm) ratios of shrub species at each time since fire. The lower and upper ba rs of the boxplot represent the 25 th and 75 th percentiles, respectively; the solid middle bar represents the median and the dotted bar represents the mean. The lower and upper circl es show the 5 th and 95 th percentiles; outliers are not shown. Different lowercase letters below the boxplots indicate significant differences among species within each time since fire. Different uppercase letters above the boxplots indicate significant dif ferences among times since fire within a species.

PAGE 89

89 Figure 3 3 Relationship between height and diameter (both natural log transformed) for all shrub species at each time since fire. p values < 0.05 indicate that regression slopes are not homogeneous.

PAGE 90

90 Figure 3 4. Grouped boxplots of leaf:shoot biomass ratios of shrub species at each time since fire. Specifics of the boxplots are the same as in Figure 3 2. Different letters represent significant differences. Different lowercase letters below the b oxplots indicate significant differences among species within each time since fire. Different uppercase letters above the boxplots indicate significant differences among times since fire within a species.

PAGE 91

91 Figure 3 5. Grouped boxplots of new:old s hoot biomass ratios of oak species at each time since fire. Specifics of the boxplots are the same as in Figure 3 2. Different letters represent significant differences. Different lowercase letters below the boxplots indicate significant differences among species within each time since fire. Different uppercase letters above the boxplots indicate significant differences among times since fire within a species.

PAGE 92

92 Figure 3 6. Grouped boxplots of the number of new apical shoot growth increments, mean length of new apical shoot growth increments per stem, number of leaves per cm of new shoot growth, and the ratio of height to the number of new apical shoot growth increments of Q. geminata and Q. inopina at each time since fire. Specifics of the boxplots are t he same as in Figure 3 2. Different lowercase letters below the boxplots indicate significant differences between Q. geminata and Q. inopina within each time since fire. Different uppercase letters above the boxplots indicate significant differences among times since fire for Q. inopina

PAGE 93

93 Figure 3 7. Mean (+ se) total length of new apical shoot growth (top panel) and total number of new leaves (bottom panel) for Q. geminata and Q. inopina at each time since fire. Different letters represent significantly different means within a speices. indicates a significant difference between species.

PAGE 94

94 Figure 3 8. Mean (+ se) height, area, height(cm):area(cm 2 ) ratio, and number of leaves of palmettos at each time since fire. Dif ferent letters represent significantly different means.

PAGE 95

95 Figure 3 9. Mean (+ se) total aboveground biomass of S. etonia (left panel) and S. repens (right panel) at each time since fire. Different letters represent significantly different means.

PAGE 96

96 Figur e 3 10. Grouped boxplots of petiole:lamina biomass ratios of palmettos at each time since fire. The lower and upper bars of the boxplot represent the 25 th and 75 th percentiles, respectively; the solid middle bar represents the median and the dotted bar re largest and smallest values that are not outliers; outliers are not shown. Different lowercase letters below the boxplots indicate significant differences between S. etonia and S. repens within eac h time since fire. Different uppercase letters above the boxplots indicate significant differences among times since fire within a species.

PAGE 97

97 Figure 3 11. Relationships between height and total stem biomass (left panels; both natural log transformed) and between diameter and total stem biomass (right panels; both natural log transformed) for all shrub species at each time since fire; p values < 0.05 indicate that regression slopes are not homogeneous.

PAGE 98

98 Figure 3 12. Relationships between length, widt h, height, and number of leaves versus total biomass for S. etonia (left panels) and S. repens (right panels); p values > 0.05 indicate that regression slopes are homogeneous.

PAGE 99

99 CHAPTER 4 SOIL NUTRIENT DYNAMI CS ALONG A TIME SINC E FIRE CHRONOSEQUENC E IN SCRU BBY FLATWOODS Introduction Nitrogen ( N) and phosphorus (P) are essential plant nutrients that limit plant growth in most, if not all, terrestrial ecosystems (Vitousek and Howarth 1991). Fire, a natural disturbance in many ecosystems, consumes plant biomass litter, and soil organic matter, converting organic ally bound N and P into inorganic forms (Certini 2005) that may be lost to the atmosphere or returned to the ecosystem in ash. F ire can have different effects on the relative availability of N and P beca use N volatilization occurs at temperatures as low as 200C (White et al. 1973), whereas P is volatilized at temperatures above 774C (Raison et al. 1985a). Fire rapidly mineralizes P in soil organic matter, often resulting in enhanced P availabil ity after fire (e.g. Lewis 1974; Wilbur an d Christensen 1983; Adams et al. 1994; Giardina et al. 2000). Although N availability often increases after fire (e .g. Wilbur and Christensen 1983; Covington and Sackett 1992; Schmidt and Stewart 1997; Wan et al. 2001; Smit hwick e t al. 2005a; Turner et al. 2007 ), N may be relatively less available than P after fire (Chapter 1) because approximately twice as much N as P is lost to the atmosphere du ring fire (Gillon and Rapp 1989; Pivello and Coutinho 1992; Cook 1994; Mackense n et al. 1996). Over the long term, P in ash may become relatively less available as it is immobilized by plants and microbes or fixed via geochemical reactions while N may increase as N inputs accumulate ( Carter and Foster 2004 ). Previous studies, howev er, have not found consistent patterns in N and P availability across time after fire chronosequences. Ammonium (NH 4 + ), nitrate (NO 3 ), and/or available P may increase (MacKenzie et al. 2004; Durn et al. 2008 ) or decrease

PAGE 100

100 with t ime after fire (Marion and Black 1998; Wan et al. 2001; DeLuca et al. 2002; Bloom and Mallik 2006; MacKenzie and DeLuca 2006; Durn et al. 2008), and t otal soil N i ncreases (MacKenzie et al. 2004; Prez et al. 2004; Yermakov and Rothstein 2006) or does not change (Wan et al. 2001; B ond Lamberty et al. 2006) with time after fire, causing soil N:P ratios to increase (Bloom and Mallik 2006; Lagerstrm et al. 2009) or decrease with tim e after fire (Durn et al. 2008; Durn et al. 2009). Variation in patterns of N and P availability with time after fire may be due to differences in the fundamental ecosystem components that control nutrient availability or differences in fire intensity Nitrogen availability is controlled by mineralization and nitrification rates, which are positively corre lated with soil %N (Marion and Black 1988; Evans et al. 1998; Frank 2008) and may increase or decrease with time after fire (Prez et al. 2004 ; White et al. 2004; MacKenzie et al. 2006 ; Yermakov and Rothstein 2006 ). Microbial biomass, soil temperature, and soil moisture also affect N availability (Smithwick et al. 2005a) and are influenced by fire (Peet et al. 1975; Singh et al. 1991; Weekley et al. 2007; Marcos et al. 2009). Phosphorus availability is affected by soil pH (Jaggi et al. 2005), and ash on the soil surface after fire may increase soil pH (Grogan et al. 2000; Bada and Mart 2003; Molina et al. 2007). An increase in soil pH concomitant with high concentrations of calcium (Ca2+) in ash (Ewel et al. 1981; Kauffman et al. 1993) may lead to an incre ase in calcium phosphate (Hsu and Jackson 1960), thus occluding P made available by fire. H igh fire intensity causes greater availability of N and P than moderate or low intensity f ires (Gimeno Garca et al. 2000; Kennard and Gholz 2001; Romany et al. 200 1). C onsideration of the mechanisms causing changes in N and P availability over time after fire is necessary for understanding more general patterns.

PAGE 101

101 Because fire has the potential to alter the relative availability of N versus P both immediately followin g fire and over inter fire cycles, fire may cause shifts in N versus P limitation, with recently burned sites being N limited and long unburned sites being P limited, particularly in old, highly weathered soils. Fire frequency and intensity varies across e cosystems (Sousa 1984), causing differences in the time frame for changes in N and P availability after fire and in the impacts of fire on the plant community. In many ecosystems, f ire frequency and intensity has changed or is predicted to change, due to the spread of non native invasive spec Beckage 2009), human activities (DeWilde and Chapin 2006), and climate change (e.g. Beckage et al. 2003; Flannigan et al. 2005). Understanding how nutrient availability chan ges over time after fire is particularly important considering that an increase in fire frequency may prohibit accumulation of N over inter fire cycles and contribute to increased N limitation, whereas a decrease in fire frequency may exacerbate P limitati on if P remains stored in plant tissue. Much of the research investigating changes in nutrient availability with time after fire has been conducted in conifer dom inated (e.g. DeLuca et al. 2002; Bloom and Mallik 2006; MacKenzie and DeLuca 2006; Durn et a l. 2008) or chaparral (e .g. Christensen and Muller 1975; Marion and Black 1988; Fenn et al. 1993) ecosystems. Scrubby flatwo ods are a novel and intriguing ecosystem in which to investigate the effects of fire on nutrient availability for several reasons. F irst, scrubby flatwoods occur on well drained, nutrient poor quartz sand soils (Abrahamson and Hartnett 1990). Because the soil has low clay content and few weatherable minerals, the amount and type of dead organic matter is a primary control on nutrient a vailability and storage

PAGE 102

102 (Abrahamson and Hartnett 1990). High leaching, due to low clay content, contributes to low amounts of soil organic matter (Gholz and Fisher 1982), except in long unburned sites where litter accumulates. Because nutrient availability is low in scrubby flatwoods, fire related changes in nutrient availability are likely to be detectable from background fluxes in nutrient availability. Second, the dominant species resprout within weeks after fire and may utilize nutrients made available by fire or reallocate nutrients from below to aboveground tissues (El Omari et al. 2003). Regardless, low background nutrient availability suggests that a post fire flux in nutrient availability is likely to be an important source of plant nutrition. Thir d, scrubby flatwoods have been shaped over time by fire and low nutrient availability, yet little is known about nutrient availability, plant productivity, or nutrient limitation. Furthermore, large areas of land that previously supported scrubby flatwoods ecosystems have been converted to agriculture, pastureland, and urban areas, leading to increased nutrient inputs and reduced fire frequency. My main objective was to determine how soil nutrient availability varies over time after fire. Specifically, I t ested the hypotheses that: (1) N availa bility increases with time after fire, while P availability decreases with time after fire; (2) N pools increase with time after fire; and (3) N mineralization rates increase with time after fire. Because so il microbe s can be killed by fire, I hypothesized that microbial N would increase with time after fire. In addition, I tested the hypotheses that N availability is positively correlated with soil moisture and substrate availability and P availability is positively c orrelate d with soil pH. Furthermore I investigated differences in soil nutrient availability with depth.

PAGE 103

103 Methods Study Site This study was conducted at Archbold Biological Station (ABS) in Highlands County, Florida, USA (2710'50"N, 8121'0" W), which i s near the southern tip of the Lake Wales Ridge. Archbold Biological Station typically has warm wet summers and cool dry winters (Abrahamson et al. 1984). Mean annual precipitation is 136.5 cm (ABS weather records, 1932 2004), and mean annual temperature i s 22.3C (ABS weather records, 1952 2004). Archbold Biological Station is divided into burn units, which have been managed with prescribed fires for over 35 years. My research focused on scrubby flatwoods, a distinctive plant community of Florida scrub. Scrubby flatwoods are dominated by shrubby oaks (Fagaceae), palmettos (Arecaceae), and ericaceous shrubs (Ericaceae). The shrubs are primarily evergreen with an average height of 1 2 m and herbaceous species are sparse (Abrahamson et al. 1984). Scrubby fla twoods experience fire return intervals of 8 16 years (Menges 2007), and the dominant vegetation resprouts after fire (Menges and Kohfeldt 1995). Soils are entisols derived from paleo dunes (Abrahamson et al. 1984) that have no horizon development, little organic matter, and low exchange capacity and base saturation (Brown et al. 1990). Field and Lab Sampling In May 2005, I established eighteen 30 x 30 m plots in scrubby flatwood s communities (Abrahamson et al. 1984), three each in sites 1, 4, 6, 8, 10, and 13 years after fire (hereafter called the 1 yr, 4 yr, 6 yr, 8 yr, 10 yr, and 13 yr plots or sites). Within a time after fire, plots were located in different burn units when possible. Plots in the same burn unit were separated by at least 150 m and may ha ve experienced

PAGE 104

104 differences in fire intensity (Table 4 1). Therefore, each plot represents an independent replicate of the time after fire treatment. Overall, plots covered a distance of approximately four miles, and although summer thunderstorms can be pat chy, all plots experienced the same climate. All plots were established in scrubby flatwoods dominated by scrub oak ( Quercus inopina Ashe ) on flat or gently sloped sites. Thus, the climate, organisms, relief, and parent material were the same in all sites (Jenny 1941); the only state factor that varied among sites was time after fire. In each plot, I established 30m transects across each plot that were initiated at 5m, 10m, 15m, 20m, and 25m along the NW to SW side of the plot. In all plots, I used ion exc hange resins to measure soil nitrate (NO 3 ), ammonium (NH 4 + ), and phosphate (PO 4 3 ). At a random location on each transect, separate anion and cation exchange resin bags (5 x 5 cm) were placed in the top 5 cm of the soil and left in the field for 3 month i ntervals. Resin bags were in the field for one year continuously (Mid June Mid Sept. 2005, Mid Sept. Mid Dec. 2005, Mid Dec. 2005 Mid March 2006, and Mid March Mid June 2006). Before being buried in the field, anion and cation resin bags were charg ed with 2M HCl and 2M NaCl, respectively. After resin bags were removed from the field, they were rinsed with DI H 2 O to remove dirt and any attached roots. Anion and cation resin bags were extracted with 50 mL of 0.5 M HCl and 0.5 M NaCl, respectively and shaken for six hours. Resin extracts were frozen and taken to the University of Florida where NO 3 NH 4 + and PO 4 3 concentrations were determined colorimetrically on a continuous flow autoanalyzer (Astoria Pacific, Inc., Clackamas, Oregon, USA).

PAGE 105

105 In the t wo plots within each time after fire with the greatest spatial separation (12 plots total) I collected soil from a random location on each transect and one year later, recorded all species present in the plot (Table 4 2). During June and July 2005, one so il core was taken at each location and divided into 0 5, 5 10, 10 15, and 15 20 cm depths. Within 24 hours of collection, soil samples were passed through a 2 mm sieve; roots remaining in the sieve were separated into < 2 mm and > 2 mm, dried at 60C and w eighed. Soils were sub sampled for determination of gravimetric soil moisture, pH, 15 N, inorganic and organic N concentration, N mineralization rates, and microbial N. Gravimetric moisture content was determined on sampl es dried at 105C for 48 hrs. For soil pH, 10 g of air dried soil was added to 10 mL of deionized water, shaken for 30 sec, allowed to stand for 10 min (Thomas 1996), then pH was determined with an electronic pH meter (Thermo Orion 250A+, Orion Research, I nc., Boston, Massachusetts, USA). A subsample of soil was dried at 60C for 48 hrs, hand ground with a mortal and pestle, and analyzed for percentages of N, C, and 15 N at the University of Florida on an elemental analyzer (ECS 4010, Costech Analytical, Valencia, California, USA) coupled with an isotope ratio mass spectrometer (Delta Plus XL, ThermoFinnigan, Brenen, Germany). Soil bulk density increases with depth in acidic sandy soils (Skyllberg et al. 2001); thus, I used the mean bulk density for each depth for each time since fire to calculate N and C pools. To measure dissolved inorganic and organic N concentrations, 50 mL of 0.5 M K 2 SO 4 was added to 10 g of fie ld moist soil, shaken for 30 seconds, and allowed to stand overnight. Solutions were filtered through Whatman #1 filter paper that was pre leached with 0.5 M K 2 SO 4 Filtered extracts were sub sampled, frozen, then taken to the

PAGE 106

106 University of Florida. Dissol ved inorganic N (NH 4 + + NO 3 ) concentrations of one sub sample were determined colorimetrically on a segmented flow autoanalyzer (Astoria Pacific, Inc., Clackamas, Oregon, USA). I measured dissolved organic N (DON) on a separate sub sample of the soil extr act using the persulfate oxidation digestion method (Sollins et al. 1999), which converts all N into NO 3 I added 5 ml of the oxidizing reagent to 5 ml of the soil extract, then autoclaved the solution at 121C for 55 minutes. To determine the efficiency of the digestion procedure, an organic and inorganic standard were processed with each batch of samples. The NO 3 concentration of the digested sample was determined colorimetrically on a continuous flow autoanalyzer. To determine the concentration of DON, I divided the NO 3 concentration of the digested samples by the digestion efficiency, and then subtracted the DIN measured in the undigested samples. Field capacity of scrubby flatwood soils was determined by measuring gravimetric moisture content of sa turated soils (fr om sites four and 13 years after fire), and I measured N mineralization rates of soils at field capacity because soils were collected over a one month period with variation in daily precipitation (0 50 mm per day). A sub sample of 10 g o f field moist soil was contained in a specimen cup and stored in a refrigerator (~ 4C) for two days until gravimetric moisture content was determined. An average gravimetric moisture content was determined from all soils within each plot, and the same amo unt of de ionized (DI) H 2 O was added (169 226 different than the mean, a lower amount of DI H 2 O was added. Specimen cups were then stored in the dark at room tempera ture (~ 23C) for one week. After one week, 50

PAGE 107

107 mL of 0.5 M K 2 SO 4 was added to the soil, shaken for 30 sec, and allowed to stand overnight. Solutions were filtered, stored, and analyzed for NO 3 and NH 4 + as described above. Because cool temperatures (~ 4C) limit microbial activity, I considered these as seven day incubations. For half of the soil samples (from one plot for each time since fire) a second sub sample of 10 g of field moist soil was contained in a specimen cup and stored in the dark at room tem perature (~ 23C) for one week. After one week, 50 mL of 0.5 M K 2 SO 4 was added to the soil, shaken for 30 sec, and allowed to stand overnight. Solutions were filtered, stored, and analyzed for NO 3 and NH 4 + as described above. Net rates of N mineralization for both samples at field capacity and samples at ambient field moisture conditions, (NH 4 + + NO 3 ) g soil 1 of initial and one week extractions. The fumigation extraction method was used to measure microbial N (Horwath and Paul 1994). A 10 g sample of field moist soil was contained in a glass be aker and fumigated with chloroform in a desiccator for 24 hours. Soil samples were then transferred to a specimen cup, extracted with 50 ml of 0.5M K 2 SO 4 and processed the same as for DIN Fumigated soil extracts were digested to convert all N to NO 3 usi ng the same persulfate oxidation digestion used to measure DON. The NO 3 concentration of the fumigated, digested sample was determined colorimetrically on a continuous flow autoanalyzer. To determine the concentration of chloroform labile microbial N, I d ivided the NO 3 concentration of the fumigated samples by the digestion efficiency, then subtracted the DIN and DON measured in unfumigated samples. Statistical Analyses I calculated total resin exchangeable nutrients over one year by summing NH 4 + NO 3 total inorganic N, and PO 4 3 over each sample period. Resin exchangeable N:P

PAGE 108

108 ratios were determined from the total inorganic N and total PO 4 3 for each site within a plot. Missing values (when a resin bag was found on the soil surface) were replaced with the plot mean for that time period. I averaged values for each plot so that plot was the statistical unit. Data were analyzed with one way ANOVAs with post hoc comparisons with Bonferroni confidence interval adjustments. Resin exchangeable NO 3 and PO 4 3 were natural log transformed before analyses. I analyzed the correlation correlation coefficients, to control for the effects of time since fire, with one tailed significance tests. I analyzed differences in soil properties (root biomass ( < 2 mm diameter), soil pH, soil %C, soil %N, soil C:N, soil C pools, soil N pools, total dissolved inorganic N (DIN), dissolved organic N (DON), DIN:DON, N mineralization with H 2 O addition (both soil 1 day 1 1 day 1 ), nitrification rates with H 2 1 day 1 15 N) among times since fire separately for each soil depth (0 5, 5 10, 10 15, and 15 20 cm) using a nested ANOVA model with time s ince fire and plot nested within time since fire as main effects. I used a nested model because variability in soil nutrients can be spatially dependent at a scale of 30 m or less (Blair 2005). Differences among times since fire were determined with pairwi se comparisons with Bonferroni confidence interval adjustments. The effects of water addition on N mineralization and nitrification rates were analyzed separately for each time since fire and depth with paired, one tailed t tests. Differences in soil prope rties with depth were determined separately for each time since fire using repeated measures analysis with depth as the within subjects factor and plot as the between subjects factor. When the assumptions of

PAGE 109

109 sphericity were not met, the degrees of freedom were adjusted using the Greenhouse Geisser estimated epsilon values, which is a conservative correction (Field 2009), and differences among dates and times since fire were determined with post hoc pairwise comparisons with Bonferroni confidence interval ad justments. Data were transformed before analyses to meet the assumption of normality when necessary. I used separate linear regressions for each depth within each time since fire to g soil 1 day 1 ). I calculated mean pH per plot of surface soils (0 5 cm), and used linear regression to investigate the relationship between pH of surface soils and resin extractable PO 4 3 (resin bags were located in surface soils) for the plots in wh ich both soil pH and resin extractable PO 4 3 were measured. I conducted partial correlations between soil variables (%C, %N, C:N, DIN, DON, DIN:DON, chloroform labile microbial 15 N, and pH) for each depth while controlling for the effect of time since fire. I conducted a multiple regression with root biomass at the dependent variable and DIN, DON, and soil %N as ind ependent variables. Root biomass data were square root transformed before analyses. All other data were natural log transformed. I calculated microbial biomass N using the equation B N = F N / K EN where B N = microbial biomass N, F N = chloroform labile N, an d K EN = 0.54, the efficiency of the fumigation (Brookes et al. 1985). I used 0.54 for K EN because this value was determined from K 2 SO 4 extractions of a variety of soils (Brookes et al. 1985), and because this value is in the m iddle of the range of previous estimates of K EN (Shen et al. 1984, Jonasson et al. 1996). I used linear regression to analyze the relationship between microbial biomass N and DON. I conducted a multiple regression with N mineralization

PAGE 110

110 rates with H 2 1 day 1 ) as the dependent variable and microbial biomass N, DON, C:N, and %N as independent variables. I conducted a multiple 1 ) as the dependent variable and microbial biomass N, DON, N mineralization, and C:N as independent vari ables. All variables were natural log transformed before analysis. Data were analyzed using SPSS 11.5. Results Species C omposition Composition of resprouting shrubs was similar among plots (Table 4 2). Lyonia fruticosa Quercus chapmanii Quercus geminata Quercus inopina and Serenoa repens occurred in all twelve plots, while Lyonia lucida Palafoxia feayi and Sabal etonia were present in eleven plots. The two most common graminoid species, Aristida beyrichiana and Rhynchospora megalocarpa were present in 92% and 83% of the plots, respectively. A nitrogen fixing species was present in all 1 yr, 4 yr, and 6 yr plots ; whereas, a nitrogen fixing species was present in only one 8 yr, 10 yr, and 13 yr plot Herbaceous species that recruit from seed after fir e were sparse in general, but a greater number of the most common species occurred at intermediate times after fire. Lichens did not colonize plots until at least eight years after fire. Pines were found only in plots that had burned once (Table 4 2 ). Sand pine, Pinus clausa was present in both 13 yr plots and in one 8 yr plot Species composition of the plots where soil samples were not collected was similar to the ot her plots of the same time after fire ex cept for the 13 yr plot in burn unit 24A (Table 4 1) where a few Pinus elliottii var. densa individuals were found.

PAGE 111

111 Root B iomass Root ( < 2 mm diameter) biomass (g m 2 ) tended to increase with time after fire in surface soils (0 5 cm) (Table 4 3) and varied with depth in 4 yr, 6 yr, and 8 yr sites (Tab le 4 4 ; Table 4 5 ). In surface soils, root biomass was 3.0 and 2.4 times greater in 13 yr sites than in 4 yr and 6 yr sites respectively (Figure 4 1 ). Total root biomass per 20 cm depth ranged from 1938 g m 2 in 4 yr sites to 2874 g m 2 in 10 yr sites At all soil depths, soil %N was positively correlated with root biomass (Figure 4 2 ). Dissolved inorganic and organic N, however, were not significant predictors of root biomass (Table 4 6 ). Resin Exchangeable N utrients Over one year, resin exchangeable NH 4 + (F 5,12 = 2.28, p = 0.112) and total inorganic N (F 5,12 = 2.37, p = 0.102) were highest in 8 yr sites, but did n ot differ significantly across the time after fire chronosequence. By contrast, resin exchangeable NO 3 (F 5,12 = 6.17, p = 0.005) and PO 4 3 (F 5,12 = 3.46, p = 0.036) decrease d then increased with time after fire (Figure 4 3 ). Resin exchangeable NO 3 and PO 4 3 were both five times greater in 13 yr than in 6 yr sites. Resin exchangeable N:P ratios increase d then decreased with time after fire (F 5,12 = 7.82, p = 0.002), and were approximately four times lower in 13 yr sites compared to 6 yr and 8 yr sites (Figure 4 3 C). Variation in N:P ratios with time after fire was primarily driven by changes in P. Resin exchangeable N was positively correlated with resin exchangeable P (r = 0.199, p = 0.031). Mean resi n exchangeable PO 4 3 over one year was negatively correlated with mean pH of surface soils (Figure 4 4 ; r = 0.569, p = 0.053).

PAGE 112

112 Nitrogen Pools and F luxes Time after fire had a significant effect on dissolved inorganic N (DIN) only in deep soils (15 20 cm); whereas, dissolved organic N (DON) varied with time after fire in all soils except deep soils (Table 4 3). In deep soils, DIN tended to be higher in recently burned and longer unburned sites tha n in s ites at intermediate times after fire (Figure 4 5 D). In surface soils (0 5 cm) DON was 2.2 times greater in 13 yr than in 6 yr sites, and in 5 10 cm soils, DON was 1.5 times greater in 4 yr than 1 yr sites (Figure 4 5 E). Across all times since fire, DIN was 2.0 to 2.9 times greater in surface soils (0 5 cm) than in deep soils (Figure 4 5 D). Significant differences in DON among depths were not consistent among times after fire, but DON tended to be lowest in the deepest soils (Figure 4 5 E). The ratio of DIN: DON did not vary with time after fire (Table 4 4). Across all times after 1 mean 1 and mean DIN:DON was always less than one. Potential N 1 day 1 ) measured in the lab did not vary with time after fire in deeper soils (>5 cm) at field capacity (Table 4 3). In surface soils (0 5 cm) N immobilization occurred in 1 yr sites while N mineralization o ccurred at all other times after fire; N mineralization was 2.5 to 55 times greater in 13 y r sites than in other sites (Figure 4 6 A). Potential N mineralization tended to be highest in surface soils (Table 4 4 ; Table 5 5 ). 1 day 1 ) were positively correlated in surface soils (0 5 cm) in 4 yr sites (r = 0.678, p = 0.045) and 13 yr sites (r = 0.784, p = 0.007), which caused there to be no differences in N mineraliz 1 day 1 ) with time after fire in surface soils (Table 4 3) or with depth (Table 4 4 ; Table 4 5 ). Similar to N mineralization, nitrification rates

PAGE 113

113 3 g soil 1 day 1 ) tended to be highest i n surface soils (0 5 cm) (Figur e 4 6B). Differences with time after fire, however, were significant at intermediate soil depths (5 10 cm and 10 15 cm) (Table 4 3) and not in surface soils. The effect of w ater addition on N mineralization 1 day 1 ) and nitrification 3 g soil 1 day 1 ) depended on time after fire and soil depth (Figure 4 7 ). In surface soils (0 5 cm), water addition increased N mineralization in 4 yr and 6 yr sites, and in 5 10 cm soils, water addition increased N mineralization in 6 yr, 8 yr, and 10 yr sites. In deep soils (15 20 cm), water addition caused N immobilization in 1 yr and 4 yr sites (Figure 4 7 ). Chloroform labile microbial N (CLMN) was highest in long unburned sites (Table 4 3). In surface soils (0 5 cm), CLMN was 2.3 to 3.0 times grea ter in 13 y r sites than all other times after fire (Figure 4 5 G). CLMN decreased with depth at all times after fire (Table 4 4 ; Table 4 5), and on average, was 2.5 to 5.6 times greater in surface soils than in deep soils (Figure 4 5 G ). Across all times aft er fire and depths, mean CLMN 1 Bulk Soil P roperties In surface soils (0 5 cm), soil %C was 2.0 to 3.4 times greater in 13 yr sites than other times after fire (Figure 4 5A), soil %N was 2.5 to 4.5 times greater in 13 yr sites than other t imes after fire (Figure 4 5B), and s oil C:N ratios were 1.29 and 1.36 times greater in 13 y r sites than in 1 yr and 6 yr sites respectively (Figure 4 5 C) Soil %C and %N were significantly higher in surface soils than in deep soils (15 20 cm) at all times after fire, yet soil C:N varied with depth only in 6 yr sites (Table 4 4 ; Table 4 5 ). Similar to soil %C and %N, surface soil C pools (g C m 2 ) and N pools (g N m 2 ) were highest in long unburned sites (Table 4 3). C arbon pools were 2.5 and 3.4 times greater in 13 y r

PAGE 114

114 sites than in 8 yr and 6 yr sites, respectively (Figure 4 8A), and N pools were 2.5 times greater in 13 y r sites than in 6 yr sites (Figure 4 8 B) Carbon and N pools varied with depth, but these differences were not as strong as or as cons istent as differences in soil %C and soil %N because of variation in bulk density (Table 4 4 ; Table 4 5 ). Total C pools per 20 cm depth ranged from 1285 g m 2 in 6 yr sites to 3729 g m 2 in 13 yr sites Total N pools per 20 cm depth ranged from 45.6 g m 2 in 6 yr sites to 93.2 g m 2 in 13 yr sites Differences in soil pH with time since fire depended on soil depth (Table 4 3) and were most pron ounced in the 0 5 cm soil layer where soil pH was approximately 15% greater in 1 yr, 4 yr, 6 yr, and 10 yr sites compared to 13 y r sites (Figure 4 5 I). In 1 yr, 4 yr, and 6 yr sites soil pH was highest in surface (0 5 cm) soils. Differences in soil pH with depth were most pronounced one year after fire, when mean pH was at least 0.44 units greater in surface soils (0 5 cm) than in any other soil depths. 15 N tended to increa se then decrease with time after fire for surface (0 5 cm) and 10 15 cm soils (Table 4 15 N was negative in 1 yr sites and p ositive at all other times after fire (Figure 4 5 H), but was only significantly different between 1 yr and 8 yr sites. At all times after 15 N was lowest in surface soils (Table 4 4 ; Table 4 5 15 N was more enriched in 5 10 cm soils than in surface so ils, increasing 1.2 (Figure 4 5 H). Relationships Among Soil V ariables Dissolved organic N (DON) and dissolved inorganic N (DIN) were positively correlated with soil %C and %N (Table 4 7). DON was positively correlated with CLMN (Table 4 7) and m icrobial biomass N (Figure 4 9) for all soil depths except 5 10 cm Across soil depths, microbial biomass N was the only consistent predictor of DIN (Table

PAGE 115

115 4 8). In surface soils (0 5 cm), DON was the only significant predict or of N mineralization rates; whereas, in sub sur face soils (5 10 cm), microbial biomass N was the only significant predictor of N mineralization rates (Table 4 9 ). Soil %N was positively correlated with soil %C, while CLMN was positively correlated with both soil %C and %N. Both soil %N and CLMN were ne gatively correlated with soil pH (Table 4 7). Discussion Over one year, PO 4 3 was highest in recently burned and long unburned sites. Total inorganic N was highest at intermediate times after fire, but did not vary significantly. This caused N:P ratios t o be greatest at intermediate times after fire and lowest in long unburned sites. In surface soils, soil %C and %N, C and N pools, dissolved organic N (DON), net N mineralization, and chloroform labile microbial N were all highest in long unburned sites. Effects of T ime A fter F ire on S oil C haracteristics and N utrient A vailability Inorganic N availability, measured as resin exchangeable N or K 2 SO 4 extractable N, did not vary significantly with time after fire in scrubby flatwoods. In many ecosystems, post fire increases in NH 4 + and NO 3 persist for only five months or less (Schmidt and Stewart 1997; Jensen et al. 2001; Bennett et al. 2002). In palmetto flatwoods, which contains some of the same species as scrubby flatwoods but has more mesic soils, the post fire pulse of NH 4 + persisted for less than three months (Chapter 2). Recovery of aboveground biomass, which occurs within months in scrubby flatwoods, may contribute to short lived increases in nutrient availability (Hobbs and Schimel 1984). Thus, fire in duced changes in inorganic N availability may not persist for a year after fire. While fire induced changes in inorganic N availability occur over short time

PAGE 116

116 scales (Chapter 2), fire appears to have little direct effect on differences in N availability ove r longer time scales in scrubby flatwoods. Net N immobilization occurred in surface soils (0 5 cm) of recently burned sites. Turner et al. (2007) also measured net N immobilization after fire, which, in their case, may be due to the N deficiency of the sit e and high quantities of low N wood remaining on the soil surface after fire. An increase in net N mineralization shortly after fire (Hobbs and Schimel 1984; Adams and Attiwill 1991; Kaye and Hart 1998) due to the positive correlation between N mineralizat ion and soil temperature (Wilson et al. 2002) would not have been detected in my lab incubations. Nitrogen soil 1 day 1 ) in surface soils (0 5 cm) were highest in long unburned sites, likely because dissolved organic N (DON) was highest in long unburned sites and mineralization rates were positively correlated with DON (Table 4 9 ). Higher N mineralization rates in longer unburned sites may also be related to the positive correlation between N mineralization and soil %N (Marion and Black 1988; Carreira et al. 1994). In addition, N mineralization decreases with an increase in fire frequency (Reich et al. 2001), and all of my 13 yr sites have burned only once over the past 35 years. Similarly to N mineralization, dissolved organic N (DON) and chloroform labile microbial N (CLMN) in surface soils were highest in long unburned sites. DON and CLMN were positively correlated with soil %N and %C, which were also highest in long unburned sites. Furthermore, CLMN was a significant predictor of DON. Fire can kill soil microbes, and the effects of fire on soil temperature decrease with depth (Ewel et al. 1981; Giardina et al. 2000; Jensen et al. 2001), so the effects of fire should be greatest

PAGE 117

117 in surface soils. The lack of a difference in CLMN from one to ten years after fire suggests that fire had limited direct effects on CLMN. In contrast t o my results, MacKenzie et al. (2006) found that microbial activity declined with time after fire. Soil %N and total N pools (g m 2 ) in surface soils (0 5 cm) tended to be greater in recently burned and long unburned sites than in sites at intermediate times after fire. This pattern is likely related to changes in litter quality feedbacks, N inputs, and accumulation of organic matter with time after fire. The first leaves flushed after fire have higher N concentrations than leaves pre fire (Chapter 2), a nd incorporation of these leaves to soil organic matter may contribute to high soil %N and N pools in recently burned sites. Both an increase in the recalcitrance of litter and a decrease in photodegradation of litter (Austin and Vivanco 2006), caused by d ecreased light availability over time, could lead to N accumulation in long unburned sites. Over the short term, inputs of N through symbiotic N fixation and rainfall are not high enough to replace the amount of N volatilized in fire (Carter and Foster 200 4; Cook 1994), but over longer time intervals, N fixation can lead to an accumulation of N in the soil (Bormann and Sidle 1990). Soil crusts in Florida scrub have a high N fixing capacity, which decreases immediately after fire (Hawkes 2003). The abundance of photosynthetic microbes present in soil crusts peaks at intermediate times since fire (10 15 years) in rosemary scrub (Hawkes and Flechtner 2002) and declines in long unburned sites likely because of increased litter cover and light limitation. Non sym biotic N 2 fixation increases with time since fire (Prez et al. 2004), so high N fixing capacity of soil crusts may contribute to high soil %N and large soil N pools in long unburned sites. In addition, N

PAGE 118

118 fixing plants, which contribute N inputs to scrubby flatwood soils, occurred across the scrubby flatwoods time after fire chronosequence (Table 4 1). Soil %N was correlated with soil %C (Table 4 7 ), suggesting that accumulation of organic matter may regulate N availability and soil N concentrations. Yerma kov and Rothstein (2006) found that total soil N increased with time after fire due to an increase in organic soil N with time after fire. In my study, 13 yr scrubby flatwoods sites had more organic matter than any other time after fire (J. Schafer, person al observation). In addition, total soil N may decrease with an increase in fire frequency (DeLuca and Sala 2006; Cech et al. 2008), and all of my 13 yr sites have burned only once over the past 35 years (Table 4 1). 15 N was more enriched at intermediate times since fire compared to recently burned and long unburned sites (Figure 4 5H). This suggests that N losses are greater at intermediate times after fire (Martinelli et al. 1999). Resin exchangeable N was high est at intermediate times after fire, but N availability was not significantly different among times after fire. Fire consumes surface soil layers and volatilizes N, which can leave post fire soils enriched in 15 N (Hgberg 1997); however, soils were not en riched in 15 N after fire likely because scrubby flatwoods soils generally have little soil organic matter. Fractionation during the transfer of N from mycorrhizal fungi to a host plants results in plant tissue depleted in 15 N relative to the N source (Evan s 2001; Hobbie and Colpaert 2003). Soil %C and soil C pools (g m 2 ) followed the same pattern as soil %N and soil N pools; soil %C and C pools were highest in long unburned sites, and tended to be higher in recently burned sites than in sites at interme diate times since fire. Soil %C and

PAGE 119

119 C content are high in recently burned sites likely due to the presence of charcoal, because soil C is positively correlated with charcoal C (MacKenzie et al. 2008). Soil charcoal has a lower C concentration than recently produced charcoal, suggesting that the C content of charcoal decreases over time (Ohlson et al. 2009), and that the contribution of charcoal C to soil %C decreases with time after fire. Carbon accumulation over time after fire (MacKenzie et al. 2004; Varg as et al. 2008) is likely related to litter accumulation, which is positively correlated with canopy cover (Hall et al. 2006), and the transformation of plant material into soil organic matter, which increases over time after fire (Treseder et al. 2004). I n fact, total soil C can increase due to an increase in organic soil C with time after fire (Yermakov and Rothstein 2006). Thus, as organic material accumulates in long unburned scrubby flatwoods, soil %C and C content also increase. In surface soils, soi l pH tended to be higher in more recently burned sites than in longer unburned sites. Ash on the soil surface after fire may increase soil pH (Grogan et al. 2000; Bada and Mart 2003; Molina et al. 2007) due to the high pH of ash (Jensen et al. 2001; Gofo rth et al. 2005; Molina et al. 2007; Marcos et al. 2009) and the high concentration of cations, such as Ca 2+ and K + in ash (Raison et al. 1985b; Arocena and Opio 2003). Persistence of high soil pH with time since fire is likely related to persistence of a sh in the soil. Resin exchangeable PO 4 3 tended to be higher in recently burned and long unburned sites than intermediate times after fire (Figure 4 3 ). High PO 4 3 availability in recently burned sites is likely due to high concentrations of P in ash post fire (Wilbur and Christensen 1983; Raison et al. 1985b), since mineral soil PO 4 3 is correlated with

PAGE 120

120 ash depth (Rice 1993). Microbial immobilization or plant uptake may contribute to the decrease in resin exchangeable PO 4 3 at intermediate times since fir e. Furthermore, PO 4 3 may become bound in calcium phosphate (Hsu and Jackson 1960; Stephens et al. 2004) due to high concentrations of calcium (Ca 2+ ) in ash (Ewel et al. 1981; Kauffman et al. 1993). There are several explanations for the increase in resin exchangeable PO 4 3 in long unburned sites. First, phosphatase activity is highest when plants are fertilized with organic P (DeLucia et al. 1997), and increased phosphatase activity in long unburned sites, which had high amounts of soil organic matter, cou ld increase PO 4 3 availability. Second, soil phosphatase activity (Chen et al. 2004) and the quantity and quality of root exudates (Grayston et al. 1996) varies among plant species, and the biggest difference in plant species composition across my time aft er fire chronosequence was the presence of pines ( Pinus clausa and Pinus elliottii ) in long unburned sites. Several Pinus species, including Pinus elliottii produce root exudates (Agnihotri and Vaartaja 1969; Fox and Comerford 1990; van Schll et al. 2006 ) such as low molecular weight organic acids, which can bind cations covalently linked to P, liberating inorganic P (Jurinak et al. 1986; DeLucia et al. 1997). Third, an increase in the abundance of lichens with time after fire (Menges and Kohfeldt 1995) m ay affect P availability if lichens differ from other scrubby flatwood species in their effect on soil pH or their production of organic acids. Fourth, fire frequency may have an impact on P availability. Availability of PO 4 3 is lower in sites with more f requent fires (Hernndez and Hobbie 2008), and 13 yr sites had burned only once in the last 35 years. Resin exchangeable N was highest in 8 yr sites, while resin exchangeable PO 4 3 was higher in recently burned and long unburned sites. Thus, resin exchang eable N:P

PAGE 121

121 ratios increased from one to eight years after fire, then decreased from eight to 13 years after fire. My results suggest that N limitation may be greater in recently burned and longer unburned sites and that P limitation may be greater at interm ediate times after fire; however, soil N:P ratios have not been used to indicate N limitation, P limitation, or co limitation by N and P as have foliar N:P ratios (Koerselman and Meuleman 1996; Gsewell 2004). The relationship between soil N:P ratios and n utrient limitation is unclear because foliar N and P concentrations are not always correlated with soil N and P availability (Frank 2008; Litaor et al. 2008). In many cases, there were differences in soil characteristics and nutrient availability between plots within a time after fire (Table 4 3 ), suggesting that differences in fire intensity and fire frequency affect nutrient availability. High fire intensity causes greater availability of soil N and P than moderate or low intensity fires (Gimeno Garca e t al. 2000; Kennard and Gholz 2001; Romany et al. 2001), and there may be little difference in soil properties between unburned sites and sites burned by low severity fires (Hatten et al. 2005). Fire intensity is positively correlated with ash depth (Rice 1993) and high intensity fires cause a greater increase in soil pH (Kennard and Gholz 2001), which affects P availability (Jaggi et al. 2005). Furthermore, losses of P may be greater in high intensity fires (Gimeno Garca et al. 2000). Fire intensity can be highly variable over distances of only 10 m (Rice 1993), so variation in fire intensity among my plots (Table 4 1) likely contributed to variation in soil nutrients. Nitrogen availability (Hernndez and Hobbie 2008) and N mineralization (Reich et al. 20 01) decrease with an increase in fire frequency, while nutrient losses increase with an increase in fire frequency (Wanthongchai et al. 2008). Differences in species composition can affect the

PAGE 122

122 nutrient content of ash (Qian et al. 2009), and greater plant b iomass can lead to an increase in fire severity (Romany et al. 2001). Thus, differences in fire history among scrubby flatwoods sites may cause variation in nutrient availability with time after fire. Variation in S oi l Characteristics and Nutrient A vailab ility with Soil D epth Differences in nutrient availability and bulk soil characteristics with depth were relatively consistent among times after fire. Soil %N, soil %C, dissolved inorganic N (DIN), chloroform labile microbial N (CLMN), N mineralization, an d nitrification tended to 15 N was more negative in surface soils (Figure 4 5; Figure 4 6 ). Concentrations of NH 4 + and NO 3 (Fenn et al. 1993) and net ammonification rates (Bloom and Mallik 2006) are affected by the identity and cover of plant species, respectively, and soil %N is higher under rather that outside the plant canopy (Aguilera et al. 1999). It is likely that the influence of plant leaf litter is greatest in surface soils, which may contribute to higher DI N, soil %N, and CLMN in surface soils. Soil N and soil C are positively correlated with charcoal content (MacKenzie et al. 2008), so persistence of charcoal in surface soils may contribute to high soil %N and %C. Because plants are more depleted in 15 N tha n their N source (Chapter 2; Michelsen et a l. 1998, Schmidt and Stewart 200 3), surface soils depleted in 15 N reflect inputs of plant litter to soil organic matter. Patterns in soil pH with depth depended on time after fire. Soil pH was higher in surface s oils (0 5 cm) than in deeper soils only in recently burned sites (one to six years after fire). Ash on the soil surface may increase soil pH (Grogan et al. 2000; Bada and Mart 2003; Molina et al. 2007). Leaching of ash covered soils increases soil pH (Mo lina et al. 2007), so it is likely that as ash becomes integrated through the soil profile over time after fire, soil pH becomes more similar across depths.

PAGE 123

123 Effects of A b iotic Factors on Soil Nutrient A vailability In scrubby flatwood surface soils, N mine ralization and nitrification rates tended to be higher in soils at field capacity than in soils at ambient field moisture conditions (Figure 4 7 ), likely due to the positive effects of soil moisture on microbial biomass and activity. Abundance of soil bact eria and fungi is higher after the wet season than after the dry season (Aguilera et al. 1999); in litter subjected to wetting and drying, microbial biomass is higher after rewetting (Schimel et al. 1999); and microbial biomass C is positively correlated w ith soil moisture (Tate and Terry 1980). Furthermore, dehydrogenase activity, a measure of microbial activity, is positively correlated with soil moisture (Tate and Terry 1980; Paradelo and Barral 2009). In deeper scrubby flatwoods soils, however, the effe ct of water addition on N mineralization and nitrification rates depended on time since fire (Figure 4 7 ). Increased soil moisture can inhibit growth of bacteria deeper in the soil profile (Tate and Terry 1980), and specific respiration increment can be ne gatively correlated with soil moisture (Meril and Ohtonen 1997), suggesting that too much moisture may limit oxygen, and thus microbial activity. Across a variety of ecosystems, net mineralization is positively correlated with precipitation (Prez et al. 2004) and soil moisture (Cassman and Munns 1980; Powers 1990; Evans et al. 1998; Frank 2008), but soil may have a maximum mineralization capacity above which more water does not increase mineralization (Cassman and Munns 1980). Resin exchangeable PO 4 3 w as negatively correlated with pH of surface soils (Figure 4 4 ). Over one year, resin exchangeable PO 4 3 was highest in sites with the most acidic soil. Phosphatase activity is negatively correlated with soil pH (Sinsabaugh et al. 2008), and low phosphatase activity leads to low PO 4 3 availability. Frank (2008) found that soil pH (ranging from 6.3 to 7.9) was negatively correlated with net P

PAGE 124

124 mineralization in grasslands, so it is not surprising that the change in mean pH from 3.95 to 4.98 across my scrubby fl atwoods sites had significant effects on P availability. Furthermore, PO 4 3 can form minerals with Ca 2+ and high concentrations of Ca 2+ in ash may limit mobility and availability of PO 4 3 in more basic soils (Hsu and Jackson 1960). Conclusion In scrubby f latwoods soils, inorganic N availability is not affected by time after fire, but is related to soil moisture, while variation in PO 4 3 availability is related to soil pH. Soil %N, soil %C, N mineralization rates, dissolved organic N, chloroform labile micr obial N, and PO 4 3 availability were highest in long unburned sites. Variation in measures of nutrient availability may be due to the differences in species composition and fire history of 13 yr sites compared to scrubby flatwoods sites of other times afte r fire rather than the effects of fire. In this study, 13 yr sites had numerous pine trees, high soil organic matter, and had only been burned once in the past 35 years, which may have contributed to high PO 4 3 soil %N, and soil %C in surface soils; howev er, in scrubby flatwoods sites 20 years after fire with low soil organic matter, and where pines are less abundant or absent, soil %N and %C are lower (J. Schafer, unpublished data). My 1 yr to 10 yr sites are characteristic of scrubby flatwoods sites in g eneral; whereas, my 13 yr sites differ from the characteristic scrubby flatwood sites. Thus, my results indicating high nutrient availability in long unburned scrubby flatwoods may not apply to all scrubby flatwoods. Species composition and fire frequency, as well as time after fire, appear to be important in affecting nutrient availability in Florida scrub soils.

PAGE 125

125 Table 4 1. Description of study sites. Fire intensity 1 = low intensity, 2 = intermediate intensity, 3 = high intensity. The total number of bu rns is based on the record of fire history at Archbold Biological Station ( 197 0 present) and includes the most recent burn after which measurements were made. Multiple numbers in a column indicate that there may have been differences in fire intensity or fire history within the plot area. Intensive sampling indicates plots where soil samples were collected and species composition was assessed. Time after fire (years) Date of Burn Burn Unit Area (acres) Fire Intensity Total # of Burns Intensive Sampling 1 22 June 2004 54B 49 3 3 X 1 22 June 2004 54C 53 3 1,2 1 24 May 2004 61A 40 3 3 X 4 12 Feb 2001 44 163 1,3 2 X 4 12 Feb 2001 45C 81 3 3,4 4 12 Feb 2001 47A 67 3 3 X 6 7 July 1999 48B 100 1 3 2 X 6 7 July 1999 48B 100 3 1 6 26 July 1999 56B 27 2 3 3 X 8 7 May 1997 55 117 2 3 1 X 8 7 May 1997 55 117 3 1,2 8 7 May 1997 55 117 3 1 X 10 15 June 1995 49B 76 3 3,4 X 10 15 June 1995 49B 76 2 3 3,4 10 24 May 1995 56C 24 3 2 X 13 19 May 1992 25B 5 2 3 1 X 13 20 May 1992 24A 12 2 3 1 13 9 Dec 19 92 31B 15 1 3 1 X

PAGE 126

126 Table 4 2. Species composition (in 2006) of the twelve plots where soil samples were collected (in 2005). Values are the number of plots in which species occurred except for the total number of species found in plots of each t ime since fire. Only shrubs and sub shrubs present in at least two plots are included. Only herbs present in at least three plots are included. Species identifiable to genus only are included if present in at least three plots. Years since fire in 2005 (in 2006) Species Family 1 (2) 4 (5) 6 (7) 8 (9) 10 (11) 13 (14) Total Trees Pinus clausa Pinaceae 1 2 3 Shrubs Asimina obovata Annonaceae 1 1 2 Bejaria racemosa Ericaceae 1 2 2 5 Ilex opaca var. arenicola Aquifol iaceae 1 1 1 3 Lyonia ferruginea Ericaceae 1 1 2 Lyonia fruticosa Ericaceae 2 2 2 2 2 2 12 Lyonia lucida Ericaceae 2 2 1 2 2 2 11 Palafoxia feayi Asteraceae 1 2 2 2 2 2 11 Persea borbonia var. humilis Lauraceae 1 1 2 Quercus champanii Fa gaceae 2 2 2 2 2 2 12 Quercus geminate Fagaceae 2 2 2 2 2 2 12 Quercus inopina Fagaceae 2 2 2 2 2 2 12 Sabal etonia Arecaceae 2 2 2 1 2 2 11 Serenoa repens Arecaceae 2 2 2 2 2 2 12 Sideroxylon tenax Sapotaceae 1 1 2 Ximenia americana ^ Olacacea e 2 2 2 1 1 8 Suffrutescent/Sub shrubs Gaylussacia dumosa Ericaceae 1 2 2 2 1 8 Helianthemum nashii Cistaceae 1 1 1 3 Lechea deckertii Cistaceae 1 2 2 2 1 1 9 Licania michauxii Chrysobalanaceae 1 2 2 1 1 1 8 Opuntia humifusa Cactaceae 1 1 2 2 1 1 8 Polygala polygama Polygalaceae 2 1 2 1 1 7 Smilax auriculata Smilacaceae 2 1 2 2 2 2 11 Vaccinium darrowii Ericaceae 1 1 1 2 5 Vaccinium myrsinities Ericaceae 2 2 2 2 2 2 12 N fixers Chapmanii floridana Fabaceae 1 1 2 1 5 Galactia regularis/elliottii Fabaceae 1 2 1 1 1 6 Mimosa quadrivalvis Fabaceae 1 1 1 3 Graminoids Andropogon floridanus Poaceae 2 1 1 4 Aristida beyrichiana Poaceae 2 2 2 2 2 1 11 Dicanthelium sp. Poaceae 2 1 3 Rhynchospora meg alocarpa Cyperaceae 2 2 2 1 2 1 10

PAGE 127

127 Table 4 2 continued. Years since fire in 2005 (in 2006) Species Family 1 (2) 4 (5) 6 (7) 8 (9) 10 (11) 13 (14) Total Herbs Ambrosia artemisiifolia Asteraceae 1 1 1 3 Balduina angustifolia Astera ceae 1 1 1 3 Cnidoscolus stimulosus Euphorbiaceae 1 1 1 1 4 Commelina sp. Commelinaceae 1 1 1 1 1 1 6 Liatris ohlingerae Asteraceae 1 1 1 3 Paronychia chartacea Caryophyllaceae 1 1 1 3 Other Selaginella arenicola Selaginellaceae 1 2 2 2 2 1 10 Lichens ( Cladina sp., Cladonia sp.) 1 2 2 5 Total # of species 28 28 32 33 27 28 40 endemic to Florida ^ hemi parasitic

PAGE 128

128 Table 4 3. Results of nested analyses of variance of soil variables at each depth. Degrees of freedom = 5 for YSF and 6 for P lot(YSF). Soil Depth (cm) 0 5 5 10 10 15 15 20 F p F p F P F p Root Biomass (g/m 2 )^ YSF 3.09 0.017 0.87 0.510 0.98 0.440 0.79 0.558 Plot(YSF) 1.89 0.101 3.49 0.006 1.68 0.146 0.83 0.555 pH YSF 8.29 <0.001 1.85 0.121 2.55 0.040 2.33 0.057 Plot(YSF) 2.64 0.027 2.61 0.028 3.72 0.004 4.10 0.002 %N* YSF 6.58 <0.001 1.38 0.248 0.39 0.855 1.60 0.178 Plot(YSF) 1.28 0.282 1.66 0.152 2.61 0.029 0.97 0.453 %C* YSF 6.50 <0.001 0.99 0.435 0.24 0.944 1.82 0.127 Plot(YSF) 0.92 0.486 1.21 0.319 2.91 0.017 1.38 0.243 CN* YSF 3.07 0.017 0.70 0.627 0.29 0.917 1.53 0.197 Plot(YSF) 1.40 0.235 1.11 0.372 2.88 0.017 1.75 0.129 N ( g m 2 )* YSF 3.75 0.006 1.37 0.250 1.20 0.324 1.54 0.194 Plot(YSF) 1.28 0.282 1.66 0.152 2.61 0.029 0.97 0.453 C ( g m 2 )* YSF 6.14 <0.001 4.06 0.004 2.30 0.060 3.91 0.005 Plot(YSF) 0.71 0.646 5.18 <0.001 4.13 0.002 0.81 0.565 15 N YSF 2.52 0.042 1.05 0.402 2.74 0.030 0.23 0.947 Plot(YSF) 1.41 0.231 2.01 0.083 2.25 0.054 1.61 0.165 Inorganic N (NH 4 + + NO 3 ) 1 ) YSF 0.66 0.657 0.41 0.837 1.51 0.205 2.51 0.043 Plot(YSF) 1.12 0.362 0.74 0.618 2.16 0.063 2.38 0.043 Dissolved Organic N* 1 ) YSF 2.46 0.046 3.48 0.009 4.33 0.003 1.57 0.187 Plot(YSF) 1.63 0.159 0.57 0.753 1.17 0.338 1.10 0.373 DIN:DON* 1 ) YSF 1.50 0.208 1. 48 0.213 0.92 0.478 1.19 0.328 Plot(YSF) 1.57 0.176 0.98 0.450 1.40 0.234 2.14 0.066 Chloroform labile microbial N* 1 ) YSF 6.14 <0.001 4.06 0.004 2.30 0.060 3.91 0.005 Plot(YSF) 0.71 0.646 5.18 <0.001 4.13 0.002 0. 81 0.565 N mineralization (+ H 2 O)* 1 day 1 ) YSF 3.42 0.010 1.20 0.322 2.02 0.093 1.82 0.127 Plot(YSF) 0.64 0.698 2.86 0.019 1.67 0.150 1.02 0.425 N mineralization (+ H 2 O) 1 day 1 ) YSF 0.85 0.524 0.97 0.443 2.53 0.042 2.42 0.049 Plot(YSF) 1.19 0.330 3.42 0.007 2.97 0.015 1.90 0.100

PAGE 129

129 Table 4 3 continued. Soil Depth (cm) 0 5 5 10 10 15 15 20 F p F p F P F p Nitrification (+ H 2 O)* 1 day 1 ) YSF 1.70 0.154 3.47 0.009 2.73 0.030 0.84 0.526 Plot(YSF) 1.53 0.189 2.73 0.023 0.96 0.460 1.07 0.395 data were natural log transformed before analyses ^ data were square root transformed before analyses

PAGE 130

130 Table 4 4 Results of repeated measures analyses of d ifferences with depth for sites one, four, and six years after fire 1 year after fire 4 years after fire 6 years after fire F df p F df p F df p Root Biomass^ ( g m 2 ) Depth 1.27 3 0.306 6.34 3 0.003 3.32 3 0.037 Plot 7.34 1 0.027 0 .22 1 0.651 14.61 1 0.005 Depth Plot 1.67 3 0.201 0.89 3 0.457 0.21 3 0.887 pH Depth 8.87 1.56 0.006 3.33 3 0.036 5.22 3 0.006 Plot 0.001 1 0.981 3.05 1 0.698 13.22 1 0.007 Depth Plot 0.46 1.56 0.594 0.48 3 0.119 5.38 3 0.006 %C* Depth 10.90 3 <0.001 10.34 3 <0.001 11.98 1.69 <0.001 Plot 0.42 1 0.532 1.02 1 0.342 1.38 1 0.273 Depth Plot 5.77 3 0.004 0.69 3 0.564 4.69 1.69 0.033 %N* Depth 17.20 3 <0.001 16.57 3 <0.001 13.04 3 <0.001 Plot 0.65 1 0.443 2.71 1 0.138 0.46 1 0.514 Depth Plot 3.32 3 0.037 1.22 3 0.324 3.20 3 0.041 C:N* Depth 0.73 3 0.544 1.16 3 0.344 5.38 1.75 0.021 Plot 0.10 1 0.759 0.20 1 0.665 2.99 1 0.122 Depth Plot 7.34 3 0.001 0.89 3 0.460 3 .55 1.75 0.061 N ( g m 2 )* Depth 6.11 3 0.003 9.50 3 <0.001 2.68 3 0.069 Plot 0.65 1 0.443 2.71 1 0.138 0.46 1 0.514 Depth Plot 3.32 3 0.037 1.22 3 0.324 3.20 3 0.041 C ( g m 2 )* Depth 3.95 3 0.020 6.03 3 0.003 4.49 1.69 0.037 Plot 0.42 1 0.532 1.02 1 0.342 1.38 1 0.273 Depth Plot 5.77 3 0.004 0.69 3 0.564 4.69 1.69 0.033 Depth 6.60 3 0.002 4.40 3 0.013 23.69 3 <0.001 Plot 0.02 1 0.881 0.003 1 0.959 22.66 1 0.001 Depth Plot 1.16 3 0.346 0.09 3 0.965 3.89 3 0.021 DIN* Depth 17.76 3 <0.001 6.94 3 0.002 22.93 3 <0.001 Plot 0.16 1 0.696 3.90 1 0.089 0.28 1 0.609 Depth Plot 1.63 3 0.208 2.70 3 0.071 2.55 3 0.079 DON* Depth 1.41 3 0.265 6.12 3 0.004 3.32 3 0.037 Plot 0.24 1 0.639 1.32 1 0.287 0.63 1 0.449 Depth Plot 4.93 3 0.008 2.80 3 0.065 2.45 3 0.088 DIN:DON* Depth 4.02 3 0.019 2.72 3 0.070 7.97 3 0.001 Plot 0.02 1 0.875 1.58 1 0.249 0.07 1 0.801 Depth Plot 1.26 3 0.311 4.10 3 0.019 3.13 3 0.044 Microbial N* Depth 18.16 3 <0.001 23.43 3 <0.001 38.48 3 <0.001 Plot 1.95 1 0.200 10.62 1 0.014 8.23 1 0.021 Depth Plot 3.38 3 0.035 0.87 3 0.471 4.53 3 0.012

PAGE 131

131 Table 4 4 continued. 1 year after f ire 4 years after fire 6 years after fire F df p F df p F df p N min eralization 1 day 1 ) Depth 0.40 3 0.750 2.48 3 0.089 0.26 1.69 0.740 Plot 1.41 1 0.269 0.01 1 0.932 2.87 1 0.129 Depth Plot 0.10 3 0.958 1.34 3 0.287 1.19 1.69 0.325 N min eralization g N 1 day 1 ) Depth 1 .96 3 0.146 2.30 1.22 0.165 0.38 3 0.764 Plot 2.13 1 0.183 0.01 1 0.926 3.48 1 0.099 Depth Plot 0.98 3 0.418 1.52 1.22 0.258 0.72 3 0.549 Nitrification* Depth 1.73 3 0.188 4.80 1.46 0.039 1.19 1.09 0.310 Plot 2.44 1 0.156 0.11 1 0.752 1.78 1 0.219 Depth Plot 0.68 3 0.616 0.49 1.46 0.569 0.89 1.09 0.381 data were natural log transformed ^ data were square root transformed

PAGE 132

132 Table 4 5 Results of repeated measures analyses of differences with depth for sites eight, ten, and thirteen years after fire. 8 year after fire 10 years after fire 13 years after fire F df p F df p F df p Root Biomass^ ( g m 2 ) Depth 5.57 3 0.005 2.54 1.39 0.133 1.40 3 0.267 Plot 3.62 1 0.093 0.01 1 0.927 0.38 1 0.556 Dept h Plot 0.17 3 0.916 0.85 1.39 0.412 0.98 3 0.420 pH Depth 2.62 3 0.074 0.68 2.04 0.524 1.03 1.85 0.376 Plot 0.83 1 0.388 0.02 1 0.892 1.09 1 0.327 Depth Plot 26.89 3 <0.01 9.58 2.04 0.002 1.63 1.85 0.228 %C* Depth 11 .01 1.87 0.001 10.83 3 0.001 13.12 3 <0.001 Plot 12.37 1 0.008 0.31 1 0.594 0.18 1 0.988 Depth Plot 4.41 1.87 0.033 0.39 3 0.761 0.04 3 0.679 %N* Depth 21.00 3 <0.001 15.87 3 <0.001 28.82 2.02 <0.001 Plot 11.08 1 0.010 0.08 1 0. 781 0.08 1 0.783 Depth Plot 3.84 3 0.022 0.51 3 0.677 0.21 2.02 0.814 C:N* Depth 1.22 3 0.323 1.10 3 0.367 0.14 3 0.934 Plot 7.09 1 0.029 3.24 1 0.110 0.40 1 0.546 Depth Plot 4.99 3 0.008 0.06 3 0.977 0.16 3 0.921 N (g m 2 ) Depth 5.22 3 0.006 7.47 3 0.001 10.10 2.02 0.001 Plot 11.08 1 0.010 0.08 1 0.781 0.08 1 0.783 Depth Plot 3.84 3 0.022 0.51 3 0.677 0.21 2.02 0.814 C ( g m 2 )* Depth 3.28 1.87 0.069 5.43 3 0.005 4.91 3 0.008 Plot 12.37 1 0.008 0.31 1 0.594 0.18 1 0.679 Depth Plot 4.41 1.87 0.033 0.39 3 0.761 0.04 3 0.988 Depth 10.99 3 <0.001 17.02 3 <0.001 8.48 3 0.001 Plot 2.69 1 0.139 0.13 1 0.729 0.10 1 0.763 Depth Plot 0.21 3 0.888 2.25 3 0.108 1.11 3 0.364 DIN* Depth 10.03 3 <0.001 9.36 3 <0.001 8.78 3 <0.001 Plot 1.58 1 0 .244 0.08 1 0.785 0.65 1 0.443 Depth Plot 0.68 3 0.571 1.24 3 0.315 2.60 3 0.075 DON* Depth 5.13 1.43 0.034 24.09 3 <0.001 13.19 3 <0.001 Plot 2.13 1 0.183 0.14 1 0.713 0.01 1 0.922 Depth Plot 0.37 1.43 0.632 0.44 3 0.719 3.31 3 0.037 DIN:DON* Depth 3.56 1.77 0.060 0.80 3 0.503 0.25 3 0.857 Plot 0.20 1 0.662 <0.01 1 0.998 0.31 1 0.591 Depth Plot 0.30 1.77 0.721 1.67 3 0.199 4.99 3 0.008 Microbial N* Depth 33.80 1.90 <0.001 28.18 3 <0.001 25 .22 3 <0.001 Plot 22.64 1 0.001 0.03 1 0.876 0.05 1 0.836 Depth Plot 5.72 1.90 0.015 0.71 3 0.554 0.15 3 0.927

PAGE 133

133 Table 4 5 continued. 8 year after fire 10 years after fire 13 years after fire F df p F df p F df p N min eralization soil 1 day 1 ) Depth 0.67 3 0.578 0.55 3 0.650 4.85 3 0.009 Plot 10.29 1 0.012 4.07 1 0.078 0.06 1 0.811 Depth Plot 1.97 3 0.145 0.44 3 0.728 1.18 3 0.336 N min eralization g N 1 day 1 ) Depth 0.75 3 0.532 0.6 1 3 0.617 0.03 3 0.992 Plot 9.26 1 0.016 8.87 1 0.018 0.001 1 0.975 Depth Plot 2.40 3 0.093 1.48 3 0.244 1.96 3 0.147 Nitrification* Depth 4.71 3 0.010 2.36 3 0.096 1.12 3 0.362 Plot 4.65 1 0.063 0.01 1 0.925 3.74 1 0.029 De pth Plot 0.47 3 0.702 4.35 3 0.014 7.02 3 0.025 data were natural log transformed ^ data were square root transformed

PAGE 134

134 Table 4 6 Results of multiple regressions, with root biomass (g m 2 ) as the dependent variable and dissolved inorganic N (DI N), dissolved organic N (DON), and percent N, for each soil depth. Root biomass was square root transformed and all other variables were natural log transformed before analyses. B SE B p R 2 F df p 0 5 cm Constant 34.16 10.44 0.286 7.36 3,55 <0.001 DIN 0.77 2.81 0.035 0.783 DON 4.40 2.73 0.241 0.112 %N 5.17 2.45 0.330 0.039 5 10 cm Constant 49.78 7.54 0.226 5.36 3,55 0.003 DIN 0.68 2.23 0.037 0.762 DON 1.14 2.83 0.050 0.688 %N 6.10 1.68 0.450 0.001 10 15 cm Co nstant 52.04 6.85 0.328 8.93 3,55 <0.001 DIN 3.39 1.70 0.237 0.051 DON 0.19 2.22 0.011 0.933 %N 6.71 1.44 0.606 <0.001 15 20 cm Constant 50.34 8.99 0.256 6.31 3,55 0.001 DIN 0.95 1.78 0.069 0.593 DON 0.19 2.26 0.012 0.93 3 %N 6.36 1.88 0.523 0.001

PAGE 135

135 Table 4 7 Partial correlations between soil variables at each depth (controlling for time the diagonal. 0 5 cm %C* %N* C:N* DIN* DON* DIN:DON* CLMN* 15 N pH %C* .969 .730 .355 .617 .292 .774 .085 .496 %N* <0.001 .539 .460 .605 .196 .794 .024 .443 C:N* <0.001 <0.001 .062 .432 .456 .443 .222 .469 DIN* 0.003 <0.001 0.322 .352 .478 .431 .007 .222 DON* <0.001 <0.001 <0.001 0.003 .653 .52 2 .075 .351 DIN:DON* 0.013 0.071 <0.001 <0.001 <0.001 .141 .065 .149 CLMN* <0.001 <0.001 <0.001 <0.001 <0.001 0.146 .132 .510 15 N 0.263 0.427 0.047 0.480 0.287 0.313 0.162 .117 pH <0.001 <0.001 <0.001 0.047 0.003 0.131 <0.001 0.191 5 10 cm %C* %N* C:N* DIN* DON* DIN:DON* CLMN* 15 N pH %C* .947 .792 .105 .269 .098 .674 .357 .671 %N* <0.001 .553 .229 .242 .031 .704 .352 .597 C:N* <0.001 <0.001 .162 .238 .312 .409 .258 .603 DIN* 0.216 0.042 0.112 .226 .724 .272 .047 .1 34 DON* 0.020 0.034 0.036 0.044 .508 .192 .141 .168 DIN:DON* 0.233 0.407 0.009 <0.001 <0.001 .105 .141 .238 CLMN* <0.001 <0.001 0.001 0.019 0.074 0.217 .354 .568 15 N 0.003 0.003 0.025 0.364 0.146 0.145 0.003 .320 pH <0.001 <0.001 <0.001 0.15 7 0.104 0.036 <0.001 0.007 10 15 cm %C* %N* C:N* DIN* DON* DIN:DON* CLMN* 15 N pH %C* .957 .903 .228 .493 .156 .734 .518 .668 %N* <0.001 .739 .351 .469 .028 .733 .518 .630 C:N* <0.001 <0.001 .009 .445 .321 .617 .434 .616 DIN* 0.04 3 0.003 0.472 .261 .708 .366 .020 .077 DON* <0.001 <0.001 <0.001 0.024 .496 .441 .301 .470 DIN:DON* 0.121 0.418 0.007 <0.001 <0.001 .006 .202 .275 CLMN* <0.001 <0.001 <0.001 0.002 <0.001 0.481 .409 .542 15 N <0.001 <0.001 <0.001 0.442 0.011 0. 064 0.001 .364 pH <0.001 <0.001 <0.001 0.283 <0.001 0.018 <0.001 0.002 15 20 cm %C* %N* C:N* DIN* DON* DIN:DON* CLMN* 15 N pH %C* .939 .910 .402 .607 .118 .767 .482 .625 %N* <0.001 .713 .437 .617 .096 .799 .459 .522 C:N* <0.001 <0. 001 .294 .497 .126 .604 .430 .647 DIN* 0.001 <0.001 0.013 .208 .681 .459 .324 .294 DON* <0.001 <0.001 <0.001 0.058 .574 .541 .313 .417 DIN:DON* 0.188 0.237 0.173 <0.001 <0.001 .020 .036 .067 CLMN* <0.001 <0.001 <0.001 <0.001 <0.001 0.440 .38 3 .518 15 N <0.001 <0.001 <0.001 0.007 0.008 0.393 0.002 .429 pH <0.001 <0.001 <0.001 0.012 0.001 0.312 <0.001 <0.001

PAGE 136

136 Table 4 8 Results of multiple regressions, with dissolved inorganic N ( g N g soil 1) as the dependent variable and microbial bioma ss N (MBN), dissolved organic N 1 day 1 ) (N min), and soil C:N as independent variables, for each soil depth. All variables were natural log transformed before analyses. B SE B P R 2 F df p 0 5 cm Constant 1.864 0.776 0.296 5.66 4 0.001 MBN 0.329 0.118 0.423 0.007 DON 0.216 0.131 0.263 0.104 N min 0.266 0.630 0.060 0.674 C:N 0.827 0.276 0.420 0.004 5 10 cm Constant 0.825 0.521 0.303 5.86 4 0.001 MBN 0.353 0.100 0.460 0.001 DON 0.370 0.150 0.298 0.017 N min 2.213 0.815 0.331 0.009 C:N 0.661 0.184 0.474 0.001 10 15 cm Constant 0.230 0.444 0.294 5.61 4 0.001 MBN 0.334 0.105 0.468 0.002 DON 0.376 0.165 0.305 0.027 N min 1.776 0.849 0.247 0.041 C:N 0.448 0.170 0.392 0.011 15 20 cm Constant 1.057 0.409 0.272 5.05 4 0.002 MBN 0.339 0.126 0.415 0.009 DON 0.093 0.158 0.082 0.560 N min 2.770 1.042 0.315 0.010 C:N 0.015 0.160 0.014 0.925

PAGE 137

137 Table 4 9 1 day 1 ) as the dependent variable and microbial biomass N (MBN), dissolved organic N (DON), soil C:N, and soil %N as independent variables, for each soil depth. All variables were nat ural log transformed before analyses. B SE B p R 2 F df p 0 5 cm Constant 0.397 0.283 0.375 8.08 4 <0.001 MBN 0.015 0.033 0.085 0.652 DON 0.074 0.026 0.402 0.007 C:N 0.094 0.060 0.213 0.121 %N 0.049 0.033 0.311 0.143 5 10 cm Constant 0.012 0.179 0.138 2.16 4 0.086 MBN 0.042 0.020 0.365 0.045 DON 0.038 0.025 0.202 0.131 C:N 0.051 0.032 0.245 0.117 %N 0.015 0.022 0.134 0.497 10 15 cm Constant 0.070 0.189 0.065 0.94 4 0.445 MBN 0.008 0.01 9 0.083 0.664 DON 0.047 0.026 0.274 0.076 C:N 0.009 0.032 0.057 0.778 %N 0.006 0.024 0.053 0.814 15 20 cm Constant 0.114 0.174 0.051 0.72 4 0.580 MBN 0.004 0.020 0.048 0.824 DON 0.025 0.021 0.195 0.245 C:N 0.01 7 0.023 0.143 0.462 %N 0.016 0.025 0.153 0.536

PAGE 138

138 Figure 4 1 Mean ( + se) root biomass (g m 2 ) at each sampling depth for each time since fire.

PAGE 139

139 Figure 4 2 Relationship betwee n root biomass and soil percent N for each sampling depth.

PAGE 140

140 Figure 4 3 Mean ( + se) resin extractable NH 4 + NO 3 total inorganic N (A), PO 4 3 (B), and N:P ratios (C) over one year. Different letters represent significant differences among times since fi

PAGE 141

141 Figure 4 4 Relationship between pH of surface soils (0 5 cm) and resin exchangeable PO 4 3 Data is included from the 12 plots in which both variables were measured.

PAGE 142

142 Fig ure 4 5 Mean ( + se) soil %C (A), soil %N (B), soil C:N (C), K 2 SO 4 extractable dissolved inorganic N (DIN) (D), dissolved organic N (DON) (E), ratio of DIN 15 N (H), and soil pH (I) at each sampling depth for each time since fire.

PAGE 143

143 Figure 4 6 Mean (+ se) net N mineralization rates (A) and nitrification rates (B) with H 2 O addition at each sampling depth for each time since fire. Different letters 0.05. Differences among d epths within a time since fire are indicated by p < 0.1, ** p < 0.05, *** p < 0.01.

PAGE 144

144 Figure 4 7 Mean (+ or se) net N mineralization rates (left panels) and nitrification rates (right panels) at ambient field conditions and with H 2 O addition at each time since fire for each sampling depth. Differences between ambient and H 2 O addition for each time since fire at each depth are indicated by p < 0.1, ** p < 0.05, *** p < 0.01.

PAGE 145

145 Figure 4 8 Mean ( se) C pools (A) and N pools (B) with depth for each time since fire. Different letters indicate significant differences among times since fire in

PAGE 146

146 Figure 4 9 Relationship between microbial biomass N and dissolved inorganic N (DON) for each sampling depth.

PAGE 147

147 CHAPTER 5 DISTURBA NCE EFFECTS ON NUTRI ENT LIMITATION OF PL ANT PRODUCTIVITY IN SCRUBBY FLATWOODS : DOES FIRE SHIFT NITR OGEN VERSUS PHOSPHORUS LIMITATIO N? Introduction Fire is a common disturbance in many ecosystems and can have profound impacts on nutrient cycling and availab ility. Fire consumes plant biomass, litter, and soil organic matter, converting organic nutrients to inorganic forms (Certini 2005), which can be lost to the atmosphere or returned to the ecosystem in ash. Nitrogen (N) and phosphorus (P) are essential plan t nutrients that limit plant growth in most, if not all, terrestrial ecosystems (Vitousek and Howarth 1991). Yet fire has different effects on N and P due to fundamental differences in their biogeochemistry. Since N is more readily volatilized than P, rela tively more N than P is lost from an ecosystem during fire (Raison et al. 1985a). In highly weathered soils, most P is in organic matter. Fire rapidly mineralizes the P in these pools, often resulting in enhanced P availability after fire. Nitrogen, howeve r, may be relatively less available than P after fire due to greater combustion losses, and N inputs in unpolluted ecosystems are largely dependent upon biological N fixation which accumulates N over the inter fire cycle. During fire, nutrients can be lo st from an ecosystem to the atmosphere through volatilization (non particulate forms) or transport of ash (particulate forms). Numerous studies have measured the effect of fire on nutrient loss to the atmosphere by calculating the difference between the pr e fire nutrient content of fuel (i.e. understory plants and/or litter) and the post fire nutrient content of ash. For example, in a low intensity fire in a Mediterranean forest, 77% of N and 35% of P was lost (Gillon and Rapp 1989); during fires in Austral ian forests and woodlands, 92 94% of N and 41 53%

PAGE 148

148 of P was lost (Cook 1994); during slash and burn agriculture in Amazonia, 93 98% of N and 27 47% of P was lost (Mackensen et al. 1996); and fires in Brazilian savannas caused 95% of N and 51% of P to be los t (Pivello and Coutinho 1992). Regardless of the type of ecosystem or fire, approximately twice as much N as P is lost to the atmosphere during fire due to differences in volatilization temperatures and forms of nutrient loss. N volatilization occurs at te mperatures as low as 200C (White et al. 1973), whereas P is volatilized at temperatures above 774C (Raison et al. 1985a). The majority of N in combusted fuel is lost in non particulate forms, while P is lost in both non particulate and particulate forms. Thus, ash on the soil surface contains high concentrations of P and low concentratio ns of N (Debano and Conrad 1978; Raison et al. 1985b). The different fates of N and P in consumed fuel affect the total nutrient budget of an ecosystem. I nputs of N throu gh symbiotic N fixation and rainfall are not high enough, over the short term, to replace the amount of N volatilized in fire (Carter and Foster 2004; Cook 1994). Over time, P in ash becomes relatively less available as it is immobilized by plants and micr obes or fixed via geochemical reactions. Nitrogen availability, in contrast, tends to increase as inputs accumulate. Because fire has the potential to alter the relative availability of N versus P both immediately following fire and over inter fire cycles, a fundamental question about nutrient limitation is whether fire causes shifts in N versus P limitation. This is important because nutrient limitation of plant productivity is a fundamental control over the structure and function of ecosystems and has con sequences for biomass accumulation (Bret Harte et al. 2004), nutrient retention and loss (Hedin et al. 1995), biodi versity (Wassen et al. 2005), composition

PAGE 149

149 (Chapin et al. 1987), and the ecosystem values and services provided to humans (Daily et al. 2000). Plant growth is limited by any nutrient present in the soil below an optimum supply (Chapin et al. 2002). Thus, fire mediated differences in nutrient supply suggest that nutrient limitation may change with time since fire, with recently burned sites being N limited and long unburned sites being P limited, particularly in old, highly weathered soils. Research on the effects of fire on nutrient limitation, however, is scarce. Pines and oaks in a fire adapted Mediterranean forest were P limited in a site 5 ye ars post fi re (Sardans et al. 2004). O ne of four species was N limited in a lodgepole pine forest 3 5 years after fire (Romme et al. 2009). Nitrogen limits productivity of trees in secondary Amazonian forests, possibly due to the residual effects of fire o n N availability (Davidson et al. 2004). I n lakes within unburned catchments phytoplankton biomass was limited by P or co limited by N and P, while in lakes within burned catchments phytoplankton biomass was limited by N, likely due to increased loading of P relative to N post fire (McEachern et al. 2002). F ire dependent F lorida scrub ecosystems occur on old, highly weathered oligotrophic soils (Myers 1990). P revious research in this system suggests that the nutrient most limiting to plant production ma y change with time since fire. In palmetto flatwoods, soil and foliar N:P ratios decreased shortly after fire (Chapter 1). In scrubby flatwoods, phosphate, but not total inorganic N, varied with time since fire, causing soil N:P ratios to be highest at int ermediate times since fire (6 8 years). Although inorganic N:P ratios were low in longer unburned sites (13 years), bulk soil %N was high, suggesting that N accumulates in soil over time (Chapter 3). Thus, I tested the

PAGE 150

150 hypotheses that: (1) plant productivi ty in recently burned scrubby flatwoods is N limited; (2) plant productivity in intermediately burned scrubby flatwoods is P limited; and (3) plant productivity in long unburned scrubby flatwoods is co limited by N and P. Methods Study S ite This study was conducted at Archbold Biological Station (ABS) in Highlands County, Florida, USA (2710'50"N, 8121'0" W), which is near the southern tip of the Lake Wales Ridge. Archbold Biological Station typically has warm wet summers and cool dry winters (Abrahamson et al. 1984). Mean annual precipitation is 136.5 cm (ABS weather records, 1932 2004), and mean annual temperature is 22.3C (ABS weather records, 1952 2004). Archbold Biological Station is divided into burn units, which have been managed with prescribed fi res for over 35 years. My research focused on scrubby flatwoods, a distinctive plant community of Florida scrub. Scrubby flatwoods are dominated by shrubby oaks (Fagaceae), palmettos (Arecaceae), and ericaceous shrubs (Ericaceae). Scrubby flatwoods exper ience fire return intervals of 8 16 years (Menges 2007), and the dominant vegetation resprouts after fire (Menges and Kohfeldt 1995). Soils are entisols (Abrahamson et al. 1984) that have no horizon development, little organic matter, and low exchange capa city and base saturation (Brown et al. 1990). Experimental Design In July 2007, I located burn units 6 weeks after fire (recently burned), 8 years after fire (i ntermediate), and 20 years after fire (long unburned). For each time since fire, I established t hree blocks in scrubby flatwoods communities. Within each block, I established ten 3 x 3 m plots, with a buffer of 2 to 4.5 m between each plot, for a total of

PAGE 151

151 30 plots for each time since fire. Plots were established where there were at least three indivi duals of scrub oak ( Quercus inopina Ashe) and saw palmetto ( Serenoa repens (W. Bartram) Small (Arecaceae)) because these are two of the most dominant scrubby flatwoods species, with 37 and 22 percent cover, respectively (Abrahamson et al. 1984). Each plot was randomly assigned one of the following treatments (3 replicates of each per time since fire): control, low N addition (2 g N m 2 ), intermediate N addition (5 g N m 2 ), high N addition (10 g N m 2 ), low P addition (1 g P m 2 ), intermediate P ad dition (2.5 g P m 2 ), high P addition (5 g P m 2 ), low N + P addition (2 g N + 1 g P m 2 ), intermediate N + P addition (5 g N + 2.5 g P m 2 ), or high N + P addition (10 g N + 5 g P m 2 ). According to the National Atmospheric Deposition Program, d eposition of N in the area of ABS is 5 10 kg ha 1 Thus, my low N treatment doubled background N deposition. I added P fertilizer at half the rate of N, as is commonly done in nutrient fertilization ex periments (Vitousek et al. 1993; Davidson et al. 2004 ), because plant demand for N is greater than plant demand for P. While most fertilization studies apply one level of f ertilizer (Davidson et al. 2004; Vitousek and Farrington 1997), I applied different levels of fertilizer as this is likely more realistic in relation to different degrees of habitat alteration. Nitrogen was added as ammonium urea and ammonium nitrate (half of each) and P was added as Triple Superphosphate. The annual dose of fertilizer was divided into quarters and added several times over one year because sandy soils in Florida scrub have low sorption capacity (N. Comerford, personal communication), and because a large rain event could cause high leaching of added nutrients. During the first year of the experiment, fertilizer was added in July 2007 (middle of the wet season), October 2007 (beginning of the early dry

PAGE 152

152 season), January 2008 (end of the early dry season), and April 2008 (end of the late dry season) (Figure 1). In September 2008, the 20 year since fire sites were burned in a pre scribed fire. Thus, during the second year of the experiment, only the recentl y burned sites (now 1 year after fire) and the sites at intermediate ti mes since fire (now 9 year after fire), were used. Fertilizer was added in October 2008 (half of the yearly amount), January 2009 (one quarter of the yearly amount), and April 2009 (one quarter of the yearly amount). Hereafter, sites will be referred to by their time after fire at the beginning of the experiment in July 2007 (e.g. 8 years after fire or 8 yr sit es). Soil N utrients To monitor background nutrient availability in each site, I buried four anion and four cation exchange resin bags in the soil near each control plot before the first nutrient addition in July 2007. These resin bags were removed before the nutrient addition in October 2007 and new bags were buried. I continued this for subsequent fertilization events (i.e. bags removed and buried in January 2008; bags removed and buried in April 2008; bags removed in July 2008), thus measuring backgroun d nutrient availability in 3 month intervals over the first year of the experiment. In August 2008, I collected soil from three random locations in all plots. Each soil core (2.5 cm diameter) was separated by depths (0 10 cm and 10 20 cm), and the three co res from each depth were bulked. Within 24 hours of collection, soil samples were passed through a 2 mm sieve and sub sampled for determination of gravimetric soil moisture, pH, total percentages of N and C, inorganic N concentrations, N mineralization rat es, and inorganic P concentrations. Gravimetric moisture content was determined on samples dried at 105C for 48 hrs. For soil pH, 10 g of air dried soil was added to 10 mL of deionized water, shaken for 30 sec, allowed to stand for 10 min

PAGE 153

153 (Thomas 1996), a nd then pH was determined with an electronic pH meter (Thermo Orion 250A+, Orion Research, Inc., Boston, Massachusetts, USA). A subsample of soil was dried at 60C for 48 hrs and ground on a spex mill (8000D dual mixer/mill, Spex Certiprep Inc., Metuchen, New Jersey). One sub sample of ground surface (0 10 cm) soils was analyzed for percentages of N and C at the University of Florida on an elemental analyzer (ECS 4010, Costech Analytical, Valencia, California, USA). A sub sample of ground soils from each co ntrol plot for each time since fire (both 0 10 cm and 10 20 cm depths) was sent to the ALS Laboratory Group (www.alsglobal.com) for analysis of macro To measure inorganic N concentratio ns, 50 mL of 0.5 M K 2 SO 4 was added to 10 g of field moist soil, shaken for 30 seconds, and allowed to stand overnight. Solutions were filtered through Whatman #42 filter paper that was pre leached with 0.5 M K 2 SO 4 Filtered extracts were frozen and taken t o the University of Florida. Dissolved inorganic N (NH 4 + + NO 3 ) concentrations were determined colorimetrically on a segmented flow autoanalyzer (Astoria Pacific, Inc., Clackamas, Oregon, USA). To measure inorganic P concentrations, 30 mL of 0.05 M hydroc hloric acid (HCl) and 0.0125 M hydrogen sulfate (H 2 SO 4 ) was added to 15 g of field moist soil, shaken for 5 min, then filtered through Whatman #42 filter paper. I stored filtered samples in a refrigerator for up to three weeks before analysis for phosphate (PO 4 3 ) concentrations Tek Instruments, Inc., Winooski, Vermont, USA) using the malachite green method Ecology Research Center (Highlands, Co Florida).

PAGE 154

154 I calculated the mean resin exchangeable nutrients per plot and used a one way ANOVA to analyze differences with time since fire. I analyzed differences in ion) with time since fire separately for surface (0 10 cm) soils and deep (10 20 cm) soils. One way ANOVAs were used when data were normally distributed and Kruskall Wallis tests were used when data were not normally distributed. In addition, I used Kruska ll Wallis tests to analyze differences in total P and calcium (Ca) with depth for each time since fire. The effects of nutrient addition on soil extractable N and P, N:P ratios, soil %N, %C, C:N ratios, and soil pH were determined separately for each time after fire and depth. All data were analyzed using a general linear model with treatment as a fixed effect and block as a random effect. I used post differences were greater than the control. Data were natural log transformed when necessary to meet the assumptions of normality (SPSS 11.5). Aboveground Biomass and G rowth In each plot, before the first fertilization event, I marked three individuals of Q. inopina and S. repens and up to three individuals of Lyo nia lucida (Lam.) K. Koch, Lyonia fruticosa (Michx.) G. S. Torr, Quercus chapmanii Sarg., Quercus geminata Small, and Sabal etonia Swingle ex Nash (nomenclature follows Wunderlin and Hansen 2003) These species account for over 97% of the shrub cover in my experimental plots (Table 5 1). The oaks ( Quercus species) and ericaceous shrubs ( Lyonia species) are clonal, multi stemmed species, and the number of stems per individual (defined by a circle of 10 cm in diameter centered on the tallest stem) at the beg inning of the experiment, ranged from 1 to 22 in recently burned sites, from 1 to 11 in sites at intermediate times since fire, and from 1 to 9 in long unburned sites. For the shrubs L.

PAGE 155

155 lucida L. fruticosa Q. chapmanii Q. geminata and Q. inopina I mea sured crown length and width, and measured the diameter and height of each stem. In addition, I measured apical shoot growth on a randomly selected subset of stems for Q. inopina each individual. If a stem had many new apical shoot growth increments, I mea sured a subset of them and scaled up to the entire stem. For the palmettos S. etonia and S. repens I measured maximum crown length, minimum crown length, and height, and counted leaves. In July 2008, after one year of fertilization, I made the same measur ements listed above and estimated percent cover of all shrub species. In July 2009, after two years of fertilization, I measured only the most dominant species, Q. inopina and S. repens In addition, I estimated percent cover of all shrub species pre ferti lization and one year after fertilization. Differences in percent cover of shrubs and contribution to total shrub cover with time since fire were determined with Kruskal Wallis tests. I used age specific allometric equations (Chap ter 2) for different tim es after fire (6 weeks, 1 2 years, 8 10 years, 20 22 years) to estimate total stem biomass and leaf biomass of each stem of each shrub species and total leaf biomass of each palmetto individual. For several shrub stems in sites 6 weeks after fire, leaf bio mass estimates were greater than total stem biomass estimates, so leaf biomass was used for total stem biomass I used several methods to investigate the effects of nutrient addition on aboveground shrub biomass. I assessed differences in total shrub bioma ss among treatments at each time after fire. I used measurements of shrub length (maximum crown length) and width (minimum crown length) to calculate the area of each marked individual. For oaks and lyonias, I summed stem biomasses to determine the biomass

PAGE 156

15 6 of each individual. I then calculated the total measured biomass and total area for each species in each plot. If the area covered by a species was greater than what I measured, I used estimates of percent cover of each speci es to scale up biomass of each species to the entire plot. Total plot biomass was analyzed using a general linear model with treatment as a fixed effect and block as a random effect. I assessed differences in total plant biomass among treatments for Quercus inopina and Serenoa repens because these species occurred in all plots. For Q. inopina I summed stem biomass to determine the biomass of each individual. For both species, I calculated the percent change in total biomass for each species separately for the first and second years of fertilization. Because stem turnover occurred during the course of the experiment, I assessed differences in basal diameter and height for Q. inopina stems that were alive for multiple years. I calculated the mean percent change in basal diameter and heig ht of stems per individual. For sites six weeks and eight years since fire, I analyzed differences in stems that were alive over the first year of fertilization, over the second year of fertilization, and over both years of fertilization. For apical shoot growth, I calculated mean growth per individual (based on a subset of stems if individuals had more than 3 stems) To analyze the effects of treatment on percent change in total plant biomass, basal diameter, height, and apical shoot growth, I used general linear univariate models with treatment as a fixed effect and block as a random effect. I used post treatment plots was greater than in the control plots. Extreme outliers (> 3 standard deviations from the mean) were removed if the percent change for a measure was negative. I analyzed differences in percent change in biomass because the number of

PAGE 157

157 new leaves produced per year by S. repens and S. etonia is positively correlated with palmetto biomass ( Abrahamson 2007), suggesting that pre treatment biomass may affect the biomass response to added nutrients. For sites six weeks and eight years after fire (the sites that were fertilized for two years), I used paired t tests to compare differences in perce nt change in total biomass, basal diameter, and height between years. In addition, I assessed differences in stem percent survival, number of new stems, and percent change in stem number, which accounts for both stem death and stem recruitment, over the fi rst year of fertilization for each Q. inopina individual. I analyzed the effects of treatment separately for each time since fire using Kruskal Wallis tests after confirming with Kruskal Wallis tests that there were no block effects (SPSS 11.5). Litterfa ll In each plot pre fertilization (July 2007), I measured litter depth at 12 locations. I measured litterfall over one year by placing 4 litter traps in a cross formation on the ground of each 3 x 3 m plot. Litter traps were 0.09 m 2 10 cm deep, and made o f 2 mm hardware cloth. Litter was collected in December 2007 (the first time at which there was litter to collect), March 2008, May 2008, and July 2008. Litter was pooled at the plot level, sorted to fractions (leaf, twig, and reproductive litter), dried a t 60C for 48 hours, and weighed. Leaf and reproductive litter collected in May and July was sorted to species before being dried and weighed. Quercus inopina litter samples from May 2008 were ground on a wiley mill and analyzed for percentages of N at the University of Florida on an elemental analyzer (ECS 4010, Costech Analytical, Valencia, California, USA).

PAGE 158

158 I calculated total leaf litterfall (g m 2 year 1 ) for each plot and used a separate general linear model for each time after fire with treatment as a fixed effect, block as a random effect, and mean (of pre and one year post fertilization) total percent shrub cover as the covariate to analyze the effects of nutrient addition on leaf litterfall. Because oak leaf turnover occurs in the spring (Abram s and Menges 1992), and Quercus inopina was the dominant oak in my plots, I analyzed treatment effects on Q. inopina litterfall from March May 2008. I used a separate general linear model for each time after fire with treatment as a fixed effect, block as a random effect, and post fertilization Q. inopina percent cover (measured in July 2008) as a covariate to analyze the effects of nutrient addition on Q. inopina leaf litterfall and Q. inopina leaf litterfall N (g N m 2 2 months 1 ). I analyzed differe nces in %N of Q. inopina litter collected in May 2008 using a general linear model with treatment as a fixed effect and block as a random effect. I used post treatment plots was greater than i n the control plot. Data was natural log transformed when necessary to meet the assumptions of normality (SPSS 11.5). Foli ar N utrients I collected foliar samples of all marked Quercus inopina and Serenoa repens individuals pre fertilization. In July 2008, after one year of fertilization, and July 2009, after two years of fertilization, I collected foliar samples of all Q. inopina and S. repens individuals that were still alive. All samples from pre fertilization and one year after fertilization, but only s amples from control and high nutrient addition plots from two years after fertilization, were dried, ground, and analyzed for percentages of N following the same methods as for litter samples. I measured foliar phosphorus of pre fertilization and one year after fertilization samples of Q. inopina and S. repens from the control,

PAGE 159

159 high N, high P, and high N + P plots. Subsamples of 0.05 to 0.5 grams were weighed into crucibles, ashed in a muffle furnace at 500C for 5 hours, extracted with 6 M HCl, then brough t to volume so that the solution was 0.6 M HCl. Extracts were stored in the refrigerator for several days then analyzed colorimetrically on a spectrophotometer microplate reader (PowerWave XS Microplate Reader, Bio Tek Instruments, Inc., Winooski, Vermont, USA) at the University of Florida using the ascorbic acid molybdenum blue method (Murphy and Riley 1962). Standard NIST peach leaves were used to determine the efficiency of the digestion. To analyze differences in all measures of pre fertilization foliar nutrients (%N, %P, and N:P ratios) I fit a model with species and time since fire and their interaction as main effects. Differences among species and times since fire were determined with post hoc Tukey HSD tests (JMP 8.0). I analyzed the effect of treat ment on the percent change in foliar %N, %P, and N:P ratios one year after fertilization using general linear models with treatment as a fixed effect and block as a random effect. I used post hoc ent plots was greater (%N and %P) or less (N:P) than in the control plot (SPSS 11.5). Root Productivity I used root ingrowth cores (Cuevas and Medina 1988) to measure root productivity during the first year of fertilization. In each plot pre fertilization (July 2007), I randomly established three cylindrical, closed bottom root ingrowth cores (2 mm mesh, 20 cm deep, 8 cm diameter). I collected soil cores, sieved the soil through 2mm and 1mm sieves to remove roots (< 2 mm diameter) and underground stems (> 2 mm diameter), placed the ingrowth cores in the ground, and then filled the cores with the root free soil. I weighed the sieved roots after drying at 65C for 48 hours. Root ingrowth

PAGE 160

160 cores were removed from the ground in August 2008, and each core was imme diately put into a plastic bag. I hand picked roots from the inside of each core and sieved all sand from inside each core through 2mm and 1mm sieves. Long thin roots that passed through the sieve were picked out by hand. Roots were dried at 60C for 48 ho urs then weighed. I used a general linear univarite model with treatment as a fixed effect, block as a random effect, and pre fertilization root biomass (>2mm + 2mm fraction; g m 2 ) as the covariate to analyze the effects of nutrient addition on root pro ductivity (> 2 mm + 2 mm fraction; g m 2 year 1 ). I used post productivity in treatment plots was greater than in the control plot (SPSS 11.5). Results Soil N utrients Over one year, resin exchangeable NH 4 + (F 2,6 = 4.62, p = 0.061) and NO 3 (F 2,6 = 4.22, p = 0.072) tended to be highest during the first year after fire (Figure 5 2). Total resin exchangeable N was 3.2 and 2.6 times greater during th e first year after fire than in 8 yr and 20 yr sites respec tively (F 2,6 = 9.60, p = 0.013). Resin exchangeable PO 4 3 was 5.1 and 3.8 times greater during the first year after fire than in 8 yr and 20 yr sites, respectively (F 2,6 = 11.86, p = 0.008). There was no difference, however, in resin exchang eable N:P ratio s with time after fire (F 2,6 = 0.34, p = 0.726). In surface soils (0 10 cm) of control plots P was highest in recently burned sites, while zinc (Zn) was lowest in 8 yr sites Soil P and Zn did not differ with time after fire in deep soils (Table 5 3). In recently burned sites 2 = 5.00, df = 1, p = 0.025) and 2 = 3.33, df = 1, p = 0.068) were higher in surface soils than in deep soils. In 8 yr and 20 yr sites 2 2 2 = 2 = 0.67, p = 0.414, respectively) did not vary with soil depth. Other

PAGE 161

161 elements were below detection limits (Table 5 2) or did not vary with time after fire (Table 5 3) i n control plots After one year of fertilization, N addition had a minimal impact on soil N (Table 5 4), only increasing inorganic N availability in deep soils (10 20 cm) in 20 yr sites (Figure 5 3). Eight years after fire, the significant effect of treatment occurred because high P addition caused a decrease in inorganic N availability (Figure 5 3). A fter one year of fertilization, P addition had a minimal effect on soil extractable P in recently burned sites (Table 5 4). Eight years after fire, all levels of P addition increased soil extractable P, while 21 years after fire, only high and intermediate P addition increased soil extractable P above levels in control plots (Figure 5 4). In 8 yr and 20 yr sites, high P addition increased extractable P 10 fold in surface and deep soils. Soil extractable P in high P and high N+P plots was similar in recently burned and long unburned sites. Due to the effects of P addition on soil extractable P, soil N:P ratios were higher in control and N addition plots than in P and N+P addition plots eight and twenty years after fire. There was no effect of nutrient additio n on soil %N, soil %C, C:N ratios, or soil pH at any time after fire (Table 5 4). Aboveground C over, Biomass, and G rowth Pre fertilization, percent cover of Quercus inopina and Serenoa repens as well as total p ercent cover of dominant shrubs, was lowest in 6 wk sites (Table 5 1). In recently burned sites, aboveground biomass of dominant shrubs was 71 g m 2 + 5 g m 2 (mean + se), and dominant shrubs covered, on average, 23% (2.1 m 2 ) of the plot area (9 m 2 ) (Table 5 1). Eight years after fire, abovegro und biomass of dominant shrubs was 334 g m 2 + 20 g m 2 (mean + se), and dominant shrubs covered, on average, 58% (5.2 m 2 ) of the plot area. Twenty years after fire, aboveground biomass of

PAGE 162

162 dominant shrubs was 377 g m 2 + 25 g m 2 (mean + se), and d ominant shrubs covered, on average, 59% (5.3 m 2 ) of the plot area. At all times after fire, nutrient addition did not have a significant effect on the percent change in total shrub biomass (Table 5 5). In recently burned sites, mean percent change in tota l shrub biomass was highest in high N plots (542%) and lowest in high P plots (273%). Mean percent change in total shrub biomass across treatments was negative in two of the three blocks at intermediate times after fire and was negative in one block in lon g unburned sites Over the first year of fertilization in recently burned sites, mean percent change in S. repens biomass was 662% in high N plots, which was significantly higher than the control (Figure 5 5A). P ercent change in biomass of Serenoa repens was not affe cted by nutrient addition over the first year o f fert ilization at intermediate times after fire (Table 5 5); however, i n long unburned sites, the mean increase in percent change in biomass was approximately 12 times greater in high N addition p lots than in control plots (Figure 5 5E). Percent change in S. repens b iomass was greater in the first than in the second year of fertilization in both 6 wk (t = 21.09, df = 29, p < 0.001) and 8 yr (t = 3.67, df = 29, p = 0.001) sites, and was not affected by nutrient addition o ver the second year of fertili zation. Over the first year of fertilization i n recently burned sit es, mean percent change in biomass of Quercus inopina was 645% in high P addition plots, which was significantly greater than the contro l (Figure 5 6A). In 8 yr and 20 yr sites, however, nutrient addition had no effect on percent change in Q. inopina biomass; in 8 yr sites, mean percent change in biomass was higher in control plots than in treatment plots (Figure 5 6C).

PAGE 163

163 Percent change in Q inopina biomass was greater in the first than in the second year of fertilization in 6 wk sites (t = 9.8 4, df = 29, p < 0.001), but not 8 yr sites (t = 0.35, df = 29, p = 0.726), and was not affected by nutrient addition o ver the second year of fertili za tion. Over the first year of fertilization in recently burned sites, mean percent change in height of Q. inopina stems was significantly higher in high P plots (195%) compared to control plots (90%) (Figure 5 7A), while intermediate N + P addition signif icantly increased mean percent change in basal diameter (Figure 5 8A) and apical shoot growth (Figure 5 9A) of Q. inopina Over the first year of fertilization in 8 yr and 20 yr sites, nutrient addition did not affect mean percent change in height or basal diameter of Q. inopina stems (Table 5 5). Apical shoot growth of Q. inopina however, was significantly higher in low N + P addition, intermediate N addition, and high P addition plots than in control plots in 8 yr sites, and significantly higher in high P addition plots than in control plots in 20 yr sites (Figure 5 9). For stems alive all three years of the experiment (in 6 wk and 8 yr sites ), percent change in basal diameter ( 6 wk sites: t = 14.19, df = 29, p < 0.001 ; 8 yr sites: t = 14.19, df = 29, p < 0.001 ) and height ( 6 wk sites: t = 9.53, df = 29, p < 0.001 ; 8 yr sites: t = 2.45, df = 29, p = 0.020) were greater in the first year than in the second year of fertilization In recently burned and long unburned sites, nutrient addition had no effect on percent survival of stems, the number of new stems, or percent change in stem number of Quercus inopina after one year of fertilization (Table 5 6). At intermediate times after fire percent survival of stems was lower in control plots than in low N plo ts but nutrient addition had no effect on percent change in stem number (Table 5 6).

PAGE 164

164 Litterfall Mean + se total leaf litterfall (g m 2 year 1 ) was 75.4 + 5.3, 164.3 + 10.4, and 147.9 + 10.9 in 6 wk, 8 yr, and 20 yr sites respectively. At all times af t er fire, total leaf litterfall was not affected by nutrient addition, but in 20 yr sites percent cover of shrubs was a significant predictor of total litterfall (Table 5 7). Similarly, Q. inopina spring leaf litterfall was not affected by nutrient additio n (Figure 5 10), but post fertilization percent cover of Q. inopina was a significant predictor of Q. inopina litterfall in 6 wk and 20 yr sites (Table 5 7). In 8 yr sites Q. inopina litterfall %N was higher in high N + P treatments than in control plots (Table 5 8); however, Q. inopina litterfall N was not affected by nutrient addition at any time after fire (Table 5 7). Mean Q. inopina litterfall N (g N m 2 2 mos 1 ) was 0.12, 0.23, and 0.23 in 6 wk, 8 yr, and 20 yr sites, respectively. Foliar N utrie nts Pre fertilization, foliar %N and %P were consistently higher in Q. inopina than in S. repens (Table 5 9). The largest difference occurred in 6 wk sites, where foliar %N and %P were 1.27 and 1.53 times higher in Q. inopina than in S. repens respective ly (Figure 5 11). Proportionally, the decrease in foliar %P of Q. inopina and S. repens from 6 wk to 8 yr sites was greater than the decrease in foliar %N. Foliar N:P ratios increased 8 and 6 units for Q. inopina and S. repens respectively from 6 weeks to 20 years after fire (Figure 5 11). Foliar %N, %P and N:P ratios were similar in 8 yr and 20 yr sites In recently burned sites, foliar %N and %P decreased over the first year of fertilization. Nutrient addition had a significant effect on percent change in foliar %N of Q. inopina (Figure 5 12), and high P and N + P addition caused less of a decrease in foliar %P of Q. inopina and S. repens (Figure 5 13) Eight years after fire, nutrient addition had no effect on percent change in foliar %N of Q. inopina o r S. repens but the

PAGE 165

165 increase in foliar %P of Q. inopina was seven times greater in high P plots than in control plots (Figure 5 13A). This caused the percent change in foliar N:P ratios of Q. inopina to be positive in control plots and negative in high P addition and high N + P addition plots (Figure 5 13C). In long unburned sites, percent change in foliar %N of S. repens was approximately three times higher in high N and N + P plots than in control plots (Figure 5 12E), while high P and N + P addition sig nificantly increased percent change in foliar %P of Q. inopina (Figure 5 13A). Root P roductivity In control plots, mean + se root productivity (g m 2 year 1 ) was 193.7 + 50.7, 128.8 + 10.7, and 145.4 + 41.1 in 6 wk, 8 yr, and 20 yr sites respectivel y. In recently burned sites, nutrient addition had a slightly significant effect on root productivity, and pre fertilization root biomass was a significant predictor of root productivity (Table 5 7). Eight years after fire, root productivity was 3.6 times higher in high N+P addition plots than in control plots, and pre fertilization root biomass was a significant predictor of root productivity (Figure 5 14B). Twenty years after fire, root productivity was 2.4 times higher in high N + P plots than in control plots (Figure 5 14C). Discussion Soil N utrients Over the first year of the experiment, background nutrient availability (resin exchangeable N and P) was highest in recently burned sites, likely due to the short term effects of fire on nutrient availabi lity. High N availability in burned sites is likely due to microbial or ash derived N. Increased soil temperatures associated with fire (Ewel et al. 1981) kill soil microbes, indicated by a decrease in microbial C and N after fire (Prieto Fernndez et al. 1998), which causes the release of N from ruptured mi crobial cells

PAGE 166

166 (Dunn et al. 1985; Serrasolsas and Khanna 1995). In addition, ash can contain high concen trations of N (Ewel et al. 1981; Kauffman et al. 1993), which can cause an increase in soil N after fire. High concentrations of NH 4 + and NO 3 in burned sites may also be related to high N mineralization and nitrific ation rates (DeLuca et al. 2002; DeLuca and Sala 2006); however, previous research in Florida scrub ecosystems has found that N immobilizati on, rather than N mineralization, may occur within the first year after fire (Chapters 1 and 3). In other scrubby flatwoods sites, r esin exchangeable N was slightly lower over the second year after compared to eighth year after fire (Chapter 3), which sugg ests that the short term effects of fire on N persist for less than one year after fire (Chapter 1). High PO 4 3 availability in recently burned sites is likely due to high concentrations of P in ash post fire (Wi lbur and Christensen 1983; Raison et al. 19 85b), since mineral soil PO 4 3 is correlated with ash depth (Rice 1993). In other scrubby flatwoods sites, P availability was higher 13 years after fire than one year after fire (Chapter 3); however, this could be related to the presence of pine trees, whi ch did not occur in plots in this study, or a greater amount of organic matter than in the sites used in this study (1.47 versus 0.66 mean soil %C). Soil N:P ratios were only slightly, and not significantly lower, in recently burned sites than other times after fire. Soil macronutrients and micronutrients in deep soils (10 20 cm) of control plots did not vary with time after fire (Table 5 3). Thus, differences in soil nutrients in surface soils are likely related to fire, and variation in plant growth and b iomass across times after fire should not be due to differences in parent material.

PAGE 167

167 Nitrogen addition had a greater effect on soil extractable N 20 years after fire compared to other times after fire (Figure 5 4). Litter depth was highest 20 years after fire suggesting that some of the added fertilizer is retained in the system in organic matter. The lack of an effect of N addition on soil N six weeks and eight years after fire could be due to uptake of added N by plants and/or microbes. In coastal Florid a scrub oak ecosystems, soil microbial communities were primarily N limited with a secondary P limitation (Brown et al. 2009), and in dry tropical ecosystems, N, but not P, addition caused an increase in microbial biomass C, suggesting that N is more limit ing than P (Tripathi et al. 2008). At the global scale, microbial N:P ratios average 7:1 (Cleveland and Liptzin 2007), so microbial demand of N is greater than demand of P. Other studies from a variety of ecosystems have found that N addition increases soi l NO 3 and NH 4 + (Bret Harte et al. 2004; Caffrey et al. 2007; Hungate et al. 2007; Iversen and Norby 2008). Scrubby flatwoods differ from other sites by having sandy soils with low exchange capacity, suggesting that leaching may be important in these sites if plants and microbes do not use the added N quickly. Unlike N, P addition led to a significant increase in soil extractable P after one year of fertilization in sites nine and 21 years since fire, but not in recently burned sites (Figure 5 3). Nine ye ars after fire, high P addition (5 g P m 2 ) did not increase soil extractable P more than low P addition (1 g P m 2 ); whereas, 21 years after fire, only high P addition significantly increased soil extractable P. These results support three primary con clusions about P addition. First, the effect of added P was greater in sites with lower background P availability. Based on background soil P concentrations measured in surface soils (0 10 cm) of control plots one year after fertilization, low P

PAGE 168

168 addition i n sites eight years since fire had the potential to increase soil P content by the same factor as high P addition in recently burned sites (~60 fold). High P addition in sites eight and 20 years since fire could have increased soil P content approximately 160 and 310 fold, respectively. Second, some added P was retained in the soil, either in organic matter or bound to soil particles. In treatments with high P addition, 6 7% of the added P was retained in the top 20 cm of soil. The other 93 94% of added P w as either leached or taken up by plants and microbes. Third, plants and microbes did not use all of the available P because it was leached or bound to soil particles. Alternatively, N may have become limiting, which is possible considering that as P additi on increased eight years after fire, N availability decreased (Figure 5 3). Litaor et al. (2008) found that N addition caused a decrease in soil pH, and phosphorus availability is negatively correlated with soil pH (Chapter 3; Jaggi et al. 2005); however, N addition did not affect soil pH in my study. Similar to my study, P fertilization increased resin exchangeable PO 4 3 in tundra (Bret Harte et al. 2004) and increased soil P in alder wetlands (Gkkaya et al. 2006). Fertilization increased nutrient avail ability in scrubby flatwoods ecosystems. The independence of soil inorganic N availability six weeks and eight years after fire to N addition is more likely related to the offset in times of nutrient addition and soil collection than a lack of an effect of N addition. Soils were collected four months after the last fertilizer addition, which was only one fourth of the total amount of nutrients added, by which time added N was taken up by plants or microbes, or leached. I likely measured an increase in soil P because fertilizer P may become rapidly immobilized or sorbed (Sarmiento et al. 2006).

PAGE 169

169 Aboveground Biomass and G rowth Nutrient addition did affect biomass and growth of the dominant shrubs in several cases. High N addition increased biomass of Serenoa r epens individuals six weeks and 20 years after fire (Figure 5 6), suggesting that palmettos are N limited in recently burned sites and long unburned sites. Over the first year of fertilization in sites six weeks after fire, high P addition increased biomas s and height growth of Quercus inopina intermediate N + P addition increased basal diameter change and total apical shoot growth, suggesting that Q. inopina is co limited by N and P, with a stronger P limitation, in recently burned sites. Eight years afte r fire, low N + P addition, intermediate N addition, and high N addition increased apical shoot growth, while 20 years after fire, high P addition increased apical shoot growth (Figure 5 9). These results suggest that different scrubby flatwoods species a re limited by different nutrients. Similar results have been documented in other systems (Romme et al. 2009) Foliar N:P ratios suggest that species from the same habitat can be limited by different nutrients, possibly because of differences in mycorrhizal status (Diehl et al. 2008), but nutrient addition can have a positive (Nijjer et al. 2010) or negative (Sims et al. 2007) effect on mycorrhizal fungi. Oaks have associations with ectomycorrhizae (Langely et al. 2002) and S. repens has associations with ar buscular mycorrhizae (Fisher and Jayachandran 1999). These differen ces in mycorrhizal status may contribute to limitation by different nutrients. Nutrient addition had no effect on total aboveground shrub biomass (Table 5 5). Eight years after fire, S. re pens biomass was not affected by nutrient addition. Eight and 20 years after fire, biomass, basal diameter, and height growth of Q. inopina were not influenced by nutrient addition. The limited response of shrubs to nutrient addition on

PAGE 170

170 shrub biomass and g rowth eight and 20 years after fire was surprising given that scrubby flatwoods have low nutrient availability and that my focal species have responded to nutrient addition in previous studies in Florida scrub. Abrahamson (1999) found that NPK fertilizatio n increased biomass of Serenoa repens and Sabal etonia and Berry and Menges (1997) found that P fertilization increased growth of Quercus inopina stems, while N fertilization increased S. repens leaf production. These studies, however, added fertilizer ar ound an individual rather than over the area of a plot. Johnson and Abrahamson (2002) found that 80 to 90% of Q. inopina stems died over a 10 year period, but new stems were recruited so that the number of stems of Q. inopina individuals remained the same or increased over time. The number of stems per individual decreased only i n sites one year after fire; however, individuals in recently burned sites did not decrease in biomass, and percent change in stem number of Quercus inopina individuals was not aff ected by nutrient addition at any time after fire. Thus, the limited effect of nutrient addition on biomass and growth is likely not due to a decrease in stem number of shrubs. Rather, numerous other factors could have influenced the nutrient addition on b iomass and growth. First, I measured a random sub sample of shrubs in each plot, which is less precise than harvesting and weighing all plant biomass. In addition, I did not measure biomass of sub shrubs or herbaceous species. In other ecosystems, however, graminoids had a larger response to fertilizer addition than deciduous and evergreen shrubs (Bret Harte et al. 2004); nutrient addition led to an increase in the abundance of graminoids and forbs, while deciduous and evergreen shrubs declined in abundance (Jgerbrand et al. 2009). Nutrient addition may have had a greater impact on the

PAGE 171

171 biomass of sub shrubs, graminoids, and forbs because these species may have most of their roots in surface soils giving them greater access to added nutrients. I would not ha ve detected an effect of nutrient addition on biomass if one of the less common species showed the greatest response to added nutrients. Furthermore, P addition increased the annual diameter increment of Acacia koa trees increased only when trees were thin ned and grasses were controlled with herbicide (Scowcroft et al. 2007), suggesting that competition for nutrients may have limited growth of shrubs and the effects of nutrient addition on shrub biomass. Second, the buffer between plots may not have been l arge enough to ensure that oaks and ericaceous shrubs were not connected underground. Clones of the Australian shrub Zieria baeuerlenii can cover distances of at least 5 m (Sharma 2001), clones of Rhododendron ferrugineum range from < 1.5 to 27 m 2 (Pornon and Escaravage 1999), and clones of Quercus geminata a common species in scrubby flatwoods, are often at least 9 m 2 and can cover distances of up to 18 m (Ainsworth et al. 2003). If clones of oaks or ericaceous shrubs extended across plots and stems were connected underground, individuals that were in the control plots could have had access to nutrien ts in the fertilized plots. Such a growth pattern could have diluted the effect on nutrient addition on biomass and growth. Third, if nutrient addition le d to an increase in new growth or foliar nutrients higher herbivory may have resulted Gall wasp occurrence (Abrahamson et al. 2003) and abundance of leaf miners ( Cornelissen and Stiling 2008) are correlat ed with oak leaf chemistry. Cornelissen and Stiling ( 2006) found that fertilized Q. geminata had higher foliar %N, lower tannins, and higher densities of numerous herbivore guilds. I did

PAGE 172

172 not quantify herbivory, but I did not observe greater herbivory in fertilized plots 8 and 20 years since fire. In sites si x weeks after fire, the tops of several resprouts appeared to have been consumed by larger herbivores, but this occurred across treatments and seemed to occur most often in Quercus chapmanii not the more dominant species. Fourth, shrubs may have allocate d resources to reproductive parts that were not included in my aboveground biomass estimates. Acorn production differs between white oaks ( Quercus chapmanii Quercus geminata ) and red oaks ( Quercus inopina Quercus laevis Quercus myrtifolia ). Cycles of ac orn production range from 2 to 2.4 years for white oaks and from 3.6 to 5.5 years for red oaks (Abrahamson and Layne 2003). R esprouts of white oaks produce acorns d uring the first year after fire, whereas red oaks produce acorns three or four years after f ire (Abrahamson and Layne 2002a). In addition, the mean number of acorns per bearing individual tends to increase with ramet size (Abrahamson and Layne 2002b). I assessed acorn production by Q. inopina and Q. geminata in September 2008, one year after fert ilization. Overall, I counted 13 acorns on 735 stems, 273 acorns on 378 stems, and 103 acorns on 350 stems in sites one year, nine years, and 21 years after fire, respectively. Most stems, however, produced no acorns. Only nine stems produced more than 10 acorns, and the stems that produced multiple acorns were from a variety of nutrient treatments. Abrahamson and Layne (2003) found that precipitation accounted for 44% to 74% of the variation in crop size of the oaks, suggesting that nutrient availability m ay play a secondary role in affecting oak reproduction. Flowering of Sabal etonia or Serenoa repens increased in the first year or two after fire, but was not stimulated but fertilization (Abrahamson 1999). I counted reproductive stems, but did not quantif y flowering or fruiting, of S.

PAGE 173

173 repens individuals after the first and second years of fertilization. Similar to Abrahamson (1999), I found no effect of nutrient addition on allocation to reproductive stems. After the first year of the experiment, more repr oductive stems were produced by plants in sites one year after fire, while after the second year of the experiment, plants in sites nine years after fire produced more reproductive stems than plants in sites two years after fire. Fifth, resources other than N and P may have limited plant growth. Scrubby flatwoods sandy soils have low water holding capacity (Abrahamson et al. 1984), and low water availability may limit the ability of plants to respond to increased nutrient availability. In coastal scrub, the biomass increment of oaks was positively correlated with annual rainfall (Seiler et al. 2009). In my study, rainfall was lower than average and similar to average during the first and second years of fertilization, respectively (Figure 5 1). Quercus in opina has a shallower depth of water uptake than Q. chapmanii and Q. geminata as indicated by analysis of stable oxygen isotopes (Saha et al. 2008). Thus, Q. inopina in particular may have experienced water limitation and drought stress during months with low rainfall. Larger plants, which occur in longer unburned areas, may deplete water resources more rapidly than smaller plants; however, gravimetric soil moisture, leaf water potential, and stomatal conductance do not vary with time since fire (K. Adams and S. Saha, unpublished data), suggesting that water limitation may not be a confounding factor with time after fire. Light limitation, particularly in longer unburned sites, may hinder the ability of plants to respond to increased nutrient availability. For seedlings of the tropical tree Inga vera light levels had a greater effect on growth than N and P addition (Myster 2006). In scrubby flatwoods, oaks and ericaceous shrubs are

PAGE 174

174 the dominant overstory species, and leaf blade area and canopy area of the p almettos Serenoa repens and Sabal etonia are positively correlated with percent overstory (Abrahamson 2007). Biomass of S. repens increased with N addition in sites 20 years after fire, which is when light limitation may be highest, suggesting that plants were investing in increasing leaf and canopy area to increase light reception. In addition, other nutrients such as potassium (K) or calcium (Ca) may have limited productivity or fertilization with only one nutrient, such as N, could have induced P deficie ncy (Teng and Timmer 1995). In a flooded savanna, only addition of N, P, K, and sulfur (S) together increased plant productivity (Sarmiento et al. 2006), and biomass of Serenoa repens and Sabal etonia increased with addition of fertilizer that contained N, P, K, Ca, and magnesium (Mg) (Abrahamson 1999). The P fertilizer I used, triple superphosphate, contained Ca in addition to P. Background levels of Ca in scrubby flatwoods range from 12.6 to 25.2 g m 2 (Abrahamson et al. 1984). At my rate of P addition, however, I only added 0.66 to 3.31 g Ca m 2 which likely would not have limitation exists despite high annual rates of net N mineralization, possibly because of asynchron y in plant nutrient demand and nutrient availability; however, by adding fertilizer numerous times over one year, I hopefully mitigated asynchrony in nutrient availability and plant nutrient demand. Sixth, plants may have allocated resources to acquiring m ore resources rather than in growth. For example, nutrient addition increased root productivity eight and 20 years after fire, when nutrient addition had less of an effect on aboveground biomass

PAGE 175

175 and growth of Q. inopina and S. repens In addition, increasi ng foliar nutrient concentrations may occur at a cost to increasing biomass and growth. I found large variation in the mean percent change in biomass and growth within a treatment, even when there were not significant block effects. This variation could be related to difference in ages of clumps of stems or differences in growth among specific oak clones or palmetto individuals. Aerts (2009) found intraspecific variation in the response of leaf production to N addition and Bown et al. (2009) found that clon es of Pinus radiata differ in growth allocation, with the slowest growing genotype partitioning a smaller fraction of GPP to ANPP than the fastest growing genotype. Thus, genotypic variation may have increased variability in the effects of fertilization if all of the individuals of any species in one plot were the same clone. Assessment of nutrient limitation in an ecosystem may depend on the length of fertilization studies (Niinemets and Kull 2005), and other studies have found significant effects of n utrient addition after two years of fer tilization (Gkkaya et al. 2006; Lund et al. 2009). In my study, however, nutrient addition had a significant effect on plant biomass and growth in after one year, but not two years, of fertilization. This could be du e to the fact that in sites that were fertilized for two years, the percent change in aboveground biomass, across all treatments, was greater during the first year than during the second year of fertilization. Studies from a variety of other ecosystems hav e ack 2006; Darby and Turner 2008; Iversen and Norby 2008) and P addition (Gkkaya et al. 2006) can increase aboveground biomass and growth. This suggests that scrubby flatwoods shrubs have a limit on their ability to

PAGE 176

176 increase growth in response to nutrient addition, and that shrubs may maximize growth in the first year of increased nutrient availability. Litterfall Nutrient addition did not affect total leaf litterfall over the first year of fertilization (Table 5 7) Leaf litterfall of Quercus inopina had higher %N in high N + P treatments than in control plots eight years after fire; however, nutrient addition did not affect leaf litterfall N (g N m 2 2 mos 1 ) or biomass (g m 2 2 mos 1 ) of Q. inopina In oth er fertilization experiments, N and P fertilization increased litterfall in an N limited montane forest (Tanner et al. 1992) and a P limited Hawaiian forest (Herbert and Fownes 1995). The lack of an effect of nutrient addition on Q. inopina litterfall may be related to the fact that percent cover of Q. inopina was a predictor of leaf litterfall. There are gaps within the shrub matrix of scrubby flatwoods, and most shrub species, including Q. inopina have somewhat patchy distributions. Although litter traps were spread throughout each plot, Q. inopina litter often accumulates under the canopy, which could have contributed to the large variation in litterfall and the lack of an effect of nutrient addition. Furthermore, Q. inopina stems near litter traps that died, which happened in several plots, contributed high amounts of leaf litter. Other shrub and sub shrub species had low leaf litterfall in general, but they may have responded differently to nutrient addition than Q. inopina because the mechanisms that c ause an increase in the amount of N and P transferred to the soil with nutrient fertilization differ among species (Aerts 2009). Leaves of both Serenoa repens and Sabal etonia have an average life span of 2 to 3 years, and the number of new leaves produced per year by S. repens and S. etonia is positively correlated with palmetto biomass (Abrahamson 2007). For palmettos, leaf

PAGE 177

177 turnover may be a better indicator than litterfall of nutrient effects of litter production because leaves remained suspended above t he ground for a while after they die. Foliar Nutrients Pre fertilization foliar N:P ratios suggested that Quercus inopina and Serenoa repens in recently burned sites would be N limited and in sites eight and 20 years since fire would be P limited or c o limited by N and P (Koerselman and Meuleman 1996; Gsewell 2004). In recently burned sites, high N addition biomass of S. repens ; whereas, high P and high N + P increased biomass and growth of Q. inopina suggesting that these species are limited by diff erent nutrients. Eight and 20 years after fire, nutrient addition had little effect on growth of Q. inopina but high N addition increased biomass of S. repens Thus, foliar N:P ratios do not appear to be accurate indicators of aboveground nutrient limitat ion in scrubby flatwoods. Similarly, in South African grasslands, foliar N:P ratios suggested sites were N limited when all sites were co limited by N and P (Craine et al. 2008). In recently burned sites, foliar %N and %P of Q. inopina and S. repens decre ased from pre fertilization to one year post fertilization regardless of fertilization treatment because the effects of fire on foliar nutrient concentrations were greater than the effects of nutrient addition. High N and N + P addition, however, caused le ss of a decrease in foliar %P of both species. Eight years after fire, high P addition caused a significant increase in foliar %P of Q. inopina which led to a significant decrease in foliar N:P ratios. This is consistent with a study in an alpine ecosyste m where P fertilization caused a decrease in foliar N:P ratios (Litaor et al. 2008). Twenty years after fire, high P and high N+ P increased foliar %P of Q. inopina but this did not cause significant decline in foliar N:P ratios. Eight and 20 years after fire, foliar %P and N:P

PAGE 178

178 ratios of S. repens were not affected by nutrient addition. Pre fertilization, foliar %N was significantly higher in Q. inopina than in S. repens 20 years after fire, and high N and N +P addition increased foliar %N of S. repens bu t not foliar %N of Q. inopina Similarly, Abrahamson (1999) found that N fertilization increased foliar N concentrations of Serenoa repens My results suggest that Q. inopina may be a better competitor for N in long unburned sites and Q. inopina has come c lose to maximizing foliar N content. On the other hand, because S. repens had a lower initial foliar %N, it likely had a greater capacity to increase foliar %N. Regardless, across all nutrient treatments, foliar %N of Q. inopina tended to increase over the first year of fertilization; whereas, foliar %N of S. repens decreased in many treatments (Figure 5 12). Overall, Q. inopina had a greater percent change in aboveground biomass, growth, and foliar nutrient concentrations in response to P addition, while S repens had a greater percent change in aboveground biomass and foliar nutrient in response to N addition. This could in part be related to the different mycorrhizal associations of each species. Ectomycorrhizae may be better able to increase N to Q. inop ina leading to greater P limitation; whereas arbuscular mycorrhizae may be better able to increase P to S. repens leading to greater N limitation. This suggests that within scrubby flatwoods ecosystems, differences among species may have a larger effect on nutrient limitation than time after fire. Root P roductivity Nutrient addition did not affect root productivity in recently burned sites; however high N + P addition significantly increased root productivity eight and 20 years after fire (Figure 5 14). It is not clear if higher root biomass was due to an increase in root length or an increase in root width, but Blair and Perfecto (2008) found that root length, and not

PAGE 179

179 root width, was greater in soils with added N and P. Because aboveground biomass showed a stronger response to nutrient additi on in recently burned sites than other times after fire, my results suggest that there is a trade off between allocation to aboveground and belowground resources with nutrient addition. Although an increase in soil nutri ent availability can cause a decrease in allocation to belowg round biomass (Nadelhoffer 2000; Darby and Turner 2008), allocation to belowground biomass is proportionally greater in low nutrient and dry ecosystems (Chapin et al. 2002) Eighty percent of bio mass may be belowground in Florida scrub ecosystems (Saha et al. in review), and within oak domes, two thirds of the biomass is belowground (Guerin 1993). Furthermore, belowground reserv es (McPherson and Williams 1998; Paula and Ojeda 2009) affect resprout ing ability (Moreno and Oechel 1991) and biomass of resprouts (Llo ret and Lpez Soria, 1993; Cruz et al. 2002). Because dominant scrubby flatwoods shrubs resprout after fire, in longer unburned sites (e.g. eight and 20 years after fire), increasing belowgr ound biomass is likely important for future survival; aboveground biomass will be consumed in the next fire. I did not sort roots to species, but roots from all the dominant shrub groups (oaks, palmettos, ericaceous shrubs) were present in root ingrowth co res, suggesting that increased root growth is a common response of dominant scrubby flatwoods species. Conclusions The effects of nutrient addition on biomass and growth depend both on time after fire and species Over the first year of fertilization in r ecently burned sites, aboveground biomass and growth of Quercus inopina responded to P and N + P addition, while aboveground biomass of Serenoa repens responded to N addition. Nutrient addition had no effect on leaf litterfall or root productivity. Thus, r esprouts of scrubby flatwoods

PAGE 180

180 shrubs are able to respond to nutrient addition above nutrients made available by fire, and they do so by increasing aboveground growth. Different species, however, appear to be limited by different nutrients. Eight years afte r fire, apical shoot growth and foliar %P of Q. inopina increased with nutrient addition, and high N + P addition increased root productivity. At intermediate times after fire, scrubby flatwoods shrubs appear to invest more in belowground than aboveground productivity and show co limitation by N and P with a stronger P limitation. Twenty years after fire, N addition increased biomass and foliar %N of S. repens P addition increased apical shoot growth of Q. inopina and N + P addition increased root product ivity. In long unburned sites, scrubby flatwoods appear to invest in both aboveground and belowground productivity and show co limitation by N and P. Thus, my results support my hypothesis that plant productivity in long unburned scrubby flatwoods is co li mited by N and P. My hypotheses that plant productivity in recently burned scrubby flatwoods is N limited and plant productivity in intermediately burned scrubby flatwoods is P limited, however, are only partially supported. In sites that were fertilized f or two years, any increase in biomass or growth with nutrient addition occurred during the first year of fertilization. Dominant scrubby flatwoods species were only able to respond to one years of increased nutrient availability, which is what would occur naturally after fire (Chapter 1). This suggests that shrubs are adapted to respond to nutrients made available by fire, but may not be capable of increased growth above what is possible with nutrients made available by fire. Alternatively, microbes may be better competitors for nutrients than plants, and if fire caused a decrease in microbial biomass, over the second year of fertilization in

PAGE 181

181 recently burned sites, once microbes had recovered, microbial uptake of nutrients could have limited the response of plants to nutrient addition. In addition, the sandy soils of scrubby flatwoods likely have a limited ability to retain added nutrients, and in the second year of fertilization added nutrients could have leached from the soil such that plants were not able to acquire the added nutrients. I only fertilized scrubby flatwoods for one or two years, which is a short time in the 8 16 year fire return interval of scrubby flatwoods (Menges 2007). Perhaps over a longer time period, shrubs would be able to adapt to el evated nutrient availability rather than just a pulse of nutrients. Large areas of land along the Lake Wales Ridge have been converted to agriculture and pastureland, leading to increased nutrient inputs. Some of these areas are now being restored to Flori da scrub, and my results suggest that increased nutrient availability may not benefit native shrub species.

PAGE 182

182 Table 5 1. Mean (median) percent of total shrub cover of focal shrubs in all plots pre fertilization at different times after fire and results of Kruskall Wallis tests analyzing differences with time after fire (df = 2 for all tests). Different letters represent significant di Percent Cover Percent of Total Shrub Cover 2 P 6 weeks 8 years 20 years 2 p 6 weeks 8 years 20 years Ericaceous shrubs Lyonia fruticosa 5.83 0.054 0.7 (0) 4.9 (0.5) 2.7 (0) 4.97 0.083 2.3 (0) 7.8 (0.7) 4.5 (0) Lyonia lucida 8.46 0.015 1.5 a (1) 1.4 b (0) 3.1 ab (1) 12.12 0.002 6.8 a (3.6) 2.0 b (0) 5.2 ab (1.54) Oaks Quercus chapmanii 0.73 0.692 1.4 (0) 1.9 (0) 3.4 (0) 0.14 0.931 5.6 (0) 3.4 (0) 5.4 (0) Quercus gemin ata 0.26 0.879 1.3 (1) 3.2 (1) 3.3 (1) 7.34 0.026 5.9 a (4.6) 5.2 b (1.5) 5.8 ab (1.4) Quercus inopina 44.6 <0.001 11.2 a (10) 28.0 b (25) 31.8 b (30) 3.52 0.172 44.6 (48.2) 47.0 (45.4) 51.9 (54.9) Palmettos Sabal etonia 29.31 <0.001 1.3 a (1) 4.7 b (5) 0.8 c (0) 24.59 <0.001 6.0 a (4.3) 7.9 a (8.0) 1.6 b (0) Serenoa repens 43.61 <0.001 6.0 a (5) 14.0 b (12.5) 13.7 b (10) 0.68 0.711 25.9 (25.1) 24.2 (22.1) 23.6 (21.8) Total 58.16 <0.001 23.4 a 58.1 b 58.9 b 0.99 0.608 96.9 97.5 98.0

PAGE 183

183 Table 5 2. Mean (se) concentrations of macronutrients and micronutrients in scrubby flatwoods soils. Time since fire Al (%) Ca (%) Co (ppm) Cu (ppm) Fe (%) K (%) Mg (%) Mn (ppm) Mo (ppm) Na (%) Ni (ppm) P (ppm) S (%) Zn (ppm) 1 year 0 10 cm 0 .023 (0.003) 0.027 (0.007) 0.67 (0.03) 10.47 (0.96) 0.36 (0.022) <0.01 <0.01 31.67 (2.18) 7.72 (1.23) <0.01 5.13 (0.73) 20 (0) <0.01 7.67 (1.45) 10 20 cm 0.017 (0.003) 0.01 (0) 0.77 (0.18) 8.5 (0.49) 0.36 (0) <0.01 <0.01 25.67 (2.18) 8.87 (1.60) <0.01 4. 60 (0.78) 10 (0) <0.01 4.0 (1.0) 8 years 0 10 cm 0.02 (0) 0.017 (0.003) 0.70 (0.11) 8.43 (0.62) 0.38 (0.017) <0.01 <0.01 30.83 (5.26) 8.88 (1.72) <0.01 4.52 (0.57) 10 (0) <0.01 3.0 (0.58) 10 20 cm 0.02 (0) 0.01 (0) 0.73 (0.09) 14.7 (3.12 ) 0.48 (0.07) <0.01 <0.01 32.33 (3.76) 8.54 (0.59) <0.01 5.23 (0.58) 10 (0) <0.01 4.0 (1.0) 20 years 0 10 cm 0.023 (0.003) 0.013 (0.003) 0.73 (0.03) 13.6 (3.18) 0.43 (0.036) <0.01 <0.01 31.0 (0.58) 9.11 (0.88) <0.01 5.47 (0.42) 13.3 (3.3) <0.01 11.0 (4.0) 10 20 cm 0.02 (0) 0.01 (0) 0.70 (0.10) 10.86 (3.03) 0.37 (0.015) <0.01 <0.01 24.0 (1.73) 8.99 (0.91) <0.01 4.43 (0.45) 10 (0) <0.01 3.0 (0.58)

PAGE 184

184 Table 5 3. Results of one way ANOVAs (df = 2,6) and Kruskall Wallis tests (df = 2) analyzin g differences in soil nutrients with time after fire in surface (0 10 cm) and deep (10 20 cm) soils. Element 2 p df 0 10 cm Al (%) 1.14 0.565 2 Ca (%) 3.44 0.179 2 Co (ppm) 0.96 0.619 2 Cu (ppm) 1.80 0.244 2,6 Fe (%) 2.03 0.213 2,6 Mn (ppm) 0.02 0.982 2,6 Mo (ppm) 0.32 0.739 2,6 Ni (ppm) 0.67 0.547 2,6 P (ppm) 5.60 0.061 2 Zn (ppm) 5.14 0.077 2 10 20 cm Al (%) 2.00 0.368 2 Ca (%) 0 .00 1 .00 2 Co (ppm) 0.16 0.921 2 Cu (ppm) 1.53 0.290 2,6 Fe (%) 2.59 0.154 2,6 Mn (ppm) 2.66 0 .149 2,6 Mo (ppm) 0.04 0.958 2,6 Ni (ppm) 0.47 0.647 2,6 P (ppm) 0 .00 1 .00 2 Zn (ppm) 0.92 0.631 2

PAGE 185

185 Table 5 4 Results of general linear univariate models with treatment as a fixed effect and block as a random effect for soil nutrients. 1 year after fire 9 years after fire 21 years after fire F df p F df p F df p Total Inorganic N b 1 ) 0 10 cm Treatment 0.98 9,18 0.485 3.44 9,18 0.012 1.40 9,18 0.257 Block 8.50 2,18 0.003 12.16 2,18 <0.001 4.20 2,18 0.032 1 0 20 cm Treatment 0.80 9,18 0.618 0.75 9,18 0.663 3.70 9,18 0.009 Block 15.80 2,18 <0.001 5.06 2,18 0.018 7.32 2,18 0.005 NH 4 + b 1 ) 0 10 cm Treatment 1.13 9,18 0.391 3.25 9,18 0.016 1.15 9,18 0.381 Block 6.96 2,18 0.006 11.99 2,18 <0.001 2.93 2,18 0.079 10 20 cm Treatment 0.97 9,18 0.495 0.79 9,18 0.630 2.95 9,18 0.024 Block 16.09 2,18 <0.001 5.10 2,18 0.018 6.65 2,18 0.007 Total Inorganic P a 1 ) 0 10 cm Tr eatment 2.32 9,18 0.061 20.51 9,18 <0.001 6.48 9,18 <0.001 Block 5.59 2,18 0.013 0.45 2,18 0.643 0.24 2,18 0.789 10 20 cm Treatment 3.87 9,18 0.007 27.57 9,18 <0.001 6.96 9,18 <0.001 Block 5.18 2,18 0.017 0.41 2,18 0.668 0.13 2,18 0.879 N:P a 0 10 cm Treatment 3.51 9,18 0.011 27.77 9,18 <0.001 8.26 9,18 <0.001 Block 3.34 2,18 0.058 5.15 2,18 0.017 1.35 2,18 0.285 10 20 cm Treatment 4.60 9,18 0.003 11.45 9,18 <0.001 18.80 9,18 <0.001 Block 0. 39 2,18 0.680 4.48 2,18 0.026 6.17 2,18 0.009

PAGE 186

186 Table 5 4 continued 1 year after fire 9 years after fire 21 years after fire F df p F df p F df p %N bd 0 10 cm Treatment 0.39 9,18 0.923 1.87 9,18 0.124 0.41 9,18 0.912 Block 2.99 2 ,18 0.076 6.99 2,18 0.006 1.13 2,18 0.346 %C bd 0 10 cm Treatment 0.54 9,18 0.829 1.65 9,18 0.175 0.45 9,18 0.890 Block 0.85 2,18 0.445 0.96 2,18 0.402 0.52 2,18 0.603 C:N c 0 10 cm Treatment 1.44 9,18 0.242 1.27 9,18 0.3 17 0.50 9,18 0.858 Block 12.27 2,18 <0.001 3.45 2,18 0.054 11.60 2,18 0.001 pH c 0 10 cm Treatment 0.55 9,18 0.822 1.99 9,18 0.102 1.71 9,18 0.158 Block 0.98 2,18 0.393 0.13 2,18 0.878 1.20 2,18 0.324 10 20 cm Treat ment 1.05 9,18 0.441 0.22 9,18 0.986 0.38 9,18 0.927 Block 1.42 2,18 0.267 0.63 2,18 0.544 1.47 2,18 0.256 a all data natural log transformed b data for 1 year after fire natural log transformed c data for 9 years after fire natural log transformed d data for 20 years after fire natural log transformed

PAGE 187

187 Table 5 5 Results of general linear univariate models with treatment as fixed eff ect and block as random effect testing for treatment effects on percent change in biomass and growth. 6 weeks afte r fire 8 years after fire 20 year after fire 1 st year of fertilization 2 nd year of fertilization 1 st year of fertilization 2 nd year of fertilization 1 st year of fertilization F df p F df p F df p F df p F df p Total plot biomass T reatment 0.53 9,18 0.837 1.84 9,18 0.130 1.04 9,18 0.450 Block 0.56 2,18 0.582 7.13 2,18 0.005 0.53 2,18 0.600 S. repens biomass Treatment 3.01 9,78 0.004 1.25 9,78 0.275 0.93 9,78 0.500 1.54 9,78 0.148 1.84 9,78 0.074 Block 11.30 2,78 <0.001 2.92 2,78 0.060 2.07 2,78 0.133 0.24 2,78 0.788 0.13 2,78 0.876 Q.inopina biomass Treatment 3.15 9,77 0.003 1.32 9,76 0.239 0.35 9,77 0.953 0.95 9,72 0.485 0.75 9,76 0.661 Block 2.75 2,77 0.070 1.83 2,76 0 .167 0.43 2,77 0.650 1.25 2,72 0.293 0.26 2,76 0.773 Q. inopina basal diameter bc Treatment 2.48 9,76 0.015 1.86 9,75 0.071 2.06 9,77 0.043 1.54 9,70 0.151 1.44 9,73 0.189 Block 0.15 2,76 0.862 4.91 2,75 0.010 2.93 2,77 0.059 0.51 2,70 0.601 0.16 2,73 0.854 Q. inopina height a Treatment 1.91 9,77 0.062 0.86 9,75 0.560 1.32 9,77 0.241 1.57 9,72 0.141 0.65 9,73 0.750 Block 0.98 2,77 0.381 0.77 2,75 0.466 1.20 2,77 0.306 2.20 2,72 0.118 0.68 2,73 0.509 Q. inopi na new growth (mm) a Treatment 3.88 9,77 <0.001 1.10 9,76 0.107 2.14 9,78 0.036 0.98 9,71 0.463 2.78 9,74 0.007 Block 2.07 2,77 0.134 1.33 2,76 0.138 5.30 2,78 0.007 2.64 2,71 0.079 3.81 2,74 0.027 a all data natural log transforme d b data for 1 year after fire natural log transformed c data for 20 years after fire natural log transformed

PAGE 188

188 Table 5 6. Results of Kruskal Wallis tests analyzing the effects of treatment on Quercus inopina stem survival and turnover over the first year of fertilization. Degrees of freedom = 9 for all tests. There were no block effects. 6 weeks after fire 8 years after fire 20 years after fire 2 p 2 p 2 p Percent survival of stems 7.90 0.544 19.57 0.021 10.72 0.295 Number of new stems 7.70 0.565 13.06 0.160 12.57 0.183 Percent change in stem number 10.73 0.294 4.90 0.843 10.87 0.285

PAGE 189

189 Table 5 7. Results of general univariate models with treatment as a fixed effect, block as a random effect, and a covariate. Time after fire 6 weeks 8 years 2 0 years F p F p F p Litterfall (g m 2 year 1 ) a Treatment 0.44 9,17 0.896 1.51 9,17 0.221 0.19 9,17 0.992 Block 0.26 2,17 0.776 0.97 2,17 0.398 0.99 2,17 0.391 Mean (Pre and Post fertilization) Q.inopina % cover 0.01 1,17 0.947 0.21 1,17 0.651 5.19 1,17 0.036 Q. inopina litterfall (g m 2 2 mos 1 ) b Treatment 0.40 9,17 0.920 1.45 9,17 0.242 0.25 9,17 0.981 Block 1.61 2,17 0.230 0.97 2,17 0.400 0.56 2,17 0.582 Post fertilization Q. inopina % cover 10.05 1,17 0.006 3.23 1,17 0.091 11.75 1,17 0.003 Q. inopina litterfall N (g N m 2 2 mos 1 ) b Treatment 0.56 9,17 0.811 1.28 9,17 0.316 0.29 9,17 0.969 Block 1.77 2,17 0.199 1.58 2,17 0.235 0.44 2,17 0.653 Post fert ilization Q. inopina % cover 11.03 1,17 0.004 2.83 1,17 0.111 10.50 1,17 0.005 Root Productivity (g m 2 year 1 ) Treatment 2.08 9,77 0.041 3.15 9,77 0.003 2.55 9,77 0.013 Block 5.28 2,77 0.007 0.74 2,77 0.479 0.05 2,77 0.952 Pre fertilization root biomass (g m 2 ) 4.99 1,77 0.028 5.52 1,77 0.021 2.81 1,77 0.098 a natural log transformed b data from sites 6 weeks and 20 years after fire natural log transformed

PAGE 190

190 Table 5 8. Effects of treatment on percent change in foliar and l itter nutrients Results of general linear univariate models with treatment as a fixed effect and block as a random effect. 6 weeks after fire 8 years after fire 20 years after fire F df p F df p F df p Q. inopina litterfall %N a Treatmen t 0.71 9,18 0.693 3.44 9,18 0.012 1.03 9,18 0.451 Block 0.07 2,18 0.983 1.21 2,18 0.320 0.07 2,18 0.936 Q. inopina foliar %N Treatment 2.11 9,77 0.038 1.95 9,78 0.056 1.33 9,75 0.236 Block 6.12 2,77 0.003 2.40 2,78 0.097 9.31 2,75 <0 .001 Q. inopina foliar %P Treatment 2.96 3,29 0.048 9.07 3,30 <0.001 4.82 3,28 0.008 Block 0.42 2,29 0.661 1.82 2,30 0.179 0.14 2,28 0.868 Q. inopina foliar N:P Treatment 2.32 3,29 0.096 6.41 3,30 0.002 2.28 3,28 0.101 Block 2.16 2,29 0.133 0.10 2,30 0.905 2.69 2,28 0.086 S. repens foliar %N (g) Treatment 0.50 9,78 0.868 0.96 9,78 0.477 2.56 9,78 0.012 Block 2.32 2,78 0.105 0.04 2,78 0.958 6.51 2,78 0.002 S. repens foliar %P Treatment 2 .85 3,30 0.054 0.24 3,30 0.866 0.77 3,30 0.519 Block 0.79 2,30 0.463 0.27 2,30 0.764 2.91 3,30 0.070 S. repens foliar N:P (g) Treatment 1.76 3,30 0.176 0.13 3,30 0.941 0.73 3,30 0.544 Block 1.20 2,30 0.315 0.07 2,30 0.931 0.46 2,30 0.634 a all data natural log transformed

PAGE 191

191 Table 5 9. Results of one way ANOVAs analyzing pre fertilization foliar nutrients of Q uercus inopina and S erenoa repens F df p %N YSF 154.62 2 <0.001 Species 101.15 1 <0.001 YSF*Species 13.50 2 <0.001 %P YSF 403.21 2 <0.001 Species 88.14 1 <0.001 YSF*Species 40.46 2 <0.001 N:P YSF 169.96 2 <0.001 Species 4.64 1 0.035 YSF*Species 3.10 2 0.052 Figure 5 1. Mean monthly precipitation and tota l monthly precipitation during the study period at Archbold Biological Station. Arrows indicate months in which fertilizer was added.

PAGE 192

192 Figure 5 2. Mean ( + se) resin exchangeable inorganic N (A), phosphorus (B), and N :P ratios (C) in control plots over one year. Different letters represent significant

PAGE 193

193 Figure 5 3. Mean ( + se) soil extractable inorganic N in surface (0 10 cm) and deep (10 20 cm) soils in co ntrol and treatment plots in sites 1 year, 9 years, and 21 Figure 5 4. Mean ( + se) soil extractable P in surface (0 10 cm) and dee p (10 20 cm) soils in control and treatment plots in sites 1 year, 9 years, and 21 years since

PAGE 194

194 Figure 5 5. Mean ( + se) percent change in Seren oa repens biomass during the first and second years of fertilization in sites 6 weeks (A,B), 8 years (C,D), and 20 years (E) since fire. indicates a treatment is significantly different from the

PAGE 195

195 Figure 5 6. Mean ( + se) percent change in Quercus inopina plant biomass during the first and second years of fertilization in sites 6 weeks (A,B), 8 years (C,D), and 20 years (E) since fire. indicates a treatment is different from the control .05.

PAGE 196

196 Figure 5 7 Mean ( + se) percent change in Quercus inopina stem height during the first and se cond years of fertilization in sites 6 weeks (A,B), 8 years (C,D), and 20 years (E) since fire for stems alive duri ng each year of fertilization. indicates

PAGE 197

197 Figure 5 8 Mean ( + se) percent change in Quercus inopina stem basal diameter during the first and second years of fe rtilization in sites 6 weeks (A,B), 8 years (C,D), and 20 years (E) since fire for stems alive during each year of fertilization.

PAGE 198

198 Figure 5 9. Mean ( + se) length of Quercus inopina total apical shoot growth increments during the first and second years of fertilization in sites 6 weeks (A,B), 8 years (C,D), and 20 years (E) since fire. indicates a treatment that is different from .05.

PAGE 199

199 Figure 5 10. Mean ( + se) Quercus inopina leaf litterfall from March to May (g m 2 2 mos 1 ) during the first year of fertilization in sites 6 weeks (A), 8 years (B), and 20 years (C) after fire.

PAGE 200

200 Figure 5 11. Mean ( + se) foliar %N (A), foliar %P (B), and foliar N:P ratios of Quercus inopina and Serenoa repens 6 weeks, 8 years, and 20 years since fire in all plots pre fertilization. Different letters represent significantly different means

PAGE 201

201 Figure 5 12. Mean ( + se) percent change in foliar %N of Q. inopina an d S. repens over the first year of fertilization in sites 6 weeks (A,B), 8 years (C,D), and 20 years

PAGE 202

202 Figure 5 13. Mean ( + se) perce nt change in foliar %P (A,B) and N:P ratios (C,D) of Q. inopina and S. repens over the first year of fertilization. indicates a treatment

PAGE 203

203 Figure 5 14. Mean ( + se) root productivity during the first year of fertilization. indicates

PAGE 204

204 CHAPTER 6 CONCLUSION Previous research in Florida scrub ecosystems has found limited effects of fire on soil nutrient av ailability (Schmalzer and Hinkle 1991). I found that a pulse of nutrients is detectable if soils are sampled soon enough after fire. F ire caused a short term increase in soil NH 4 + and PO 4 3 in a palmetto flatwoods ecosystem; PO 4 3 remained elevated above p re fire levels twice as long as NH 4 + possibly due to differences in microbial uptake and mobility of NH 4 + and PO 4 3 Both foliar %N and %P of resprouting plants increased over the short term after fire. The relative increase in soil and foliar P was great er than that of soil and foliar N, leading to a decrease in the soil N:P ratio and the foliar N:P ratio of the palmetto Serenoa repens shortly after fire. The difference in the magnitude of the decrease in soil and foliar N:P ratios after fire coupled with 15 N suggest that both increased soil nutrient availability and reallocation of nutrients from belowground to aboveground tissue contribute to the increase in foliar %N and %P shortly after fire. M y results suggest that eve n a short term increase in soil nutrient availability can be important for plant nutrient status, especially in ecosystems with low nutrient availability. Overall, growth, allocation of resources to aboveground tissues, and allometric equations tended to d iffer with time after fire, but the majority of differences occurred among species and between recently burned and intermediate and longer unburned sites. Resprouting ability of scrubby flatwood species appears to be important in determining growth and all ometry immediately after fire. Six weeks after fire, Quercus inopina the dominant species in my scrubby flatwoods study sites, had the highest height:diameter ratio and the highest leaf:stem biomass ratio. The ability of Q. inopina to

PAGE 205

205 resprout and aquire aboveground space after fire may contribute to the maintenance of its dominance in the sites where it occurs. The fire return interval for scrubby flatwoods in 8 16 years (Menges 2007), and there was little or no difference in plant size, biomass, or allo metric rel ationships from 8 22 years after fire within a species; however, there were differences among species within these times after fire. Thus, as time after fire increases, species specific constraints in growth and allometry appear to become more im portant. My results also suggest that caution should be taken in using allometric equations developed for plants from longer unburned sites to estimate biomass of plants in recently burned sites. In scrubby flatwoods soils, inorganic N availabili ty was not affected by time after fire, but is related to soil moisture, while variation in resin exchangeable PO 4 3 wa s related to soil pH. Resin exchangeable N was highest eight years after fire and resin exchangeable PO 4 3 was highest 13 years after fire, so N:P ratios increased then decreased with time after fire. Other measures of soil N (mineralization rates, %N, chloroform labile microbial N) were highest in surface soils (0 5 cm) in sites 13 years after fire. In addition, soil %C was highest in sites 13 years after fire. The increased in soil N and C in long unburned sites may be due to the differences in species composition and fire history of sites 13 years after fire compared to scrubby fla twood sites of other times after fire rather than the direct effect s of time after fire. In this study, sites 13 years since fire had numerous pine trees, high soil organic matter, and had only been burned once in the past 35 years, which may have contributed to high PO 4 3 soil %N, and soil %C in surface soils; however, in scrubb y flatwoods sites 20 years after fire with low soil organic matter, and where pines are less

PAGE 206

206 abundant or absent, soil %N and %C are low er (J. Schafer, unpublished data ). Thus, my results indicating high nutrient availability in long unburned scrub by flatwoods may not apply to all scrubby flatwoods. Species composition and fire frequency, as well as time after fire, appear to be important in affecting nutrient availability in Florida scrub soils. Management of scrubby flatw oods ecosystems should inc lude burning every 8 16 years (Menges 2007). My results suggest that the trajectory of nutrient availability and accumulation may change if fire is suppressed or not prescribed at the suggested return intervals. The effects of nutrient addition on biomass and growth depend on time after fire. Over the first year of fertilization in recently burned sites, aboveground biomass and growth of Quercus inopina responded to P and N + P addition, while aboveground biomass of Serenoa repens responded to N addition. N utrient addition had no effect on leaf litterfall or root productivity. Thus, resprouts of scrubby flatwoods shrubs are able to respond to nutrient addition above nutrients made available by fire, and they do so by increasing aboveground growth. Different species, however, appear to be limited by different nutrients. Eight years after fire, apical shoot growth and foliar %P of Q. inopina increased with nutrient addition, and high N + P addition increased root productivity. At intermediate times after fire, scrubby flatwoods shrubs appear to invest more in belowground than aboveground productivity and show co limitation by N and P with a stronger P limitation. Twenty years after fire, N addition increased biomass and foliar %N of S. repens P addition increas ed apical shoot growth of Q. inopina and N + P addition increased root productivity. In long unburned sites, scrubby flatwoods appear to invest in both aboveground and belowground productivity and show co limitation by N

PAGE 207

207 and P. Thus, my results support my hypothesis that plant productivity in long unburned scrubby flatwoods is co limited by N and P. My hypotheses that plant productivity in recently burned scrubby flatwoods is N limited and plant productivity in intermediately burned scrubby flatwoods is P limited, however, are only partially supported. In sites that were fertilized for two years, any increase in biomass or growth with nutrient addition occurred during the first year of fertilization. Dominant scrubby flatwoods species were only able to resp ond to one years of increased nutrient availability, which is what would occur naturally after fire (Chapter 1). This suggests that shrubs are adapted to respond to nutrients made available by fire, but may not be capable of increased growth above what is possible with nutrients made available by fire. Alternatively, microbes may be better competitors for nutrients than plants, and if fire caused a decrease in microbial biomass, over the second year of fertilization in recently burned sites, once microbes h ad recovered, microbial uptake of nutrients could have limited the response of plants to nutrient addition. In addition, the sandy soils of scrubby flatwoods likely have a limited ability to retain added nutrients, and in the second year of fertilization a dded nutrients could have leached from the soil such that plants were not able to acquire the added nutrients. I only fertilized scrubby flatwoods for one or two years, which is a short time in the 8 16 year fire return interval of scrubby flatwoods (Menge s 2007). Perhaps over a longer time period, shrubs would be able to adapt to elevated nutrient availability rather than just a pulse of nutrients. Large areas of land along the Lake Wales Ridge have been converted to agriculture and pastureland, leading t o increased nutrient inputs. Some of these areas are now being restored to Florida scrub. M y results suggest that managing with fire

PAGE 208

208 should cause pulses of nutrient availability that can be used by plants and maintaining the proper fire return intervals co uld lead to patterns of nutrient availability similar to native scrub soils. Phosphorus increased growth of Quercus inopina and N increased growth of Serenoa repens suggesting that these species may occupy different niches in scrubby flatwoods communities Because scrub species are adapted to low nutrient availability and fire related increases in nutrient availability, and only increased aboveground growth during the first year of nutrient addition, the i ncreased availability of nutrients in restoration s ites may have limited benefits for native shrub species.

PAGE 209

209 APPENDIX A SHORT TERM EFFECTS OF FIRE ON SOIL AND PLANT NU TRIENTS IN SCRUBBY FLATWOODS Scrubby flatwoods are dominated by scrub oak ( Quercus inopina ), sand live oak ( Q uercus geminata ), palmettos, an d ericaceous shrubs. The shrubs are evergreen with an average height of 1 2 m, and herbaceous species are sparse (Abrahamson et al. 1984). Soils are entisols (Abrahamson et al. 1984) with no horizon development, little organic matter, and low exchange capa city and base saturation (Brown et al. 1990). Scrubby flatwoods experience fire return intervals of 8 16 years (Menges 2007), and dominant species resprout after fire. Because scrubby flatwoods have low standing biomass and patches of bare sand, fires ofte n create a mosaic of unburned to intensely burned areas (Abrahamson et al. 1984). I randomly selected six sampling locations within the scrubby flatwoods vegetation association in a 34 acre burn unit that had previously burned in 1994. At all sampling loc ations, which were separated by at least 10m, we marked a soil sampling site and the nearest individual of six different species (when present within 1m of the soil sampling location). My focal species, all of which resprout after fire, were the shrubs L yo nia lucida Q uercus geminata and Q uercus inopina the palmettos Sabal etonia and S erenoa repens and the sub shrub V accinium myrsinities On March 14 th 2006, I collected a soil sample (0 15cm depth, 8cm diameter core) at each sampling site and collected a foliar sample from all marked individuals. A prescribed fire was initiated on March 15 th 2006, but had to be extinguished due to high winds (K. Main, pers. comm.), and the fire was re ignited on March 16 th 2006. The fire was patchy with b urned and unbu rned areas; of my marked plants, 15 were consumed by the fire, 13 were scorched and two were unburned. The maximum one minute average temperature recorded was

PAGE 210

210 680C, while the absolute maximum temperature recorded was 870C (E. Menges, unpubl. data). Post fire soil samples were collected on March 17 th May 8 th July 17 th and October 5 th 2006. The first post fire foliar samples were collected on May 8 th 2006, but all marked plants had not yet resprouted. Subsequent post fire foliar collections were made o n July 17 th 2006 and October 6 th 2006. Soil samples were returned to the lab, passed through a 2 mm sieve, and then sub sampled for determination of pH, total percentages of N and C, and extractable P. Soil pH was measured as described above. Soil sub samples for determination of percent N and C were dried at 60C for 48 hr then ground to a fine powder with a mortar and a pestle. Leaf samples were dried at 60C for 48 hours and ground on a Wiley mill (Thomas Scientific, Swedesboro, NJ, USA) with a no. 4 0 screen. Soil and foliage was analyzed for percentages of N and C at the University of Florida on an ECA 4010 elemental analyzer (Costech Analytical, Valencia, California, USA). Foliar P was measured as describe above. To measure inorganic P concentration s, 30 mL of 0.05 M hydrochloric acid (HCl) and 0.0125 M hydrogen sulfate (H 2 SO 4 ) was added to 15 g of air dried soil, shaken for 5 min, then filtered through Whatman #42 filter paper. Filtered samples were stored in a refrigerator for several days before a nalysis for phosphate (PO 4 3 ) concentrations on a n absorbance microplate reader at the University of Florida using the ascorbic acid molybdenum blue method (Murphy and Riley 1962). All data were analyzed using repeated measures analysis of variance to exam ine changes in variables over time after fire. Variable means were compared with

PAGE 211

211 Bonferroni confidence interval adjustments. Non normal data were transformed to meet the assumptions of normality. All other data were analyzed using SPSS 11.5. S oil pH tended to increase over time after fire; however, there were no differences in soil %N, %C, or C:N with time since fire. Although there was not a significant difference in soil P over time after fire, soil P tended to increase then decrease (Table A 1 ). For S. r epens present at the first sam pling date, foliar %N and foliar %P increased over time after fire (Figure A 1) and foliar N:P var ied over time (Table A 2 ). Foliar %N of Q. geminata and Q. inopina increased after fire (Figure A 2 ). S abal etonia had new leav es by the second sampling date and foliar %N increased over time (Table A 2 ). Foliar %N of ericaceous species tended to be higher post fire than pre fire regardless of the time of first resprouting (Figure A 2) Foliar %N of the palmettos S. repens and S. etonia increased over time, while foliar %N of oaks and ericaceous species tended to increase then decreased over time after fire (Figures A 2 and A 3). This suggests that different mechanisms may be causing the observed changes in foliar %N for different species in the same habitat. The ratio of belowground to aboveground biomass of palmettos and oaks in scrubby flatwoods is 7.35 and 2.13, respectively (Saha et al. in review ). Higher root to shoot ratios may confer palmettos with a greater capacity to ret ranslocate nutrients from belowground to aboveground, allowing palmettos in scrubby flatwoods to continue to increase foliar %N after soil N availability has returned to pre fire levels.

PAGE 212

212 Table A 1 Mean ( + se ) of soil variables measured in the scrubby f latwoods si te before fire and after fire with results of repeated measures analyses. Degrees of freedom = 4 for all variables Pre Fire Days Post Fire Variable F p 1 52 122 202 pH* 5.2 8 0.028 4.44 + 0.17 4.56 + 0.1 7 4.65 + 0.2 5 5.11 + 0.26 5.09 + 0.2 6 P g g soil 1 ) 2.10 0.118 2.25 + 0.39 5.02 + 1.33 4.16 + 0.8 5 3.87 + 0.85 3.39 + 1.6 4 %C* 2.06 0.124 1.46 + 0.32 2.09 + 0.61 1.14 + 0. 30 1.38 + 0.2 7 1.11 + 0.35 %N* 2.28 0.096 0.056 + 0.011 0.076 + 0.021 0.041 + 0.009 0.049 + 0.008 0.042 + 0.014 C:N 0.8 5 0.512 25.68 + 1.68 26.98 + 1.5 5 26.47 + 2.04 27.73 + 1.4 3 26.56 + 1.32 data were natural log transformed

PAGE 213

213 Table A 2. Results of repeated measures analyses of variance for foliar %N, %P, and N:P ratios. Ericaceae includes L. lucida and V. my rsinities T he number in parentheses indicates the number of post fire sampling dates based on time of resprouting. %N %P N:P F df p F df p F df p Quercus geminata (3) 86.41 3 < 0.001 Quercus geminata (2) 5.66 2 0.068 Quercus inopina (3) 11.12 3 0.001 Sabal etonia (2) 31.71 2 < 0.001 Serenoa repens (3) 14.50 3 < 0.001 16.32 3 < 0.001 5.35 3 0.014 Ericaceae (3) 10.13 3 0.009 Ericaceae (1) 80.41 1 0.071

PAGE 214

214 Figure A 1. Mean ( + se) foliar %N (A), foliar %P (B ), and foliar N:P ratios (C ) of S. repens in scrubby flatwoods (n = 5) Different l etters represent significant differences in mean values at

PAGE 215

215 Figure A 2 Mean ( + se) foliar %N of plants in scrubby flatwoods pre fire and over time post fire. Top panel: plants present at three sampling dates post fire (n = 3, 5, 5, and 3 for Q. geminata Q. inopi na S. repens and Ericaceae, respectively). Q. geminata : lowercase a and b; Q. inopina : uppercase; S. repens : lowercase c and d) ; Middle panel: plants present at two sampling dates pos t fire (n = 3, 6, 1, and 1 for Q. geminata S. etonia S. repens and V. myrsinities respectively). Different letters represent significant 0.015 ; Bottom panel: plants present at one sampling date post fire (n = 2 ). Ericaceae incl udes V. myrsinities and L. lucida

PAGE 216

216 APPENDIX B SEASONAL VARIATION I N RESIN EXCHANGEABLE NITROGEN AND PHOSPHORUS IN SCRUBB Y FLATWOODS Seasonal effects may mediate fire effects on N and P availability. For instance, seasonality of precipitation causes varia tion in soil moisture, despite fire history (Todd et al. 2000), and net N mineralization is correlated with soil moisture (Evans et al. 1998 ; Frank 2008). Differences in NH 4 + and NO 3 with time since fire may be greater during the winter compared to the sp ring (Durn et al. 2009), suggesting that seasonality of plant growth and nutrient uptake influences fire effects on nutrient availability. Furthermore, fire severity varies with burn season, and short term effects of fire on soil abiotic conditions and nu trient availability depend on burn season (Hamman et al. 2008). Thus, I was interested in how availability of N and P varies seasonally. In May 2005, I established eighteen 30 x 30 m plots in scrubby flatwood communities (Abrahamson et al. 1984), three eac h in sites 1, 4, 6, 8, 10, and 13 years since fire. Within a time since fire, plots were located in different burn units when possible. Plots in the same burn unit were separated by at least 150 m and may have experienced differe nces in fire intensity Ove rall, plots covered a distance of approximately four miles, and although summer thunderstorms can be patchy, all plots experience the same climate. All plots were established in scrubby flatwoods dominated by scrub oak ( Quercus inopina Ashe ) on flat or gen tly sloped sites. Thus, the climate, organisms, relief, and parent material were the same in all sites (Jenny 1941 ); the only state factor that varied among sites was time after fire. In each plot, I established 30m transects across each plot that were ini tiated at 5m, 10m, 15m, 20m, and 25m along the NW to SW side of the plot.

PAGE 217

217 In all plots, I used ion exchange resins to measure soil nitrate (NO 3 ), ammonium (NH 4 + ), and phosphate (PO 4 3 ). At a random location on each transect, separate anion and cation exc hange resin bags (5 x 5 cm) were placed in the top 5 cm of the soil and left in the field for 3 month intervals. Resin bags were in the field for one year continuously (Mid June Mid Sept. 2005, Mid Sept. Mid Dec. 2005, Mid Dec. 2005 Mid March 2006, a nd Mid March Mid June 2006). Before being buried in the field, anion and cation resin bags were charged with 2M HCl and 2M NaCl, respectively. After resin bags were removed from the field, they were rinsed with DI H 2 O to remove dirt and any attached root s. Anion and cation resin bags were extracted with 50 mL of 0.5 M HCl and 0.5 M NaCl, respectively and shaken for six hours. Resin extracts were frozen and taken to the University of Florida where NO 3 NH 4 + and PO 4 3 concentrations were determined colori metrically on a continuous flow autoanalyzer (Astoria Pacific, Inc., Clackamas, Oregon, USA). To analyze differences in resin exchangeable nutrients, I averaged transect values for each plot so that plot was the statistical unit. Resin bags that were foun d on the soil surface were not included in plot means; thus, the number of resin bags per plot ranged from one to five. For total inorganic N and N:P ratios, if either the cation or anion bag was found out of the ground, the site was not included in the pl ot mean. I used repeated measures analysis with date as the within subjects factor and time since fire as the between subjects factor to analyze differences in resin exchangeable nutrients. All data were natural log transformed before analyses. One N:P rat io outlier from the June September sampling period was removed from the analysis because the N:P ratio was 232, which was nine standard deviations above the overall mean for June

PAGE 218

218 September. For all repeated measures analyses, when the assumptions of spheri city were not met, the degrees of freedom were adjusted using the Greenhouse Geisser estimated epsilon values, which is a conservative correction (Field 2009), and differences among dates and times since fire were determined with post hoc pairwise comparis ons with Bonferroni confidence interval adjustments. Monthly rainfall during the year of my study tended to be above average during the wet season and below average during the dry season (Figure B 1). Resin exchangeable NH 4 + NO 3 total inorganic N, PO 4 3 and N:P ratios tended to increase from the wet to the dry season (Table B 1). Resin exchangeable NH 4 + increased from the wet season (June Sept) to the dry wet transition period (March June) and was 19 times greater in the dry wet transition period than i n the wet season (Figure B 2 ). Resin exchangeable NO 3 was lowest during the late wet/early dry season (Sept Dec). The ratio of NH 4 + to NO 3 was always greater than one and increased 23 fold between the wet season and the dry wet transition period; thus, t otal inorganic N increased from the wet season to the dry wet transition period, mirroring the change in resin exchangeable NH 4 + over time. Resin exchangeable PO 4 3 showed the least amount of seasonal variation; the highest values, which occurred during th e dry wet transition period, were only 1.5 times greater than the lowest values, which occurred during the wet season. Resin extractable N:P ratios were greater during the dry/early wet season (Dec. June) than the wet/early dry season (June Dec.) and incre ased 4 fold from the wet season to the late dry/early wet season (Figure B 2 ). Within a season, resin exchangeable N was not significantly correlated with resin exchangeable P (Table B 2). Resin exchangeable NH 4 + total inorganic N, and N:P

PAGE 219

219 ratios were ne gatively correlated with rainfall, but there was not a significant correlation between resin exchangeable NO 3 or PO 4 3 and rainfall (Figure B 3). Resin exchangeable NH 4 + and thus total resin exchangeable N, varied seasonally in scrubby flatwoods, but alt hough N mineralization increased with water addition, resin exchangeable N was negatively correlated with rainfall. Several mechanisms could explain why resin exchangeable NH 4 + was high during the dry wet transition period (March June) when rainfall was lo w, or why resin exchangeable NH 4 + was low during the early wet season (June September) when rainfall was high. First, soil microbial biomass and activity is higher in wet than in dry soils (Tate and Terry 1980; A guilera et al. 1999; Paradelo and Barral 200 9). When rainfall, and thus soil moisture, is low, nitrification rates may be low, causing NH 4 + to accumulate in soils. This is unlikely, however, because mineralization rates, as well as nitrification rates, are water limited in surface soils, suggesting that resin exchangeable NH 4 + should be low. Second, plant nutrient uptake affects soil nutrient availability, and NH 4 + and NO 3 availability may be higher during the season when plants are not growing (Durn et al. 2009). Rainfall was lowest during the lat e dry season (March June), and NH 4 + could have accumulated on resins if plant N uptake low during this time period. This is unlikely, however, because oaks, the dominant shrubs in scrubby flatwoods, turn over their leaves from March May (Abrams and Menges 1992), so oaks should be taking up N during this time period. Thus, I hypothesize that resin exchangeable NH 4 + is negatively correlated with rainfall due to high leaching losses of N. Dissolved organic nitrogen (DON) contributes a large fraction of N lost from terrestrial ecosyste ms in runoff (Hedin et al. 1995; Schlesinger et al. 1999), particularly in unpolluted ecosystems

PAGE 220

220 (Perakis and Hedin 2002). Atmospheric N deposition is relatively low in central peninsular Florida, and the ratio of DON:DIN in scrubb y flatwood soils is greater than one, suggesting that DON makes up a large part of N losses in Florida scrub ecosystems. Losses of DON are positively related to rainfall (Perakis and Hedin 2002), and high rainfall in scrubby flatwoods may lead to high loss es of DON. Early wet season rainfall in central peninsular Florida most often occurs as afternoon thunderstorms, and even short, heavy rain events could lead to N losses because DON losses are greatest during the beginning of rainfall events (Schlesinger e t al. 1999). DON is a significant predictor of N mineralization rates in surface soils (0 5 cm) of scrubby flatwoods ( Chapter 4 ). Low concentrations of DON contribute to low N mineralization rates, which contributes to low resin exchangeable NH 4 + Prez et al. (2004) found that inorganic N availability is positively correlated with precipitation in forest soils, but the low water holding capacity of sandy scrub soils likely contributes to greater leaching losses. Total rainfall during my one year study peri od (1130 mm) was slightly lower than, but similar to, mean annual rainfall (1365 mm) at my study site, suggesting that during the average year, resin exchangeable N in scrubby flatwoods is similar to or lower than what I measured. Similar to my results, re sin exchangeable N decreased over a rainfall gradient in tropical montane forests (Schuur and Matson 2001); however, the rainfall in all sites was higher than in scrubby flatwoods, and differences in resin exchangeable N were likely related factors such as redox potential. Although N mineralization rates were positively affected by water availability (Chapter 3), resin exchangeable N was negatively correlated with rainfall. This difference in the effects of water on N availability is likely related to diff erences in

PAGE 221

221 sampling methods. Resin exchangeable N was measured in an open system, which allowed for leaching losses; whereas, N mineralization and nitrification were measured in a closed system, which did not allow for leaching losses. In the field, N can be leached from the system; whereas, in lab incubations, N is contained. In addition, rainfall was correlated with resin exchangeable N over three months, while N mineralization was determined from one week incubations. In some cases, concentrations of NH 4 + are positively correlated with soil water content (Todd et al. 2000) and soil moisture tracks rainfall in scrubby flatwoods (Weekley et al. 2007), suggesting that over short time scales, rain events that do not cause high leaching of DON may increase N m ineralization, and thus NH 4 + in scrubby flatwood soils. Resin exchangeable PO 4 3 varied seasonally in scrubby flatwoods, but was not significantly correlated with rainfall. Similarly, across a larger rainfall gradient in montane forests, resin available P was not correlated with rainfall (Schuur and Matson 2001). The ratio of resin exchangeable N:P was negatively correlated with rainfall. Although resin exchangeable PO 4 3 was not correlated with rainfall, there was a negative relationship between rainfall and total inorganic N, which drove the relationship between N:P and rainfall. The N:P acquisition activity ratio of soil enzymes is negatively related to mean annual precipitation (Sinsabaugh et al. 2008). This suggests that precipitation affects the acti vity of soil microbes in addition to affecting leaching of nutrients.

PAGE 222

222 Table B 1. Results of repeated measures analyses of resin exchangeable nutrients. Date and the date by years since fire (YSF) interaction are within subject factors, and YSF is a betwe en subject factor. All data were natural log transformed before analyses. When the assumption of sphericity was not met, the Greenhouse Geisser correction for degrees of freedom was used. F D f P NH 4 + 1 day 1 ) Date 65.26 3 <0.001 YSF 1.93 5 0.162 Date YSF 1.11 15 0.381 NO 3 1 day 1 ) Date 11.51 2.006 <0.001 YSF 6.24 5 0.004 Date YSF 1.98 10.028 0.083 Total Inorganic N 1 day 1 ) Date 32.20 1.952 <0.001 YSF 1.02 5 0.446 Date YSF 0.66 9.759 0.742 PO 4 3 de 1 day 1 ) Date 6.33 3 0.001 YSF 2.96 5 0.058 Date YSF 1.41 15 0.195 N:P Date 19.83 3 <0.001 YSF 3.92 5 0.024 Date YSF 0.97 15 0.501

PAGE 223

223 Table B 2. Partial correlation coefficients (controlling for the effects of time since fire) for resi n exchangeable total inorganic N and PO 4 3 for each sampling period and over one year. Time period R df p June September .026 82 0.409 September December .154 79 0.085 December March .109 65 0.190 March June .049 73 0.336 Total per year .1 99 87 0.031 Figure B 1. Mean monthly rainfall (1932 2005) and total monthly rainfall during my study period (June 2005 June 2006) at Archbold Biological Station by soil moisture season. Modified from Weekley et al. 2007.

PAGE 224

224 Figure B 2. Mean ( + se ) resin extractable NH 4 + (A), NO 3 (B), total inorganic N (C), PO 4 3 (D), and N:P (E) during each sampling period. Different letters

PAGE 225

225 Figure B 3. Relationship between resin extractable NH 4 + (A), NO 3 (B), total inorganic N (C), PO 4 3 (D), and N:P (E) and rainfall during each 3 month sampling period.

PAGE 226

226 APPENDIX C SCRUBBY FLATWOODS SO IL ANALYSIS Abrahamson et al. (1984) provides one of the few measurements of multiple soil elements in scrubby flatwood soils. To my knowledge, however, a complete soil ana lysis has not been conducted on scrubby flatwood soils. A sub sample of ground soils from each control plot (from the fertilization experiment; Chapter 4) for each time since fire (both 0 10 cm and 10 20 cm depths) was sent to the ALS Laboratory Group ( www .alsglobal.com ) for analysis Most elements were similar among times after fire and depth s (Table C 1) Several elements were below detectable limits in scrubby flatwoods soils. A rsenic (As) was below detec tion limits nine a nd 21 years after fire, but in recently burned sites arsenic was high in both surface and deep soils This suggests that fire may increase soil arsenic, which could affect soil microorganisms; however, an increase in phosphate after fire (Chapter 1 ) may reduce the toxic effects of arsenic compounds (DaCosta 1972). Lead (Pb) and strontium (Sr) were also highest in surface soils of recently burned sites. Further sampling is needed to determine the effects of fire on these soil elements.

PAGE 227

227 T able C 1 Mean (se) concentrations of elements in scrubby flatwoods soils at different times after fire and soil depths. Elements were measured in parts per million (ppm) or percents (%). Time after fire Ag (ppm) As (ppm) Ba (ppm) Ce (ppm) Cr (ppm) Cs (p pm) Ga (ppm) Ge (ppm) Hf (ppm) Li (ppm) Nb (ppm) Pb (ppm) Rb (ppm) Re (ppm) 1 year 0 10 cm 0.013 (0.003) 1.73 (1.14) <10 1.18 (0.11) 122 (33.2) <0.05 0.16 (0) <0.05 0.17 (0.03) 1.27 (0.07) 0.43 (0.03) 2.80 (1.05) 0.23 (0.03) <0.002 10 20 cm 0.01 (0) 1.00 (0.56) <10 0.91 (0.04) 211 (70.8) <0.05 0.14 (0.01) <0.05 0.10 (0) 1.20 (0.10) 0.43 (0.03) 1.23 (0.28) 0.10 (0) 0.002 9 years 0 10 cm 0.01 (0) <0.2 <10 1.38 (0.29) 142 (51.4) <0.05 0.17 (0.01) <0.05 0.15 (0.03) 1.13 (0.0 3) 0.42 (0.06) 1.60 (0.10) 0.20 (0) 0.0025 (0.001 ) 10 20 cm 0.02 (0.01) <0.2 <10 1.25 (0.23) 156 (22.3) <0.05 0.20 (0.01) <0.05 0.17 (0.03) 1.23 (0.14) 0.60 (0.06) 1.20 (0.10) 0.17 (0.03) 0.002 (0) 21 years 0 10 cm 0.01 (0) <0.2 <10 1.02 (0.04) 162 (31.9) <0.05 0.21 (0.003) <0.05 0.13 (0.03) 1.33 (0.07) 0.53 (0.03) 1.77 (0.03) 0.20 (0) 0.002 10 20 cm 0.01 (0) <0.2 <10 0.97 (0.07) 171 (75.9) <0.05 0.17 (0.12) <0.05 0.13 (0.03) 1.33 (0.09) 0.57 (0.03) 0.93 (0.06) 0.17 (0.06) 0.002

PAGE 228

228 Tab le C 1 continued. Time after fire Sb (ppm) Sc (ppm) Se (ppm) Sn (ppm) Sr (ppm) Ta (ppm) Te (ppm) Th (ppm) Ti (%) Tl (ppm) U (ppm) V (ppm) W (ppm) Y (ppm) Zr (ppm) 1 year 0 10 cm 0.16 (0.03) 0.40 (0) 1.67 (0.33) 0.57 (0.12) 2.43 (0.79) <0. 05 0.10 0.27 (0.03) 0.019 (0.001) <0.02 0.13 (0.03) 3.67 (0.33) 0.53 (0.12) 0.30 (0) 5.10 (0.20) 10 20 cm 0.10 (0.02) 0.27 (0.09) 1.00 (0) 0.40 (0.15) 0.80 (0.21) <0.05 0.20 (0.12) 0.23 (0.03) 0.020 (0.001) <0.02 0.10 (0) 4.33 (0.33) 0.57 (0.13) 0.23 (0.0 3) 4.53 (0.14) 9 years 0 10 cm 0.11 (0.003) 0.35 (0.08) 1.00 (0) 0.45 (0.03) 1.15 (0.18) <0.05 0.11 0.27 (0.03) 0.022 (0.003) <0.02 0.13 (0.03) 4.00 (0.58) 0.37 (0.07) 0.27 (0.03) 5.73 (0.86) 10 20 cm 0.12 (0.03) 0.37 (0.03) 1.67 (0.33) 0 .57 (0.13) 0.70 (0) <0.05 0.11 (0.02) 0.23 (0.03) 0.030 (0.006) <0.02 0.10 (0) 5.00 (0.58) 0.53 (0.18) 0.33 (0.03) 2.85 (1.59) 21 years 0 10 cm 0.12 (0.01) 0.40 (0) 1.67 (0.33) 0.50 (0.06) 1.30 (0.21) <0.05 0.15 0.23 (0.03 0.022 (0.002) <0 .02 0.10 (0) 4.33 (0.33) 0.60 (0.10) 0.30 (0) 4.57 (0.17) 10 20 cm 0.10 (0.03) 0.30 (0.06) 1.33 (0.33) 0.43 (0.14) 0.70 (0.15) <0.05 0.36 (0.16) 0.20 (0) 0.025 (0.003) <0.02 0.10 (0) 4.00 (0.58) 0.50 (0.15) 0.27 (0.03) 4.63 (0.35)

PAGE 229

229 APPENDIX D DIFFERENC ES IN ABOVEGROUND BI OMASS AND STEM TURNO VER WITH TIME AFTER FIRE IN SCRUBB Y FLATWOODS To measure the effects of nutrient addition on total aboveground biomass (Chapter 5), I used allometric equations to estimate shrub biomass. I used measurements of shrub length (maximum crown length) and width (minimum crown length) to calculate the area of each marked individual. For oaks and lyonias, I summed stem biomasses to determine the biomass of each individual. I then calculated the total measured biomass and tota l area for each species in each plot. Using my estimates of percent cover of each species, I scaled up biomass of each species to the entire plot. In many cases, the biomass I measured was greater than total plot biomass determined from scaling up, suggest ing that I underestimated shrub cover. In these instances, I used the biomass of the plants I measured rather than the biomass scaled to my estimates of shrub percent cover. Differences in pre fertilization biomass among times after fire were determined wit h a Kruskal Wallis test. Biomass of dominant shrubs was significantly higher eight and 20 years after fire than six weeks after 2 = 60.01, df = 2, p < 0.001). There was no difference in total shrub biomass 8 and 20 years after fire, suggesting that biomass does not increase over this time period. Oak aboveground biomass increased linearly over time after fire (Seiler et al. 2009). Several mechanisms could explain the lack of change in total shrub biomass eight to 20 years after fire. First, there is a negative relationship between abundance and total plant mass (Allen et al. 2008); however, percent cover of shrubs overall an d the number of stems within clumps does not vary between eight and 20 years since fire (J. Schafer, unpublished data). Second, as stem diameter increases, resources are allocated to the root system to increase stability (Drexhage et al. 1999).

PAGE 230

230 Stem diamet er did not increase from eight to 20 years since fire, and root productivity does not differ between sites 8 and 20 years since fire. I hypothesize that shrub stems experience die back due to water (Saha et al. 2008) or nutrient limitation. Although the lo nger unburned sites have received similar amounts of precipitation over the last eight to nine years, the 20 to 21 years since fire sites were located at the southern end of Archbold Biological Station, where scrubby flatwoods occur 2 m higher in elevation than further North, where the eight years since fire sites were located (Abrahamson et al. 1984). Thus, the 20 to 21 years since fire sites are located further above the water table, suggesting that during times of drought, species such as Q. chapmanii an d Q. geminata which take up water from 40 200 cm (Saha et al. 2008), may experience greater water stress. Percent dieback (Au and Tardif 2007) and the ratio of biomass loss to the gross production of aboveground biomass (Kawamura and Takeda 2008) increase with stem age; however, the mean and maximum life span of Q. inopina stems is 4 y ears and 9 years, respectively (Johnson and Abrahamson 2002), so it is unlikely that stem age alone causes the observed patterns in stem size. Previous research found that th e height of Q. inopina stems did not vary over a nine year period in a long unburned site (Johnson and Abrahamson 2002). Aboveground biomass of Q. inopina did not vary from 2 34 years after fire in Florida rosemary scrub (Johnson et al. 1986); whereas, abo veground biomass of oak shoots increased, while aboveground biomass of other shrub species, including Lyonia species, did not vary, from 3 25 years since fire in scrubby flatwoods (Saha et al. in review ). Biomass of shrubs in Florida Keys pine forests did not increase from 12 to 30 years after fire (Sah et al. 2004). Stimulation of

PAGE 231

231 biomass increment of oaks in coastal scrub (increase in biomass per year) decreased over time (Seiler et al. 2009). For all focal Quercus inopina individuals, I followed stem sur vival and recruitment. I used Kruskall Wallis tests to analyze differences in mean stem percent survival and percent change in stem number among times since fire. Post hoc differences among times since fire were determined with Bonferron i adjusted signific ance values 2 = 18.31, df = 2, p < 0.001) and 2 = 25.76, df = 2, p < 0.001) varied with time since fire (Figure D 1). Mean percent survival was 7% and 13% lower 20 years and 6 weeks after fire, respectivel y, than 8 years after fire. Mean percent change in stem number was negative six weeks after fire and positive both 8 and 20 years after fire. Johnson and Abrahamson (2002) found that 80 to 90% of Q. inopina stems died over a 10 year period, but new stems w ere recruited so that the number of stems of Q. inopina individuals remained the same or increased over time. Survival of Quercus inopina stems was lower 1 and 20 years after fire than 8 years since fire, but individuals in sites 20 years since fire recrui ted new stems so that percent change in stem number was similar 8 and 20 years since fire.

PAGE 232

232 Figure D 1. Boxplots of percent survival (A) and percent change in stem number (B) of Quercus inopina plants 6 weeks, 8 yea rs, and 20 years since fire. The lower and upper bars of the boxplot represent the 25 th and 75 th percentiles, respectively; the solid middle bar represents the median and the dotted bar nd smallest values that are not outliers. The circles indicate outliers. Different lowercase letters below the boxplots indicate significant differences among times since fire.

PAGE 233

233 LIST OF REFERENCES Abrahamson WG (1995) Habitat distribution and competitive neighborhoods of two Florida palmettos. Bull Torrey Bot Club 122:1 14 Abrahamson WG (1999) Episodic reproduction in two fire prone palms, Serenoa repens and Sabal etonia (Palmae). Ecology 80:100 115. Abrahamson W G (2007) Leaf traits and leaf life spans of two x eric adapted palmettos. Am J Bot 94:1297 1308 Abrahamson WG, Johnson AF, Layne JN, Peroni PA (1984) Vegetation of the Archbold Biological Station, Florida: An example of the southern Lake Wales Ridge. Florida Sci 47:209 250 Abrahamson WG, Hartnett DC (1990) Pine flatwoods and dry prairies. In : Myers RL, Ewel JJ (eds) Ecosystems of Florida. University of Central Florida Press, Orlando pp 103 149 Abrahamson WG, Layne JN (2002a) Post fire recovery of acorn production by four oak species in southern ridge sandhill associations in South centr al Florida. Am J Bot 89:119 123 Abrahamson WG, Layne JN (2002b) Relation of ramet size to acorn production in five oak species of xeric upland habitats in South central Florida. Am J Bot 89:124 131 Abrahamson WG, Layn e JN (2003) Long term patterns of acorn production for five oak species in xeric Florid a uplands. Ecology 84:2476 2492 Abrahamson WG, Hunter MD, Melika G, Price PW (2003) Cynipid gall wasp communities correlate with oak chemistry. J Chem Ecol 29:209 223 Ab rams MD and ES Menges (1992) Leaf ageing and plateau effects on seasonal pressure volume relationships in three sclerophyllous Quercus species in south eastern USA. Functional Ecol 6:353 360. Aciego Pietri JC, Brookes PC (2009) Substrate inputs and pH as f actors controlling microbial biomass, activity and community structure i n an arable soil. Soil Biol Biochem 41:1396 1405 Adams MA, At tiwill PM (1991) Nutrient balance in forests of northern Tasmania. 2. Alteration of nutrient availability and soil water ch emistry as a result of logging, slash burning and fertilizer application. For est Ecol Manag 44:115 131 Adams MA, Iser J, Keleher AD, Cheal DC (1994) Nitrogen and phosphorus availability and the role of fire in heathlands. Aust J Bot 42:269 281

PAGE 234

234 Aerts R (200 9) Nitrogen supply effects on leaf dynamics and nutrient input into the soil of plant species in a sub arctic tundra e cosystem. Polar Biol 32:207 214 Agnihotri VP, Vaar taja O (1969) Stimulation of Waitea circinata by root exudates of Pinus cembroides Can J Microbiol 15:1319 1323 Aguilera LE, Gutirrez JR, Meserve PL (1999) Variation in soil micro organisms and nutrients underneath and outside the canopy of Adesmia bedwellii (Papilionaceae) shrubs in arid coastal Chile following drought and above average ra infall. J Arid Environ 42:61 70 Aikawa S, Hori Y (2006) Effect of a multi stemmed growth form on matter production of an understory shrub, Stephanandra incisa Plant Spec Biol 21:31 39 Ainsworth EA, Tranel PJ, Drake BG, Long SP (2003) The clonal structure of Quercus geminata revealed by conserved micros atellite loci. Molecular Ecol 12:527 532 Allen AP, Pockman WT, Restrepo C, Milne BT (2008) Allometry, growth and population regulation of the desert shrub Larrea tridentata Funct Ecol 22:197 204 Anderson RC Menges ES (1997) Effects of fire on sandhill herbs: Nutrients, mycorrhizae, and biomass allocation. Am J Bot 84:938 948 An ten NPR, Hirose T (1998) Biomass allocation and light partitioning among dominant and subordinate individuals in Xanthium canadense stan ds. Ann Bot 82:665 673 Anten NPR, Miyazawa K, Hikosaka K, Nagashima H, Hirose T (1998) Leaf nitrogen distribution in relation to leaf age and photon flux density in dominant and subordinate plants in dense stands of a dicotyledonous herb. Oecologia 113 :314 324 Arocena JM, Opio C (2003) Prescribed fire induced changes in properties of sub boreal forest soils. Geoderma 113:1 16 Au R, Tardiff JC (2007) Allometric relationships and dendroecology of the dwarf shrub Dryas integrifolia near Churchil l, subarcti c Manitoba. Can J Bot any 85:585 597 A ustin AT, Vivanco L (2006) Plant litter decomposition in a semi arid ecosystem controlled by photodegradation. Nature 442:555 558 Bada D, Mart C (2003) Plant ash and heat intensity effects on chemical and physical pro perties of two contrasting soils. Arid Land Res Manag 17:23 41 Be ckage B, Platt WJ, Slocum MG, Panko B (2003) Influence of the El Nino southern oscillation on fire regimes in the Florida E verglades. Ecology 84:3124 3130

PAGE 235

235 Bennett LT, Judd TS, Adams MA (2002) Growth and nutrient content of perennial grasslands following burning in semi arid, sub tropical Australia. Plant Ecol 164:185 199 Bergersen FJ, Peoples MB, Turner GL (1988) Isotopic discriminations during the accumulation of nitrogen by soybeans. Aust J Plant Physiol 15:407 420 Be rry DM, Menges ES (1997) Post fire changes in resource limitation of Florida scrub plants. Proceedings Fire Effects on Rare and Endangered Species and Habitat Conference : 197 201 Blagodatskaya EV, Blagodatsky SA, Anderson T H, Kuzyakov Y (2009) Contrasting effects of glucose, living roots and maize straw on microbial growth kinetics and substrate availability in soil. Eur J Soil Sci 60:186 197 Blair, BC (2005) Fire effects on the spatial patterns of soil resources in a Nicarag uan wet tropical forest. J Trop Ecol 21:435 444 Blair BC, Perfecto I (2008) Root proliferation and nutrient limitations in a Nicaraguan rain forest. Caribb J Sci 44:36 42 Bloom RG, Mallik AU (2006) Relationships between ericaceous vegetation and soil nutr ient status in a post fire Kalmia angustifolia black spruce chronos equence. Plant Soil 289:211 226 Boerner REJ, Huang J, Hart SC (2009) Impacts of fire and fire surrogate treatments on forest soil properties: a meta analytical approach. Ecol Appl 19:338 35 8 Bond Lamberty B, Gower ST, Wang C, Cyr P, Veldhuis H (2006) Nitrogen dynamics of a boreal black spruce wildfire chr onosequence. Biogeochem 81:1 16 Bonfil C, Corts P, Espelta JP, Retana J (2004) The role of disturbance in the co existence of the evergree n Quercus ilex and the deciduous Quercus cerrioides J Veg Sc i 15:423 430 Bormann BT, Sid le RC (1990) Changes in productivity and distribution of nutrients in a chronosequence at Glacier Bay National Park, Alaska. J Ecol 78:561 578 Bosn er SP, Aarssen LW (1 994) Plastic allometry in young sugar maple ( Acer saccharum ): adaptive respons es to light availability. Am J Bot 81:400 406 Bosn er SP, Aarssen LW (2001) Allometry and plasticity of meristem allocation throughout development in Arabidopsis thaliana J Ecol 89:72 79 Bown HE, Watt MS Clinton PW, Mason EG, Whitehead D (2009) The influence of N and P supply and genotype on carbon flux and partitioning in potted Pinus radiata pla nts. Tree Physiol 29:857 868

PAGE 236

236 Bradstock RA, Williams JE, Gill AM (2001) Flammable Aus tralia: the fire regimes and biodiversity of a continent. Cambridge University Press, New York Bret Harte MS, Garca EA, Sacr VA Whorley JR, Wagner JL, Lipp ert SC, Chapin III FS (2004) Plant and soil responses to neighbour removal and fertilization in Alaskan tussock tundra. J Ecol 92:635 647 Britton AJ, Helliwell RC, Fisher JM, Gibbs S (2008) Interactive effects of nitrogen deposition and fire on plant and soil chemistry in an alp ine heathland. Environ Pollut 156:409 416 Brookes PC, Landman A, P ruden G, Jenkinson DS (1985) Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biol Biochem 17:837 842 Brown ALP, Day FP, Hungate BA, Drake BG, Hinkle RC (2007) Root biom ass and nutrient dynamics in a scrub oak ecosystem under the influence of elevated atmospheric CO 2 Plant Soil 292:219 232 Brown ALP, Garland JL, Day FP (2009) Physiological profiling of soil microbial communities in a Florida scrub oak ecosystem: spatial distribution and nutrient limitation. Microb Ecol 57:14 24 Brown RB, Stone E L, Carlisle VW (1990) Soils. In: Myers RL, Ewel JJ (eds) Ecosystems of Florida. University of Central Florida Press, Orlando pp. 35 69 Caff rey JM, Murrell MC, Wigand C, McKinney R (2007) Effect of nutrient loading on biogeochemical and microbial processes in a New England sa lt marsh. Biogeochem 82:251 264 Carrei ra JA, Niell FX, Lajtha K (1994) Soil nitrogen availability and nitrification in Mediterranean shrublands of varying fire history and succession al stage. Biogeochem 26:189 209 Car ter MC, Foster CD (2004) Prescribed burning and productivity in southern pine forests: a review. For est Ecol Manag 191:93 109 Cas sman KG, Munns DN (1980) Nitrogen mineralization as affected by soil m oisture, temperature, and depth Soil Sci Soc Am J 44:1233 1237 Cavender Bares J, Kitajima K, Bazzaz FA (2004) Multiple trait associations in relation to habitat differentiation among 17 Floridian oak species. E col Monogr 74:635 662 Cech PG, Kuster T, Ed wa rds PJ, Venterink HO (2008) Effects of herbivory, fire and N 2 fixation on nutrient limitation in a humid African savanna. Ecosystems 11:991 1004

PAGE 237

237 Certini G (2005) Effects of fire on properties of forest soils: a review. Oecologia 143:1 10 Chang SX, Handley LL (2000) Site history affects soil and plant 15 N natural abundances 15 N) in forests of Northern Vancouver Island, British Columbia. Funct Ecol 14:273 280 Chapin III FS, Matson PA, Mooney HA (2002) Principles of Terrestrial Ecosystem Ecology Sp ringer Ve rlag, New York C hapin III FS, Bloom AJ, Field CB, Waring RH (1987) Plant responses to multiple environmental factors. Bioscience 37:49 57 Chappell MA, Evangelou VP (2000) Influence of added K+ on ammonium selectivity/mobility by soils with vermiculitic b ehavior. Soil Sci 11:858 868 Chen CR, Condron LM, Davis MR, Sherlock RR (2004) Effects of plant species on microbial biomass phosphorus and phosphatase activity in a range of grassland soils. Biol Fert Soils 40:313 322 Christensen NL (1985) Shrubland fire regimes and their evolutionary consequences. In: Pickett STA, White PS (eds) The ecology of natural disturbance and patch dynamics. Academic Press, New York, pp 85 100 Christensen NL, Muller CH (1975) Effects of fire on factors controlling plant growth in Adenostoma chaparral. Ecol Monogr 45:29 55 Cleary MB, Pendall E, Ewers BE (2008) Testing sagebrush allometric relationships across three fire chronoseq uences in Wyoming, USA. J Arid Environ 72:285 301 Clemente AS, Rego FC, Correia OA (2005) Growth, water relations and photosynthesis of seedlings and resprouts after fire. Acta Oecologic a 27:233 243 the microbial biomass? Biogeochem 85:235 252 Cook GD (1994) The fate of nutrients during fires in a tropical savanna. Aust J Ecol 19:359 365 Co omes DA, Grubb PJ (1998) A comparison of 12 tree species of Amazonian caatinga using growth rates in gaps and understorey, and allometric relationships Funct Ecol 12:426 435 Cornel issen T, Stiling P (2006) Responses of different herbivore guilds to nutrient addition and natural enemy exclusion. Ecoscience 13:66 74 Cornelissen T, Stiling P (2008) Clumped distribution of oak leaf miners between and within plants. Ba sic Appl Ecol 9:67 77

PAGE 238

238 Covington WW, Sackett SS (1992) Soil mineral nitrogen changes following prescribed burning in ponderosa pine. For est Ecol Manag 54:175 191 Craine JM Morrow C, Stock WD (2008) Nutrient concentration ratios and co limitation in South African grasslan ds New Phytol 179:829 836 Cruz A, Prez B, Quintana J, Moreno J (2002) Resprouting in the Mediterranean type shrub Erica australis affected by soil resource availability. J Veg Sci 13:641 650 Cuevas E, Medina E (1988) Nutrient dynamics within Amazonian for ests.2.Fine root growth, nutrient availability and leaf litter decomposition. Oecologia 76:222 235 determination of phosphate in w ater and soil. J Environ Qual 30:2206 2209 nio CM, Tunison JT, Loh RK (2000) Variation in the impact of exotic grasses on native plant composition in relation to fire across an elevational gradient in Haw aii. Austral Ecol 25:507 522 Mack MC (2006) Nutrient limitation in a fire derived nitrogen rich Hawaiian g rassland. Biotropica 38:458 467 Daily G C Sodergvist T, Aniyar S, Arrow K, Dasgupta P, Ehrl ich PR, Folke C, Jansson A, Ja nsson BO, Kautsky N, Levin S, Lubche nco J, Maler KG, Simpson D, Starret D Tilman D, Walker B (2000) The val ue of nature and the nature of values. Science 289:395 396 Darby FA, Turner RE (2008) Below and aboveground biomass of Spartina alterniflora : response to nutrient addition in a Louisiana salt marsh. Estuarie s Coasts: J CERF 31:326 334 Davidson EA, Reis d e Carvalho CJ, Vieira ICG, Figueiredo RO, Moutinho P, Ishi da FY, Primo dos Santos MT, Guerrero JB, Kalif K, Sab RT (2004) Nitrogen and phosphorus limitation of biomass growth in a tropical secondary forest. Ecol Appl 14:S150 S163 Davies SJ, Myerscough P J (1991) Post fire demography of the Wet mallee Eucalyptus leuhmanniana F.Muell. (Myrtaceae). Aust J Bot 39:459 66 Dawson TE, Mambelli S, Plamboeck AH, Templer PH, Tu KP (2002) Stable isotopes in plant ecology. Ann Rev Ecol Syst 33:507 509 DaCosta EWB (197 2) Variation in the toxicity of arsenic compounds to microorganisms and the suppression of the inhibitory effects by phosphate. Appl Microbiol 23:46 53

PAGE 239

239 Debano LF, Conrad CE (1978) The effect of fire on nutrients in a chaparral ecosystem. Ecology 59:489 497 DeLuca TH, Sala A (2006) Frequent fire alters nitrogen transformations in ponderosa pine stands of the inland northwest. Ecology 87:2511 2522 DeLuca TH, Nilsson M C, Zackrisson O (2002) Nitrogen mineralization and phenol accumulation along a fire chronose quence in northern Sweden. Oecologia 133:206 214 De Lucia EH, Callaway RM, Tho mas EM, Schlesinger WH (1997) Mechanisms of phosphorus acquisition for ponderosa pine seedlings under high CO2 and temperatu re. Ann Bot 79:111 120 DeSouza J, Silka PA, Davis SD (1 986) Comparative physiology of burned and unburned Rhus laurina after chaparr al wildfire. Oecologia 71:63 68 DeWilde L, Chapin III FS (2006) Human impacts on the fire regime of interior Alaska: interactions among fuels, ignition sources, and fire supp ress ion. Ecosystems 9:1342 1353 den Dubbeld en KC, Knops JMH (1993) The effect of competition and slope inclination on aboveground biomass allocation of understorey ferns in subtropic al forest. Oikos 67:285 290 Diehl P Mazzarino MJ, Fontenla S ( 2008 ) Plant lim iting nutrients in Andean Patagonian woody species: effects of interannual rainfall variation, soil fertility and myc orrhizal infection. Forest Ecol Manag 255:2973 2980 Drexhage M, Chauvire M, Colin F, Nielsen CNN (1999) Development of structural root ar chitecture and allometry of Quercus petraea Can J For est Res 29:600 608 Dunn PH, Barro SC, Poth M (1985) Soil moisture affects survival of microorganisms in heated chaparral soil. Soil Biol Biochem 17:143 148 Durn J, Rodrguez A, Fernndez Palacios JM, G allardo A (2008) Changes in soil N and P availability in a Pinus canariensis fire chronosequence. For est Ecol Manag 256:384 387 Durn J, Rodrguez A, Fernndez Palacios JM, Gallardo A (2009) Changes in net N mineralization rates and soil N and P pools in a pine forest wildfire chronosequenc e. Biol Fert Soils 45:781 788 El Omari B, Aranda X, Verdaguer D, Pascual G, Fleck I (2003) Resource remobilization in Quercus ilex L. resprouts. Plant Soil 252:349 357 Enquist BJ, Niklas KJ (2002) Global allocation rules for patterns of biomass partitioning in see d plants. Science 295:1517 1520

PAGE 240

240 Enquist BJ, Niklas KJ (2001) Invariant scaling relations across tree dominated communities. Nature 410:655 660 Eom A H, Hartnett DC, Wilson GWT, Figge DAH (1999) The effect of fire, mowing and fertilizer amendment on arbuscular mycorrhizas in tallgrass prairie. Am Midl Nat 142:55 70 Evans RD (2001) Physiological mechanisms influencing plant nitrogen isotope composition. Trends Plant Sci 6:121 126 Evans RD, Bloom AJ, Sukrapanna SS, Eh leringer JR (1996) Nitrogen isotope composition of tomato ( Lycopersicon esculentum Mill. Cv. T 5) grown under ammonium or nitrate nutrition. Plant Cell Environ 19:1317 1323 Evans CA, Miller EK, Friedland AJ (1998) Nitrogen mineralization associated with bi rch and fir under different soil moisture regi mes. Can J For est Res 28:1890 1898 Ewel, J, Berish C, Brown B, Price N, Raich J (1981) Slash and burn impacts on a Costa Rican wet forest site. Ecology 62:816 829 Farnswor th EJ, Ellison AM (1996) Sun shade adap tability of the red mangrove, Rhizophora mangle (Rhizophoraceae): changes through ontogeny at several levels of b iological organization. Am J Bot 83:1131 1143 Fenn ME, P oth MA, Dunn PH, Barro SC (1993) Microbial N and biomass, respiration and N mineralizat ion in soils beneath two chaparral species along a fire induced age gradien t. Soil Biol Biochem 25:457 466 Ferran A, Delitti W, Vallejo VR (2005) Effects of fire recurrence in Quercus coccifera L. shrublands of the Valencia region (Spain): II. plant and so il nutrients. Plant Ecol 177:71 83 Field A (2009) Discovering statistics using SPSS, 3rd edn. SAGE Fisher JB, Jayachandran K (1999) Root structure and arbuscular mycorrhizal colonization of the palm Serenoa repens under field conditions. Plant Soil 217:229 241 Flannigan MD, Amiro BD, Logan KA, Stocks BJ, Wotton BM (2005) Forest fires and climate change in the 21 st century. Mitigation Adaptation Strategi es Global Change 11:847 859 Fox TR, Comerford NB (1990) Low molecular weight organic acids in selected for ests of the Southea stern USA. Soil Sci Soc Am J 54:1139 1144 Franco Vizcano E, Sosa Ramirez J (1997) Soil properties and nutrient relations in burned and unburned Mediterranean climate shrublands of Baja Cal ifornia, Mexico. Acta Oecol 18:503 517

PAGE 241

241 Frank DA (2008) Ungulate and topographic control of nitrogen: phosphorus stoichiometry in a temperate grassland; soils, plants and minerali zation rates. Oikos 117:591 601 Frank DA, Evans RD (1997) Effects of native grazers on grassland N cycling in Yellowstone Nati onal park. Ecology 78:2238 2248 Gholz HL, Fisher RF (1982) Organic matter production and distribution in slash pine ( Pinus elliottii ) pl antations. Ecology 63:1827 1839 Giardina CP, Sanford Jr. RL, Dckersmith IC (2000) Changes in soil phosphorus and nitrog en during slash and burn clearing of a dry tropical forest. Soil Sci Soc Am J 64:399 405 Gilliam FS (1988) Interactions of fire with nutrients in the herbaceous layer of a nutrient poor Coastal Plant forest. Bull Torrey Bot Club 115:265 271 Gillon D, Rapp M (1989) Nutrient losses during a winter low intensity prescribed fire in a Mediterranean forest. Plant Soil 120:69 77 Gimeno Garca E Andreu V, Rubio JL (2000) Changes in organic matter, nitrogen, phosphorus and cations in soil as a result of fire and wa ter erosion in a Medit erranean landscape. European J Soil Sci 51:201 210 Goforth BR, Graham RC, Hubbert KR, Zanner CW, Minnich RA (2005) Spatial distribution and properties of ash and thermally altered soils after high severity forest fire, southern Califo rnia. Int J Wildland Fire 14:353 354 Gkkaya K, Hurd TM, Raynal DJ (2006) Symbiont nitrogenase, alder growth, and soil nitrate response to phosphorus addition in alder ( Alnus incana ssp. rugosa ) wetlands of the Adirondack Mountains, New York State, USA. En v iron Exp Bot 55:97 109 Grayston SJ Vaughan D, Jones D (1996) Rhizosphere carbon flow in trees, in comparison with annual plants: the importance of root exudation and its impact on microbial activity and nutrient availabilit y. Appl Soil Ecol 5:29 56 Grog an P, Burns TD, Chapin III FS (2000) Fire effects on ecosystem nitrogen cycling in a Californian bishop pine forest. Oecologia 122:537 544 Guerin DN (1993) Oak dome clonal structure and fire ecology in a Florida longleaf pine dominated community. Bull Torr ey Bot Club 120:107 114 Gsewell S (2004) N:P ratios in terrestrial plants: variation and functional significance. New Phytol 164:243 266 Ha ll SA, Burke IC, Hobbs NT (2006) Litter and dead wood dynamics in ponderosa pine forests along a 160 year chronos equ ence. Ecol Appl 16:2344 2355

PAGE 242

242 Hamm an ST, Burke IC, Knapp EE (2008) Soil nutrients and microbial activity after early and late season prescribed burns in a Sierra Nevada mixed conifer forest. For est Ecol Manag 256:367 374 Han Q, Kawasaki T, Nakano T, Chiba Y (2008) Leaf age effects on seasonal variability in photosynthetic parameters and its relationships with leaf mass per area and leaf nitrogen concentration within a Pinus densiflora crown. Tree Physiol 28:551 558 Hatten J, Zabowski D Scherer G, Dolan E (2 005) A comparison of soil properties after contemporary wildf ire and fire suppression. For est Eco l Manag 220:227 241 Hawkes CV (2003) Nitrogen cycling mediated by biological soil crusts and arbuscular mycorrhizal fungi. Ecology 84:1553 1562 Hawkes CV, Flec htner VR (2002) Biological soil crusts in a xeric Florida shrubland: composition, abundance, and spatial heterogeneity of crusts with different disturbance histories. Microb Ecol 43:1 12 Hedin LO, Armesto JJ, Johnson AH (1995) Patterns of nutrient loss fro m unpolluted, old growth temperate forests: Evaluation of biogeochem ical theory. Ecology 76:493 509 Henry HAL, Aarssen LW (1999) The interpretation of stem diameter height allometry in trees: biomechanical constraints, neighbor effects, or biased regre ssio ns? Ecol Lett 2:89 97 Herb ert DA, Fownes JH (1995) Phosphorus limitation of a forest leaf area and net primary production on a highly weathered soil. Biogeochem 29: 223 235 Hernndez DL, Hobbie SE (2008) Effects of fire frequency on oak litter decompositio n and nitrogen dynamics. Oecologia 158:535 543 Hikosaka K, Terashima I, Katoh S (1994) Effects of leaf age, nitrogen nutrition and photon flux density on the distribution of nitrogen among leaves of a vine ( Ipomoea tricolor Cav.) grown horizontally to avou d mutual shading of leaves. Oecologia 97:451 457 Hobbie EA, Colpaert JV (2003) Nitrogen availability and colonization by mycorrhizal fungi correlate with nitrog en isotope patterns in plants. New Phytol 157:115 126 Hobbs NT, Schimel DS (1984) Fire effects o n nitrogen mineralization and fixation in mountain shrub and grassland communities J Range Manage 37:402 405 Hgberg P (1997) Tansley Review No. 95. 15 N natural abundance in soil plant systems. New Phytol 137:179 203

PAGE 243

243 Hol brook NM, Putz FE (1989) Influence of neighbors on tree form: effects of lateral shade and prevention of sway on the allometry of Liquidambar styraciflua (Sweet gum). Am J Bot 76:1740 1749 Horwath WR, Paul EA (1994) Microbial biomass. In: Weaver RW, Angle S, Bottomley P, Bezdicek D, Smith S Tabatabai A, Wollum A (eds) Methods of soil analysis, part 2. Microbiological and Biochemical properties. Soil Science Society of America, Inc., Madison pp 753 773 Hsu PH, Jackson ML (1960) Inorganic phosphate transformation by chemical weathering in soi ls as influenced by pH. Soil Sci 90:16 24 Huenneke LF, Clason D, Muldavin E (2001) Spatial heterogeneity in Chihuahuan desert vegetation: implications for sampling methods in semi arid ecosystems. J Arid Environ 47:257 270 Hungate BA, Dijkstra P, Johnson D W, Hinkle CR, Drake BG (1999) Elevated CO 2 increased nitrogen fixation and decreases soil nitrogen mineralization in Florida scrub oak. Global Change Biol 5:781 789 Hungate BA, Hart SC, Selmants PC, Boyle SI, Gehring CA (2007) Soil responses to management, increased precipitation, and added nitrogen in ponderosa pine forests. Ecol Appl 17:1352 1365 Iversen CM, Norby RJ (2008) Nitrogen limitation in a sweetgum plantation: implications for carbon allocation and stor age. Can J For Res 38:1021 1032 Jgerbra nd AK, Alatalo JM, Chrimes D, Molau U (2009) Plant community responses to 5 years of simulated climate change in meadow and heath ecosystems at a subarctic alp ine site. Oecologia 161:601 610 Jaggi RC, Aulakh MS, Sharma R (2005) Impacts of elemental S applie d under various temperature and moisture regimes on pH and available P in acidic, neutral and alkaline so ils. Biol Fert Soils 41:52 58 James BR, Riha SJ (1986) pH buffering in forest soil organic horizons: relevance to acid precipitation. J Environ Qual 15 :229 234 Jenny H (1941) Factors of soil f ormation. McGraw Hill, New York Jensen M, Michelsen A, Gashaw M (2001) Responses in plant, soil inorganic and microbial nutrient pools to experimental fire, ash and biomass addition in a woodland savanna. Oecologia 128:85 93 Johnson AF, Abrahamson WG, McCrea KD (1986) Comparison of biomass recovery after fire of a seeder ( Ceratiola ericoides ) and a sprouter ( Quercus inopina ) species from South central Florida. Am Midl Nat 116:423 428

PAGE 244

244 Johnson AF, Abrahamson WG (2002) Stem turnover in the rhizomatous scrub oak, Quercus inopina from South central Florida. Am Midl Nat 147:237 246 Johnson DW, Hungate BA, Dijkstra P, Hymus G, Drake B (2001) Effects of elevated carbon dioxide on soils in a Florida scrub oak ecosystem. J En viron Qual 30:501 507 Johns on DW, Hungate BA, Dijkstra P, Hymus G, Hinkle CR, Stiling P, Drake BG (2003) The effects of elevated CO 2 on nutrient distribution in a fire adapted scrub oak forest. Ecol Appl 13:1388 1399 Jonasson S, Michelsen A, Schmidt IK, Ni elsen EV, Callaghan TV (1996) Microbial biomass C, N, and P in two Arctic soils and responses to addition of NPK fertilizer and sugar: implications for plant nutrien t uptake. Oecologia 106:507 515 Jurinak JJ, Dudley LM, Allen MF (1986) The role of calcium oxalate in the availability of phosphorus in soils of semiarid regions a thermodynamic study. Soil Sci 142:255 261 Kalisz PJ, Stone EL (1984) The longleaf pine islands of the Ocala National Forest, Florida: a soil study. Ecology 65:1743 1754 Kauffman JB, Sanford RL, Cummings DL, Salcedo IH, Sampaio EVSB (1993) Biomass and nutrient dynamics associated with slash fires in Neotropical dry forests. Ecology 74:140 151 Kawamura K, Takeda H (2008) Development programmed and plastic processes of growth in the mul tistemmed understory shrub Vaccinium hirtum (Ericaceae). Botany 86:268 277 Kaye JP, Hart SC (1998) Ecological restoration alters nitrogen transformations in a ponderosa pine bunchgrass e cosystem. Ecol Appl 8:1052 1060 Keeley JE (1977) Seed production, seed populations in the soil, and seedling production after fire for two congeneric pairs of sprouting and non sprouting chaparral shrubs. Ecology 58:820 829 Keeley JE, Zedler PH (1978) Reproduction of chaparral shrubs after fire: A comparison of sprouting and seeding strategies. Am Midl Nat 99:142 161 Keeley JE, Keeley SC (1988) Chaparral. In: Barbour MG, Billings WD (eds) North American terrestrial vegetation. Cambridge University Press, New York, pp 166 207 Ken nard DK, Gholz HL (2001) Effects of high and lo w intensity fires on soil properties and plant growth in a Bolivian dry for est. Plant Soil 234:119 129

PAGE 245

245 Khanna PK, Raison RJ (1986) Effect of fire intensity on solution chemistry of surface soil under a Eucalyptus pauciflora forest. Aust J Soil Res 24:423 4 34 K imura M, Simbolon H (2002) Allometry and life history of a forest understory palm Pinanga coronata (Arecaceae) on Mount Halimum, West Java. Ecol Res 17:323 338 King DA (1986) Tree form, height growth, and susceptibility to wind damage in Acer saccharum Ecology 67:980 990 Koerper GJ, Richardson CJ (1980) Biomass and net annual primary production regressions for Populus grandidentata on three sited in northern lower Michigan. Can J For Res 10:92 101 Koerselman W, Meuleman AFM (1996) The vegetation N:P ra tio: a new tool to detect the nature of nutrient limitation. J Appl Ecol 33:1441 1450 Kolb KJ, Evans RD (2003) Influence of nitrogen source and concentration on nitrogen isotopic discrimination in two barley genotypes ( Hordeum vulgare L.) Plant Cell Enviro n 26:1431 1440 Konstantinidis P, Tsiourlis G, Xofis P (2006) Effect of fire season, aspect and pre fire plant size on the growth of Arbutus unedo L. (st rawberry tree) resprouts. For est Eco l Man ag 225:359 367 Kozovits AR, Matyssek R, Winkler JB, Gttlein A, B laschke H, Grams TEE (2005) Above ground space sequestration determines competitive success in juvenile beech and spruce tre es. New Phytol 167:181 196 Lagerstrm A, Esber g C, Wardle DA, Giesler R (2009) Soil phosphorus and microbial response to a long te rm wildfire chronosequence in norther n Sweden. Biogeochem 95:199 213 Langely JA, Drake BG, Hungate BA (2002) Extensive belowground carbon storage supports roots and mycorrhizae in regenerating scrub oaks. Oecologia 131:542 548 Lewis Jr. WM (1974) Effects o f fire on nutrient movement in a South Carolina pine forest. Ecology 55:1120 1127 Litaor MI Seastedt TR, Sackett LC (2008) Nutrient status in alpine soils of the Colorado front range using the nitrogen/phosphorus ratio index. Soil Sci Soc Am J 72:1628 163 6 Littell RC, Milliken GA, Stroup WW, Wolfinger RD, Schabenberger O (2006) SAS for mixed models, 2nd edn. SAS Institute Inc., Cary, North Carolina

PAGE 246

246 Little S (1979) Fire and plant succession in the New Jersey pine barrens, In: Forman RTT (ed) Pine barrens: e cosystem and landscape. Academic Press, New York, pp 297 314 Llor et F, Lpez Soria L (1993) Resprouting of Erica multiflora after experi mental fire treatments. J Veg Sci 4:367 374 Lufafa A, Didh iou I, Ndiaye NAS, Sn M, Kizito F Dick RP, Noller JS (2009 ) Allometric relationships and peak season community biomass stocks of native shrubs in 266 Lund M, Christensen TR Mastepanov M, Lindroth A, Strm L (2009) Effects of N and P fertilization on the greenhouse ga s exchange in two northern peatlands with contrasting N deposition rates. Biogeosci 6:2135 2144 MacKen zie MD, DeLuca TH ( 2006 ) Resin adsorption of carbon and nitrogen as influenced by season and time s ince fire. Soil Sci Soc Am J 70:2122 2129 MacKenzie MD DeLuca TH, Sala A (2004) Forest structure and organic horizon analysis along a fire chronosequence in the low elevation forests of western Montana. Forest Eco l Manag 203:331 343 MacKen zie MD, DeLuca TH, Sala A (2006) Fire exclusion and nitrogen mineraliz ation in low elevation forests of western Montan a. Soil Biol Biochem 38:952 961 MacKenzie MD, McIntier E JB, Quideau SA, Graham RC (2008) Charcoal distribution affects carbon and nitrogen contents in forest soils of California. Soil Sci Soc A m J 72:1774 178 5 Mack MC, Treseder KK, Manies KL Harden JW, Schuur EAG, Vogel JG, Rand erson JT, Chapin FS III (2008) Recovery of aboveground plant biomass and productivity after fire in mesic and dry black spruce forests of interio r Alaska. Ecosystems 11:209 225 Mackens en J, Hlscher D, Klinge R, Flster H (1996) Nutrient transfer to the atmosphere by burning of debris in easte rn Amazonia. For est Ecol Manag 86:121 128 Magdoff FR, Bartlett RJ (1985) Soil pH buffering revisited. Soil Sci Soc Am J 49:145 148 Main KN, Menges ES (1997) Archbold Biological Station, Station Fire Management Plan. Land Management Publ. 97 1 Marcos E, Villaln C, Calvo L, Luis Calabuig E (2009) Short term effects of experimental burning on soil nutrients in the Cantabr ian heathlands. Ecol Eng 35:8 20 828

PAGE 247

247 Ma rion GM, Black CH (1988) Potentially available nitrogen and phosphorus along a Chaparral fire cycle chron osequence. Soil Sci Soc Am J 52:1155 1162 Martinelli LA, Piccolo MC, Townsend AR, Vitousek PM, Cuevas E, McDowell W, Robertson GP, Santos OC, Treseder K (1999) Nitrogen stable isotopic composition of leaves and soil: Tropical versus temperate forests. Biogeochem 46:45 65 McEachern P, Prepas EE, Planas D (2002) Phytoplankton in boreal SubArctic lakes following enhanced phosphorus loading from for e fire: impacts on species richness, nitrogen and light limitation. L ake Reserv Manage 18:138 148 McKinley DC, Romero JC, Hungate BA, Drake BG, Megonigal JP (2009) Does deep soil N availability sustain long term ecosystem responses to elevated CO 2 ? Global Change Biol 15:2035 2048 McPh erson K, Williams K (1998) The role of carbohydrate reserves in the growth, resilience, and persistence of cabbage palm seedlings ( Sabal palmetto ). Oecologia 117:460 468 Menges ES (1999) Ecology and conservation of Florida sc rub. In: Anderson RC, Fralish JS, Baskin JM (eds) Savannas, Barrens, and Rock Outcrop Plant Communities of North America. Cambridge University Press, USA pp 7 22 Menges ES (2007) Integrating demography and fire management: an example from Florida scrub. Au st J Bot 55:261 272 Menges ES, Kohfeldt N (1995) Life history strategies of Florida scrub plants in relation to fire. Bull Torrey Bot Club 122:282 297 Menges ES, Quintana Ascencio PF (2004) Population viability with fire in Eryngium cuneifolium : decipheri ng a decade of demographic data. Ecol Monogr 74:79 99 Meril P, Ohtonen R (1997) Soil microbial activity in the coastal Norway spruce [ Picea abies (L.) Karst.] forests of the Gulf of Bothnia in relation to humus layer quality, moisture and soil type s. Bio l Fert Soils 25:361 365 Miao S, Sindhj K, Edelstein C (2008) Allometric relationships of field populations of two clonal species with contrasting life histories, Cladium jamaicense and Typha domingensis Aquat Bot 88:1 9 Michelsen A, Quarmby C, Sleep D, J onasson S (1998) Vascular plant 15 N natural abundance in heath and forest tundra ecosystems is closely correlated with presence and type of mycorrhizal fungi in roots. Oecologia 115:406 418 Molina M, Fuentes R, Caldern R, Escudey M, Avendao K, Gutirrez M, Chang AC (2007) Impact of forest fire ash on surface charge characteristics of andisols. Soil Sci 172:820 834

PAGE 248

248 Moreno JM, Oechel WC (1994) The role of fire in Mediterranean type ecosystems. Springer Verlag, Berlin Mor eno JM, Oechel WC (1991) Fire intens ity and herbivory effects on postfire resprouting of Adenostoma fasciculatum in southern California chaparral. Oecologia 85:429 433 Murphy J, Riley JP (1962) A modified single solution method for determination of phosphate in n atural waters. Anal Chim Acta 26:31 36 Murray RB, Jacobson MQ (1982) An evaluation of dimension analysis for predicting shrub biomass. J Range Manag 35:451 454 Myers RL (1990) Scrub and high pine In: Myers RL, Ewel JJ (eds) Ecosystems of Florida University of Central Florida Press, Orlando p p 150 193 Myster RW (2006) Light and nutrient effects on growth and allocation of Inga vera (Leguminosae), a successional tree of Puerto R ico. Can J For Res 36:1121 1128 Nadelhoffer KJ (2000) The potential effects of nitrogen deposition on fine r oot production in forest ecosystems. New Phytol 147:131 139 Nadelhoffer K, Shaver G, Fry B, Giblin A, Johnson L, McKane R (1996) 15 N natural abundance and N use by tundra plants. Oecologia 107:386 394 Niinemets Kull K (2005) Co limitation of plant prim ary productivity by nitrogen and phosphorus in a species rich wooded meadow on calcareous soils. Acta Oecologica 28:345 356 Nijjer S, Rogers WE, Siemann E (2010) The impacts of fertilization on mycorrhizal production and investment in Western Gulf Coast g ras slands. Am Midl Nat 165:124 133 Niklas KJ (1995a) Plant height and the properties of some herbaceous ste ms. Ann Bot 75:133 142 Niklas KJ (1995b) Size dependent allometry of tree height, diameter, and trunk tap er. Ann Bot 75:217 227 N iklas KJ (1994) Pla nt Allometry; t he scaling of form and proces s. University of Chicago Press, Chicago Niklas KJ, Enquist BJ (2002a) Canonical rules for plant organ biomass partitioning and annual allocation. Am J Bot 89:812 819 Niklas KJ, Enquist BJ (2002b) On the vegetativ e biomass partitioning of seed plant leaves, stems and roots. Am Nat 159:482 497

PAGE 249

249 Niklas KJ, Enquist BJ (2001) Invariant scaling relationships for interspecific plant biomass production rates a nd body size. P Natl Acad Sci 98:2922 2927 Ohlson M, Dahlberg B kland T, Brown KJ, Halvorsen R (2009) The charcoal carbon pool in boreal forest soils. Nat Geosci 2:692 695 Olano JM, Menges ES, Martnez E (2006) Carbohydrate storage in five resprouting Florida scrub plants across a fire chronosequence. New Phytol 170: 99 106 Oa te M, Munn Bosch S (2008) Meristem aging is not responsible for age related changes in growth and abscisic acid levels in the Mediterranean shrub, Cistus clusii Plant Biol 10:148 155 Paradelo R, Barral MT (2009) Effect of moisture and disaggreg ation on the microbial activity of soi l. Soil Till Res 104:317 319 Paula S, Ojeda F (2009) Belowground starch consumption after recurrent severe disturbance in three resprouter species of the genus Erica Botany 87:253 259 Pearson V, Read DJ (1973) The bio logy of mycorrhiza in the Ericaceae. I. The isolation of the endophyte and synthesis of mycorrhizas in aseptic cultures. New Phytol 72:371 379 Peet M, Anderson R, Adams MS (1975) Effect of fire on big bluestem production. Am Mid Nat 94:15 26 Perakis SS, He din LO (2002) Nitrogen loss from unpolluted South American forests mainly via dissolved organi c compounds. Nature 415:416 419 Prez, CA, Carmona M R, Aravena JC, Armesto JJ (2004) Successional changes in soil nitrogen availability, non symbiotic nitrogen fi xation and carbon/nitrogen ratios in southern Chilean forest ec osystems. Oecologia 140:617 625 Pivello VR, Coutinho LM (1992) Transfer of macro nutrients to the atmosphere during experimental burnings in an open cerrado (Brazilian savanna). J Trop Ecol 8:4 87 497 Pornon A, Escaravage N (1999) Genotypic structure in clonal Rhododendron ferrugineum L. (Ericaceae) populations: origin and maintenance. Plant Ecol 141:145 150 Power s RF (1990) Nitrogen mineralization along an altitudinal gradient: interactions of soil temperature, moistur e, and substrate quality. For est Ecol Manag 30:19 29 Prieto Fernndez A, Acea MJ, Carballas T (1998) Soil microbial and extractable C and N after wildfire. Biol Fert Soils 27:132 142

PAGE 250

250 Qian Y Miao SL, Gu B, Li YC (2009) Effects of b urn temperature on ash nutrient forms and availability from cattail ( Typha domingensis ) and sawgrass ( Cladium jamaicense ) in the Florida Everglades. J E nviron Qual 38:451 464 Quintana Ascencio PF, Menges ES, Weekley CW ( 2003 ) A Fire Explicit Population Via bility Analysis of Hypericum cumulicola in Florida Rosemary Scrub. Conserv Biol 17:433 449 Raison RJ, Khanna PK, Woods PV (1985a) Mechanisms of element transfer to the atmosphere during vegetation fires. Can J For est Res 15:132 140 Raison RJ, Khanna PK, W oods PV (1985b) Transfer of elements to the atmosphere during low intensity prescribed fires in three Australian subalpine eucalypt forests. Can J For est Res 15:657 664 Reich PB, Peterson DW, Wedin DA, Wrage K (2001) Fire and vegetation effects on producti vity and nitrogen cycling across a forest grassland continuum. Ecology 82:1703 1719 Riba M (1998) Effects of intensity and frequency of crown damage on resprouting of Erica arborea L. (Erica c eae). Acta Oecol 19:9 16 Rice SK (1993) Vegetation establishment in post fire Adenostoma chaparral in relation to fine scale pattern in fire intens ity and soil nutrients. J Veg Sci 4:115 124 Robertson KM, Ostertag TE (2009) Biomass equations for hardwood resprouts in fire maintained pinelands in the Southeastern United States. South J Appl For 33:121 128 Romany J, Casals P, Vallejo VR (2001) Short term effects of fire on soil nitrogen availability in Mediterranean grasslands and shrubland s growing in old fields. For est E col Manag 147:39 53 Romme WH, Tinker DB, Stakes GK Turner MG (2009) Does inorganic nitrogen limit plant growth 3 5 years after fire in a Wyoming, USA, lodgepole pine forest? For est Ecol Manag 257:829 835 Sah JP, Ross MS, Koptur S, Snyder JR (2004) Estimating aboveground biomass of broadleaved woody plan ts in the understo ry of Flordia Keys pine forests. Forest Ecol Manag 203:319 329 Saha S, Menges ES, Catenazzi A (in review) Does time since fire explain plant biomass allocation in the Flori da scrub ecosystem? Fire Ecol Saha S Strazisar TM, Menges ES, El lsworth P, Sternberg L (2008) Linking the patterns in soil moisture to leaf water potential, stomatal conductance, growth, and mortality of dominant shrubs in the Florida scrub ecosys tem. Plant Soil 313:113 127

PAGE 251

251 Saito L, Miller WM, Johnson DW, Qually RG, Pr ovencher L, Carroll E, Szameitat P (2007) Fire effects on stable isotopes in a Sierran fores ted watershed. J Environ Qual 36:91 100 Sa lemaa M, Sievanan R (2002) The effect of apical dominance on the branching architecture of Arcostaphylos uva ursi in four contrasting environments. Flora 6:429 442 Salifu KF, Timmer VR (2001) Nutrient retranslocation response of Picea mariana seedlings to nitrogen supply. Soil Sci Soc Am J 65:905 913 Sardans J Roda F, Penuelas J (2004) Phosphorus limitation and competitive c apacities of Pinus halpensis and Quercus ilex subsp rotundifolia on different soils. Plant Ecol 174: 305 317 Sarmiento G, Pereira da Silva M, Naranjo ME, Pinillos M (2006) Nitrogen and phosphorus as limiting factors for growth and primary production in a flooded savanna in the Venezuelan llanos. J Trop Ecol 22:203 212 SAS 9.1 (2003) SAS institute Inc. Cary, North Carolina Sch imel JP, Gulledge JM, Clein Curley JS, Lindstrom JE, Braddock JF (1999) Moisture effects on microbial activity and community structu re in decomposition birch litter in the Alaskan taiga. Soil Biol Biochem 31:831 838 Schlesi nger WH, Abrahams AD, P arsons AJ, Wainwright J (1999) Nutrient losses in runoff from grassland and shrubland habitats in Southern New Mexico: I. rainfall simulation experiment s. Biogeochem 45:21 34 Schmalzer PA (2003) Growth and recovery of oak saw palmetto scrub through ten years after fire. Nat Area J 23:5 13 Schmalzer PA, Hinkle CR (1991) Dynamics of vegetation and soils of oak/saw palmetto scrub after fire: obser vations from permanent transects. NASA: 1 149 Schmalzer PA, Hinkle CR (1992) Species composition and structure of oak saw palmetto scrub vegetation. Castanea 57:220 251 Schmalzer PA, Hinkle CR (1996) Biomass and nutrients in aboveground vegetation and soil s of Florida oak saw palmetto scrub. Castanea 61:168 193 Schmidt S, Stewart GR (1997) Waterlogging and fire impacts on nitrogen availability and utilization in a subtropical wet heathland (wallum). Plant Cell Environ 20:1231 1241 Schmidt S, Stewart GR (20 15 N values of tropical savanna and monsoon forest species reflect root specialisations and soil nitrogen status. Oecologia 2003:569 577

PAGE 252

252 Schuur EAG, Matson PA (2001) Net primary productivity and nutrient cycling across a mesic to wet precipitation grad ient in Hawaiian montan e forest. Oecologia 128:431 442 Scogin gs PF, Mopipi K (2008) Effects of water, grass and N on responses of Acacia karroo seedlings to early wet season simulated browsing: Aboveground growth and biomass allocation. J Arid Environ 72:5 09 522 Scott Denton LE, Rosenstiel TN, Monson RK (2006) Differential controls by climate and substrate over the heterotrophic and rhizospheric components of soil respiration. Glob al Chang e Biol 12:205 216 Scowcroft PG, Friday JB, Idol T, Dudley N, Haraguch i J, Meason D (2007) Growth response of Acacia koa trees to thinning, grass control, and phosphorus fertilization in a secondary forest 239:69 80 Seiler TJ, Rasse DP, Li J Dijkstra P, Anderson HP, Johnso n DP, Powell TL, Hungate BA, Hinkle CR, Drake BG (2009) Disturbance, rainfall, and constrasting species responses mediated aboveground biomass response to 11 years of CO 2 enrichment in a Florida scrub oak ecosystem. Global Change Biol 15:356 367 Serrasolsas I, Khanna PK (1995) Cha nges in heated and autoclaved forest soils of S.E. Australia. I. Carbon and nitrogen. Biogeochem 29:3 24 Sharma IK (2001) Understanding clonal diversity patterns through allozyme polymorphism in an endangered and geographically restricted Australian shrub, Zieria baeuerlenni and its implicatio ns for conservation. Biochem Sys t Ecol 29:681 695 Shen SM, Pruden G, Jenkinson DS (1984) Mineralization and immobilization of nitrogen in fumigated soil and the measurement of microbial biomass nitrogen. Soil Biol Bio chem 16:437 444 Sigma Plot for Windows 11.0 (2008) Systat Software, Inc. Chicago, Illinois Silva IA, Valenti MW, Silva Matos DM (2009) Fire effects on the population structure of Zanthoxylum rhoifolium Lam (Rutaceae) in a Brazilian savanna. Braz J Biol 69 :815 820 Sims SE, Hendr icks JJ, Mitchell RJ, Kuehn KA, Pecot SD (2007) Nitrogen decreases and precipitation increases ectomycorrhizal extramatrical mycelia production in a longleaf pine forest. Mycorrhiza 17:299 309 Singh RS, Srivastava SC, Raghubanshi AS Singh JS, Singh SP (1991) Microbial C, N and P in dry tropical savanna: effects of burning and grazing. J Appl Ecol 28:869 878

PAGE 253

253 Sinsabaugh RL, Lauber CL, Weintraub MN, Ahmed B, Allison SD, Crenshaw C, Cusack D, Frey S, Gallo ME, Gartner TB, Hobbie SE, Hol land K, Keeler BL, Powers JS, Stursova M, Takacs Vesbach C, Waldrop MP, Wallenst ein MD, Zak DR, Zeglin LH (2008) Stoichiometry of soil enzyme activi ty at global scale. Ecol Lett 11:1 252 1264 Skyllberg U, Raulund Rasmussen K, Borggaard OK (2001) pH bufferin g in acidic soils developed under Picea abies and Quercus robur effects of soil organic matter, adsorbed cations and soil solution ionic strength. Biogeochem 56:51 74 Smithwick EAH, Turner MG, Mack MC, Chapin III FS (2005 a ) Postfire soil N cycling in nor thern conifer forests affected by severe, stand replacing wildfires. Ecosystems 8:163 181 Smithwick EAH, Turner MG, Metzger KL, Balser TC (2005 b ) Variation in NH 4 + mineralization and microbial communities with stand age in lodgepole pine ( Pinus contorta ) f orests, Yellowstone National Park (USA). Soil Biol Biochem 37:1546 1559 Sokal RR, Rohlf FJ (1995) Biometry 3 rd edition. Freeman, New York Sollins P, Glassman C, Paul EA, Swanston C, Lajtha K, Heil JW, Elliott ET (1999) Soil carbon and nitrogen: pools and f ractions. In: Robertson GP, Coleman DC, Bledsoe CS, Sollins P (eds) Standard soil methods for long term ecological research. Oxford University Press, Inc., New York pp 89 105 Sousa WP (1984) The role of disturbance in natural communities. Ann u Rev Ecol Sys t 15:353 391 SPSS 11.5 for Windows (2000) SPSS, Inc. Chicago, Illinois Stephens SL, Meixner T, Poth M, McGurk B, Payne D (2004) Prescribed fire, soils, and stream water chemistry in a watershed in the Lake Tahoe Basin, California. Int J Wildland Fire 13:27 35 Stevens JT, Beckage B (2009) Fire feedbacks facilitate invasion of pine savannas by Brazilian pepper ( Schinus terebinthifolius ) New Phytol 184:365 375 Stock WD, Lewis OAM (1986) Soil nitrogen and the role of fire as a mineralizing agent in a South Afr ican coastal fynbos ecosystem. J Ecol 74:317 328 Takenaka A, Takahashi K, Kohyama T (2001) Optimal leaf display and biomass partitioning for efficient light capture in an understorey palm, Licuala arbuscula Funct Ecol 15:660 668 Tanner EVJ, Kapos V, Franc o W (1992) Nitrogen and phosphorus fertilization effects on Venezuelan montane forest trunk growth and litterfall. Ecol 73:78 86

PAGE 254

254 Tate III RL, Terry RE (1980) Variation in microbial activity in histisols and its relationsh ip to soil moisture. Appl Enviro n Microbiol 40:313 317 Teng, Y. and V.R. Timmer. 1995. Rhizosphere phosphorus depletion induced by heavy nitrogen fertilization in forest nursery soils. Soil Sci Soc Am J 59 :227 233 Tessier JT, Raynal DJ (2003) Use of nitrogen to phosphorus ratios in plant tissue as an indicator of nutrient limitation and nitrogen saturation. J Appl Ecol 40:523 534 Thomas GW (1996) Soil pH and soil acidity. In: Bartels JM et al. (eds) Methods of soil analysis Part 3 Chemical Methods. Soil Science Society of America, Inc., M adison, Wisconsin, pp 475 490 Todd, M CL, Adams MA, Grierson PF (2000) Mineralisation of nitrogen in a chronosequence of rehabilitated bauxite mi nes. Aust J Soil Res 38:435 451 Toma T Ishida A, Matius P (2005) Long term monitoring of post fire aboveground biomass recovery in a lowland dipterocarp forest in East Kalimantan, Indonesia. Nutr Cyc l Agroecosys 71:63 72 Tomkins IB, Kellas JD, Tolhurst KG, Oswin DA (1991) Effects of fire intensity on soil chemistry in a Eucalypt forest. Aust J Soil Res 29:25 47 Tre s eder KK, Mack MC, Cross A (2004) Relationships among fires, fungi, and soil dynamics in Alaskan boreal forests. Ecol Appl 14:1826 1838 T ripathi SK, Kushwaha CP, Singh KP (2008) Tropical forest and savanna ecosystems show differential impact of N and P add itions on soil organic matter and aggregate structure. Global Change Biol 14:2572 2581 Turner MG, Smithwick EAH, Metzger KL, Tinker DB, Romme WH (2007) Inorganic nitrogen availability after severe stand replacing fire in the Greater Yellowstone ecosystem. P Natl Acad Sci 104:4782 4789 Turner M G Tinker DB, Romme WH, Kashian DM, Litton CM (2004) Landscape patterns of sapling density, leaf area, and aboveground net primary production in postfire lodgepole pine forests, Yellowstone National P ark (USA). Ecosys tems 7:751 775 van Schll L, Hof fland E, van Breemen N (2006) Organic anion exudation by ectomycorrhizal fungi and Pinus sylvestris in response to nutrient deficienci es. New Phytol 170:153 163 v an de Vijver CADM, Poot P, Prins HHT (1999) Causes of increase d nutrient concentrations in post fire regrowth in an East African savanna. Plant Soil 214:173 185

PAGE 255

255 Var gas R, Allen MF, Allen EB (2008) Biomass and carbon accumulation in a fire chronosequence of a seasonally dry tropical fores t. Global Change Biol 14:109 1 24 Vil M (1997) Effect of root competition and shading on resprouting dynami cs of Erica multiflora L. J Veg Sci 8:71 80 Villani EMA, Barros NF, Novais RF, Comerford NB, Costa LM, Neves JCL, Alvarez VH (1998) Phosphorus diffusive flux as affected by phosph ate source and incubation time. Soil Sci Soc Am J 62:1057 1061 Vitousek PM, Howarth RW (1991) Nitrogen limitation on land and in the sea How can it occur? Biogeochem 13:87 115 Vi tousek PM, Walker LR, W hiteaker LD, Matson PA (1993) Nutrient limitations t o plant growth during primary succession in Hawaii Volcanoes National Park. Biogeochem 23:197 215 Vitouse k PM, Farrington H (1997) Nutrient limitation and soil development: experimental test of a biogeochemical theory. Biogeochem 37:6 3 75 Vora RS (1988) Predicting biomass of five shrub species in nort heastern California. J Range Man age 41:63 65 Wan S, Hui D, Luo Y (2001) Fire effects on nitrogen pools and dynamics in terrestrial ecosystems: a meta analysis. Ecol Appl 11:1349 1365 Wanthongchai K Bauhus J, Goldammer JG (2008) Nutrient losses through prescribed burning of aboveground litter and understorey in dry dipterocarp forests of different fire history. Catena 74:321 332 Wassen MJ, Venteri nk HO, Lap shina ED, Tanneberger F (2005) Endangered plants persi st under phosphorus limitation. Nature 437:547 550. Weekley CW, Menges ES (2003) Species and vegetation responses to prescribed fire in a long unburned, endemic rich Lake Wales Ridge scrub. J Torrey Bot Soc 130:265 282 Weekley CW Gagnon D, Menges ES, Qui ntana Ascencio PF, Saha S (2007) Variation in soil moisture in relation to rainfall, vegetation, gaps, and time since fire in Florida scrub. Ecoscience 14:377 386 W einer J, Thomas SC (1992) Competition and allometry in three species of an nual plants. Ecolo gy 73:648 656 Weiner J, Berntson GM, Thomas SC (1990) Competition and growth form in a woodland annual J Ecol 78:459 469

PAGE 256

256 West GB, Brown JH, Enquist BJ (1999) A general model for the structure and allometry of plant vascu lar systems. Nature 400:664 667 Whi te EM, Thompson WW, Gartner FR (1973) Heat effects on nutrient release from soils under Po nderosa pine. J Range Manage 26:22 24 Wh ite LL, Zak DR, Barnes BV (2004) Biomass accumulation and soil nitrogen availability in an 87 year old Populus grandidentata c hronosequence. For est Ecol Manag 191:121 127 Wilbur RB, Christensen NL (1983) Effects of fire on nutrient availability in a North Carolina coastal plain pocosin. Am Midl Nat 110:54 61 Wilson CA, Mitchell RJ, Boring LR, Hendricks JJ (2002) Soil nitrogen dyn amics in a fire maintained forest ecosystem: results over a 3 year burn interval. Soil Biol Biochem 34:679 689 Wunderlin RP, Hanson BF (2003) Guide to the vascular plants of Florida, 2nd edn. University Press of Florida Yermakov Z, Rothstein DE (2006) Cha nges in soil carbon and nitrogen cycling along a 72 year wildfire chronosequence in Michigan jack pine forests. Oecologia 149:690 700. Zewdie M, Olsson M, Verwijst T (2009) Above ground biomass production and allometric relations of Eucalyptus globulus La bill. coppice plantations along a chronosequence in the central highlands of Ethiopia. B iomass Bioenerg 33:421 428

PAGE 257

257 BIOGRAPHICAL SKETCH Jennifer Schafer was born in 1979 in Columbus, Ohio. She graduated from Upper Arlington High School in 1997. Jennifer earned a Bachelor of Arts in zoology, with a minor in b otany from Miami University in May 2001. She served as an Education Intern and a Plant Ecology Intern at Archbold Biological Station in Lake Placid, Florida. Because of her love for Florida scrub eco systems, Jennifer decided to condu c t her dissertation research in the scrub preserve at Archbold Biological Station Jennifer is a Buckeye and a Gator.