Contributions of Above- and Below-Ground Litter to Short-Term Accretion and Maintenance of Surface Soil Organic Carbon i...

MISSING IMAGE

Material Information

Title:
Contributions of Above- and Below-Ground Litter to Short-Term Accretion and Maintenance of Surface Soil Organic Carbon in an Intensively Managed North Florida Loblolly Pine Stand
Physical Description:
1 online resource (145 p.)
Language:
english
Creator:
Stoppe, Aja Marie
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Soil and Water Science
Committee Chair:
Comerford, Nicholas B
Committee Members:
Jokela, Eric J
Mackowiak, Cheryl L
Higa, Rosana

Subjects

Subjects / Keywords:
carbon -- florida -- formation -- fraction -- loblolly -- mineralization -- pine -- soil
Soil and Water Science -- Dissertations, Academic -- UF
Genre:
Soil and Water Science thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The primary supply of organic carbon (C) entering the soil ecosystem originates from plant sources but the quantitative importance of the C source to the formation of SOC pools has only recently begun to be elucidated. Forest soils primarily receive C inputs from leaf litter and fine root turnover. The purpose of this study was to explore the short-term formation, maintenance, and mineralizability of SOC at the surface of a sandy soil supporting a fast growing, mid-rotation loblolly pine in the southeastern United States.This study examines the SOC processes at the surface 0-20 cm, where fine root density, macro- and microbial biomass, and the interaction between the mineral soil and forest floor are the greatest. This was accomplished by three specific objectives: 1) establish the natural short-term changes in SOC that occurred in whole soil and physical size fractions; 2) characterize the inherent mineralizability of C located in size fractions and assess aggregate stability in the soil’s large sand size macroaggregates; 3) use sequential exclusion of above ground litter inputs and above- plus below-ground inputs to investigate the importance of C sources to the development. This short-term observation measured a mean annual accretion rate of 2.3 mg SOC cm-3soil yr-1. The majority of the SOC increase occurs in the soil closest to the surface and primarily in two size fractions, 2000-250 µm and 150-53µm. Carbon mineralization was primarily determined by the C content of the fraction and losses were concentrated in the largest fraction, 2000-250 µm. On a whole soil basis C derived from below-ground sources provided the main contributions to increasing and maintaining SOC pools during this phase of stand development, specifically to the >2 mm and 150-53 µm fractions.Aggregation was present but minimally affected by the exclusion treatments. Data from this study suggest SOC development is primarily dependent on fine root turnover. The substantial SOC accretion observed may indicate a particular phase of rapid SOC development and long-term SOC developmental pattern to resemble the typical “S” curve growth pattern similar to tree growth, needle fall and fine root growth.
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 Aja Marie Stoppe.
Thesis:
Thesis (M.S.)--University of Florida, 2012.
Local:
Adviser: Comerford, Nicholas B.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-02-28

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044733:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 CONTRIBUTIONS OF ABOVE AND BELOWGROUND LITTER TO SHORT TERM ACCRETION AND MAINTENANCE OF SURFACE SOIL ORGANIC CARBON IN AN INTENSIVELY MANAGED NORTH FLORIDA LOBLOLLY PINE STAND By AJA STOPPE A THESIS PRESENTED TO THE GRAD UATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER IN SCINECE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Aja Stoppe

PAGE 3

3 To all the rocks in my life, it is through your endless support and encour agement I am here today ; and to all the h oles dug have no perspective

PAGE 4

4 ACKNOWLEDGMENTS Embodied in the accomplishment of this thesis is the guidance, support and encouragement from many inspirational individuals and I am grateful f or the opportunity to thank those who have helped bring it to fruition Foremost I would like to thank m y graduate committee: Dr. Jokela for his insight into forestry and editorial efforts, Dr. Hega for with her gracious Brazilian hospitality and mentorsh ip ; Dr. Mackowiak for providing me with an opportunity to experience agricultural research and a great source of knowledge; Dr. Comerford my major advisor, i t is through his encouragement, guidance, and trust that I have accomplished so much as a graduate student Outside of my committee comes a long list of characters that is essentially summed up through my connection to the University of Florida Forest Soils Laboratory. Through this entity countless personal and professional relationships have been cr eated and I treasure them T hey are f iends, role models, mentors, and a valuable source to count on now and in the coming future. A deep gratitude goes to the heart and center of my universe, my family I would not be here without you and I am forever gra teful for your love and protection. You Fi nal ly very essential recognition goes t o the University of Florida Employee Education Program As my primary funding source it is i mportant for the University to know that the benefits it provide s to its staff has the power to enrich and better their lives.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION TO FOREST SOIL ORGANIC CARBON FORMATION ............. 12 2 SOIL ORGANIC CARBON DISTRIBUTION, SHOR T TERM CHANGES, MINERALIZABILITY AND AGGREGATION IN A SPODOSOL UNDER LOBLOLLY PINE ................................ ................................ ................................ .... 17 Methodology ................................ ................................ ................................ ........... 21 Experimental Site ................................ ................................ ............................. 21 Soil and Litterfall Sampling ................................ ................................ ............... 24 Laboratory Methods ................................ ................................ ......................... 25 Statistical Meth ods ................................ ................................ ........................... 28 Results ................................ ................................ ................................ .................... 29 Objective 1. Determine Short Term Changes of SOC and its Distribution among Size Fractions ................................ ................................ ................... 29 Objective 2. Assess the Potential Mineralizablity of SOC within and among Size Fractions from Soil Collected during the 31 Month Sampling ................ 31 Obj ective 3. Quantify the AOC and Evaluate the Stability of Aggregates in the Large Sand Size Fraction (2000 250 m) from Soil Collected during the 31 Month Sampling ................................ ................................ ................. 31 Discussion ................................ ................................ ................................ .............. 32 Objective 1. Short Term Changes and Distribution of SOC .............................. 32 Objective 2. Assess the Potential Mineralizability of SOC within and among Soil S ize Fractions ................................ ................................ ........................ 34 Objective 3. Quantify the AOC and Evaluate the Stability of Aggregates in the Large Sand Size Fraction (2000 250 m) ................................ ............... 35 Conclusion ................................ ................................ ................................ .............. 36 3 THE CONTRIBUTIONS OF ABOVE AND BELOWGROUND sources of C SUSTAINING SURFACE SOIL ORGANIC C IN AN INTENSIVELY MANAGED LOBLOLLY PINE STAND ................................ ................................ ....................... 44 Methodology ................................ ................................ ................................ ........... 48

PAGE 6

6 Field Treatments ................................ ................................ .............................. 48 Soil Sampling ................................ ................................ ................................ ... 50 Live Root Observations ................................ ................................ .................... 50 Laboratory Methods ................................ ................................ ......................... 51 Statistical Methods ................................ ................................ ........................... 51 Results ................................ ................................ ................................ .................... 52 Discussion ................................ ................................ ................................ .............. 54 Conclusion ................................ ................................ ................................ .............. 58 4 THE RELATIVE IMPROTANCE OF CARBON SOURCES CONTIRBUTING TO SOIL ORGANIC CARBON POOLS IN AN INTENSIVLY MANAGED LOBLOLLY PINE STAND ................................ ................................ ................................ .......... 63 Methods ................................ ................................ ................................ .................. 66 Field Treatments ................................ ................................ .............................. 66 Soil Sampling ................................ ................................ ................................ ... 67 Laboratory Methods ................................ ................................ ......................... 68 Statistical Methods ................................ ................................ ........................... 70 Results ................................ ................................ ................................ .................... 71 Objective 1. Assess the Contribution of Above and Belowground Sources to S OC Pools Defined by Size Fraction ................................ ........................ 71 Objective 2. Determining the Effect of Limiting C Sources on the Stability of Macroaggregates ................................ ................................ .......................... 73 Discussion ................................ ................................ ................................ .............. 74 Objective 1. Assess the Contribution of Above and Belowground Sources to SOC Pools Defined by Size Fraction ................................ ........................ 74 Objective 2. Determining the Effect of Limiting C Sources on the Stability of Macroaggregates ................................ ................................ .......................... 76 Conclusion ................................ ................................ ................................ .............. 77 5 SYNTHESIS ................................ ................................ ................................ ........... 82 APPENDIX A STATISTICAL OUTPUT FROM SAS ................................ ................................ ...... 93 B CONFIDENCE INTERVALS OF RELATIVE PERCENT CHANGE ANALYSIS .... 131 LIST OF REFERENCES ................................ ................................ ............................. 133 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 144

PAGE 7

7 LIST OF TABLES Table page 2 1 Mean total SOC increase among the soil size fractions at two soil depths ......... 42 2 2 ANOVA table of the specific mineralization rates of cumulative C respired f rom size fraction SOC, at two soil depths, after 162 d incubation at 30 C. ........ 43 2 3 ANOVA table of soil size fraction bulk mineralization rates of cumulative C respired t two soil depths, after 162 d incubation at 30 C ................................ .. 43 4 1 Sample means and standard deviations of changes in SOC (mg cm 3 ) among soil size fractions after 31 months of exclusion treatment. ................................ 81 A 1 Chapter 2 Statistical output from SAS ................................ ................................ 93 A 2 Chapter 3 Statistical output from SAS ................................ .............................. 110 A 3 Chapter 4 Statistical output from SAS ................................ .............................. 116 B 1 The 90% confidence limits of the relative percent change in soil bulk density after 31 months of exclusion treatments ................................ ........................... 131 B 2 The 90% confidence limits of the relative percent change in coarse fraction SOC after 31 months of exclusion treatments ................................ .................. 131 B 3 The 90% confid ence limits of the relative percent change in fine earth fraction SOC after 31 months of exclusion treatments ................................ .................. 131 B 4 The 90% confidence limits of the relative percent increase in SOC among th e soil size fractions after 31 months of exclusion treatments ............................... 132

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Differences in soil bulk density in the untreate d control plots at two soil depth s ................................ ................................ ................................ ................. 38 2 2 Differences in SOC in the fine earth fraction over time for two soil depths ......... 38 2 3 Dis tribution of SOC among soil size fractions averaged across time for two soil depths ................................ ................................ ................................ .......... 39 2 4 Differences in soil size fraction SOC occuring naturally over time ...................... 39 2 5 Differences in specific mineralization rates of cumulative C respired from size fraction SOC, at two soil depths, after 162 d incubation at 30 C ........................ 40 2 6 Differences in bulk soil C respired from four soil size fractions, at two soil depths, after 162 d incubation at 30 C ................................ ............................... 41 2 7 Dispersion of AOC as % of the 2000 depths, as a function of increasing applied sonic energy ................................ ... 42 3 1 Evaluation of the effectiveness of root exclu sion treatment through live root counts ................................ ................................ ................................ ................. 60 3 2 Differences in % change in soil bulk density after to 31 months of exclusion treatment ................................ ................................ ................................ ............ 60 3 3 Differences in % change of the coarse fraction (>2 mm) after 31 months of exclusion treatment ................................ ................................ ............................ 61 3 4 Differences in % change of the fine earth fraction (<2 mm) after 31 months of exclusion treatment ................................ ................................ ............................ 62 4 1 Differences in % change of physical size fractions after 31 months of exclusion treatment ................................ ................................ ............................ 79 4 2 Diff erences in AD EC of AOC as % of total SOC in the 2000 250 m fraction after 31 months of exclusion treatment ................................ ............................... 80 5 1 Estimated annual C inputs and outputs in intensively managed, mid rotation lo blolly pine stand growing in an Ultic Alaquod in north central Florida. ............. 91 5 2 SOC development in an intensively managed loblolly pine s tand may resemble a similar trend .......... 92

PAGE 9

9 LIST OF ABBREVIATIONS ADEC Aggregate d ispersion e nergy curve AOC Aggregate o rganic c arbon C Carbon CEC Cation exchange capacity d Day DOC Dissolved o rganic c arbon DRIFTS Dif fuse reflectance i nfrared F ourier transform s pectroscopy HCl Hydrochloric acid LOI Loss on ignition POC Particulate organic carbon SOM Soil organic m atter SOC Soil organic yr Year

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the Univ ersity of Florida in Partial Fulfillment of the Requirements for the Degree of Master in Science CONTRIBUTIONS OF ABOVE AND BELOWGROUND LITTER TO SHORT TERM ACCRETION AND MAINTENANCE OF SURFACE SOIL ORGANIC CARBON IN AN INTENSIVELY MANAGE D NORTH FLORIDA LOBLOLLY PINE STAND By Aja Stoppe August 2012 Chair: Nicholas Comerford Major: Soil and Water Science The primary supply of organic carbon (C) entering the soil ecosystem originates from plant sources but the quantitative importance of t he C source to the formation of SOC pools has only recently begun to be elucidated. Forest soils primarily receive C inputs from leaf litter and fine root turnover. The purpose of this study was to explore the short term formation, maintenance, and miner alizability of SOC at the surface of a sandy soil supporting a fast growing, mid rotation loblolly pine ( Pinus taeda L .) in the southeastern United States This study examines the SOC processes at the surface 0 20 cm, where fine root density, macro and mi crobial biomass, and the interaction between the mineral soil and forest floor are the greatest. This was accomplished by three specific objectives: 1) establish the natural short term changes in SOC that occurr ed in whole soil and physical size fractions; 2) characterize the inherent large sand size macroaggregates ; 3) use sequential exclusion of aboveground litter inputs and above plus belowgro und inputs to inves tigate the importance of C sources to the development

PAGE 11

11 This short term observation measured a mean annual accretion rate of 2.3 mg SOC cm 3 soil yr 1 The majority of the SOC increase occurs in the soil closest to the surface and primarily in two size fra ctions, 2000 250 m and 15 0 53 m. Carbon mineralization wa s primarily determined by the C content of the fraction and losses were concentrated in the largest fraction, 2000 250 m. On a whole soil basis C derived from belowground sources provided the main contributions to increasing and maintaining SOC pools during this phase of stand development specifically to the >2 mm and 150 53 m fractions. Aggregation was present but minimally affected by the exclusion treatments. Data from this study suggest SOC d evelopment was primarily dependent on belowground sources of C, namely fine root turnover. The substantial SOC accretion observed may indicate a particular phase of rapid SOC development and long term urve growth pattern similar to tree growth, needle fall and fine root growth.

PAGE 12

12 CHAPTER 1 INTRODUCTION TO F OREST SOIL ORGANIC C ARBON FORMA TION Although soil organic carbon (SOC) content has been a cornerstone of soil quality parameters and plays a sign ificant role in the global carbon cycle, a complete understanding of the formation process and dynamics is still unclear. The primary supply of organic carbon entering the soil ecosystem originates from plant sources (Vogt et al. 1986) but the quantitative importance of the C source to the formation of SOC pools has only recently begun to be elucidated (Rasse et al. 2005). Recent advances in the understanding of terrestrial C cycling have come about with the employment of long term exclusion studies, the us e of stable isotopes, isotopic labeling and molecular markers ( Johnston et al. 2004 ; Kuzyakov 2011 ). The simple approach of prolonged forest floor removal has been useful in estimating the role leaf litter plays in the formation and maintenance of stable S OC pools. Garten (2009) used prolonged O horizon removal (4.5 years) to determine the aboveground litter of a Tennessee temperate hardwood forest was not a significant source of C maintaining SOC. Likewise, results from another litter and root manipulation study in place for 5 years in a northeastern hardwood forest, indicated SOC was not affected by aboveground litter additions or removal (Nadelhoffer et al. 2004). An important observation came from Froberg et al. (2007) who concluded there was a disconnec tion between the C cycling in the O horizon and the mineral soil. The authors traced dissolved organic carbon (DOC) leaching from C 14 enriched fresh litter through the soil profile and found it making limited contributions to the DOC pool at the 15 cm soi l depth. The disconnected O horizon theory was concluded when C 14 labeled leaf litter was found primarily in the microbial biomass of the O horizon and only a very small

PAGE 13

13 portion of microbial C arrived in the surface soil. Froberg et al. (2007) and other a uthors suggested that the major source of C utilized by soil microbes originated from roots or soil humus ( Karltun et al. 2005 ; Garten 2009 ; Kramer et al. 2010). Potential contributions of C from forest floor and fine root turnover have been estimated to be approximately equal in hardwoods and 1:2 in mature loblolly pine ( Pinus taeda L .) (Nadelhoffer et al. 2004; Richter et al. 1999). Aboveground sources of C contributing to SOC are limited without a mechanism for the biomass to be incorporated into the so il. Quideau et al. (2001) contrasted soils of two very different forest types in southern California, scrub oak (Quercus dumosa Nutt.) and Coulter pine (Pinus coulteri B. Don). Aboveground litter in the oak forest rapidly decomposed and mixed into the A ho rizon by a large earthworm population. In contrast the pine litter turned over very slowly, experienced no earthworm mixing and caste production. Very little aboveground litter was added to the A horizon. The authors speculate that roots provide the majori ty of C input to the formation of SOC under this particular pine ecosystem. In the short term, studies suggest the absence of aboveground litter does not affect SOC in forests that do not have a large influence of soil mixing agents (Quideau et al. 2001). The working hypothesis that the belowground C sources dominate the formation of SOC in some forested ecosystems centers around the opportunity of location and size. Large amounts of photosynthetically fixed C are drawn into the soil matrix by roots and dep osit a variety of organic materials into the rhizosphere (Farrar et al. 2003 Godbold et al. 2006). Concentrations of SOC are created in the rhizosphere where root exudations support large microbial populations and microbial residues, predominantly lysed ce ll envelopes, have been seen to persist in soils ( Kindler et al. 2006 and 2009; Simpson et

PAGE 14

14 al. 2007a; Miltner et al. 2011). The microbial necromass has recently been argued by Miltner et al. (2011) and others to be a significant source of C stabilized on m ineral surfaces or as small particulate organic matter (POM). The excretions produced by root tips and mycorrhizae are sticky with hydrophobic properties. These substances (e.g. mucilage) are an important factor in soil aggregate formation and stability (O ades 1978; Czarnes et al. 2000). Parti c le size matters when it comes to creating stable soil aggregates. The particulate organic matter that roots deposit in soil is within a size scale that can easily develop into aggregates capable of physical protectio n (Rasse et al. 2005). Aggregates that have small pore spaces exclude microbial activity and physically protect the organic C located within the aggregate (Six et al. 2000). Aboveground litter must go through a much longer process of decomposition to arriv e at that size and has less opportunity to enter into intimate interactions with the mineral soil (Schmidt et al. 2011). Our overall objective in this thesis was to explore the formation, maintenance, and mineralizability of SOC at the surface of a typical sandy soil type that supports millions of acres of planted pine in the southeastern United States. Florida southern pine plantations are commonly grown on seasonally wet Spodosols. Soil OC dynamics change with soil depth and this study examines the SOC pr ocesses at the surface 0 20 cm, where the fine root density, macro and microbial biomass, and the interaction between the mineral soil and forest floor is the greatest (Adegbidi et al. 2004) Through intensive management regimes pine plantations are succ essful on these nutrient poor soils and produce large amounts of above and belowground C biomass (Jokela et al. 2010; Vogel et al. 2011). Spodosol surface soils in this subtropical region

PAGE 15

15 are excessively sandy with low clay content and weak, crumb structu re, suggesting a low potential for protecting SOC from decomposition. Only recently has research looked aggregates with degrees of stability and identified C pools with d ifferent chemical composition in pine plantation Spodosols. The authors also discovered that some SOC pools were responsive to management while others may only respond after long time periods. This suggests the importance of further understanding the devel opment and maintenance of SOC under this cover type in order to capitalize not only on the economic worth of pine commercial products but to include the value of SOC storage that may be provided. We approach this objective by evaluating the short term cont ributions and control made by aboveground and belowground C sources to the formation and maintenance of various SOC pools. This was accomplished by three specific objectives. The first objective was to establish the natural short term changes in SOC occurr ing over a 31 month study period (Chapter 2). The transformations of SOC over this time were documented in whole soil and in physical size fractions. The process of dividing soil into physical size fractions has been a successful method in separating C poo ls with different physical and chemical characteristics (Christensen 2001). To characterize the inherent mineralizability of C located in these size fractions, a laboratory incubation study was conducted to measure the C mineralization rates under control led conditions. 250 m size class). Aggregates in this size class have been noted for containing large

PAGE 16

16 management changes (Sarkot et al. 2007b). Chapter 3 explores the second objective of this thesis which was to investigate the importance of above and belowground sources C to the development of SOC. By sequential exclusion of aboveground litter inputs a nd above plus belowground C inputs to maintaining SOC content in the whole soil in the short term. The third objective was to investigate the influence of the respect ive C source on the various SOC pools. This was accomplished with the same sequential exclusion treatments by evaluating the changes in SOC evidenced at the physical size fraction scale and changes to the stability of 2000 250 m macroagg regates (Chapter 4 ). The focus of this objective wa s to locate the SOC pools that we re most regulated by C originating from above or belowground sources and estimate the short term changes of these SOC pools. The final chapter, Chapter 5, is a summary of research on SOC fo rmation accomplished in this thesis and a synthesis of the conclusions made in each chapter. The gaps in knowledge that remain on this subject and suggestions of further research were identified.

PAGE 17

17 CHAPTER 2 SOIL ORGANIC CARBON DISTRIBUTION, SHORT TERM CHAN GES, MINERALIZABILITY AND AGGREGATION IN A SPO DOSOL UNDER LOBLOLLY PINE Spodosols of the l ower Coastal Plain of the U.S. span approximately 5.7 million ha (Adegbidi et al. 2002) and have some of the highest soil organic carbon (SOC) stocks in the region ( Stone et al. 1993). These soils commonly support intensively managed, high producing pine plantations with significant opportunity for above ground C storage (Vogel et al. 2011). In contrast, the potential for increasing belowground C storage in these fore sts has been described as limited (Shan et al. 2001) due to acidic, sandy nature of the soil and subtropical climate that encourages high mineralization rates (Richter et al. 1999; Six et al. 2002). Turnover of surface SOC in Spodosols has been estimated i n forested tropical soils to be between several years in active pools and 60 70 years in more protected pools (Telles et al. 2003). Recently various SOC pools in Florida Spodosols have been investigated and also indicated some SOC pools may be more protec ted from mineralization than others (Sarkhot et al. 2007a, b). Understanding of the processes of SOC development and the dynamics of turnover are critical to realizing the maximum potential of C sequestration. Soil is composed of C pools that have differen t chemical characteristics and different mineral soil interactions which produce a range of soil carbon residence times. Soil OC turnover has been linked to soil texture and aggregation (Hassink et al. 1997; Telles 2003; Ltzow et al. 2006). Highly active SOC pools have a mean residence time (MRT) of 1 2 years and they often contain one quarter to two thirds of the initial fresh C input (Jenkinson and Ladd 1981). This pool is composed of labile and unprotected C. Other pools have a more intermediate MRT l asting a few decades and some are much

PAGE 18

18 more persistent with MRTs of hundreds or thousands of years ( Schimel et al. 1994 ). The residence time of organic material is determined initially by its resistant chemical nature, while longer turnover is dictated by its mode of protection from microbial decomposition (e.g. incorporation into soil aggregates, complexation with soil minerals and metal oxides; Golchin et al. 1994; Six et al. 2002, Schmidt et al 2011). Subdividing soil into size fractions has become an e ffective analytical method for studying SOC pools (Christensen 1992 and 1996; Buyanov esky 1994; Sarkhot et al. 2007b ; Brunn et al. 2010 ). This approach is operationally defined, and not a true division of functionally different SOC pools (Brunn et al. 2010 ). However, it recognizes that biological processes driving SOC turnover are regulated by soil structure and that mineralization is limited when organic material is chemically or physically protected by association with soil minerals and soil aggregation, which can be related to size class (Christensen 2000; Conant et al. 2004) Carbon sto rage in the sandy soils of the l ower Coastal Plain is particularly dominates the first three so il horizons, lacks the cation exchange capacity ( CEC ) of finer textured soils (Harris and Carlisle 1987). This condition does not facilitate aggregation, while the warm subtropical climate promotes leaching, stimulates microbial activity and encourages rap id decomposition (Alvarez and Lavado 1998; Jabbagy and Jackson 2000). Finer textured soils have concentrated pools of SOC associated with silt + clay size particles resulting in a greater concentration of SOC than soils of coarser texture, even when receiv ing the same amounts of OM inputs (Telles 2003; Torn et al. 1997). Soils have a SOC saturation capacity and are limited in the amount of C it can contain

PAGE 19

19 rate of SOC minera lization is inversely proportional to a saturation deficit in the silt + clay fraction. A saturation deficit is created when there is still opportunity for chemical complexes to be formed between SOC and silt + clay minerals. It has also been proposed that a soil has a maximum capacity to stabilize SOC by other inherent mechanisms, such as biochemical resistance, physical protection and even the unprotected pools of SOC (Six et al., 2002). In sandy Coastal Plain soils, the distribution of SOC among soil si ze fractions has been shown to decrease with size fraction. When silt and clay content is low, the saturation capacity of the fine minerals is quickly satisfied. Sarkhot et al. (2007b) found that the majority of SOC content in a forested Spodosol was locat ed in the 2000 250 m size fraction (40%), while the highest SOC concentration was found in the <53 m fraction ( 8% C). Sarkhot et al. (2007b) used d iffuse reflectance infrared F ourier transform s pectroscopy (DRIFT) to describe the chemical nature of the size fraction. The authors found a C rich 2000 250 m fraction containing large amounts of recently added OM. Material indicative of more decomposed material was found in the <53 m fraction along with high amounts of recently added materi al. Estimates of SOC mineralization in Coastal Plain Spodosols have been made. A study by Ahn et al. (2009) showed SOC mineralization rates ra nging from 3.4 to 10.6 mg C kg 1 d 1 in the top 30 cm of whole soil. However the mineralizability of SOC among so il size fractions in these soils has not been investigated leaving open the question, are the

PAGE 20

20 Within a size fraction, pools of SOC are incorporated into soil aggregates; ofte n considered a mechanism of SOC protection from mineralization. Sandy Spodosols in the Coastal Plain were once considered too sandy to have an aggregate hierarchy, but several recent studies have shown the opposite (Sarkhot et al. 2007b, Azuaje et al, 201 2). Macro sand size aggregates, defined here as aggregates >250 m in diameter, are composed of small microaggregates held loosely held together by short term bonding agents (e.g. roots and fungal hyphae, biotic secretions; Tisdall and Oades 1982 ; Oades an d Waters 1991 ). Sand size microaggregates (250 53 m) are a mixture of small microaggregates, occluded organic matter, and primary organomineral complexes. Microaggregates are typically held tightly together by persistent chemical bonds and biotic transien t binding (Tisdall and Oades 1982). Roots, fungal hyphae, and the secretions roots, hyphae a nd bacteria products are important in the formation of aggregates and the protection of occluded OM (Oades 1993). Previous studies on Spodosols supporting souther n pine species have used aggregate dispersion energy curves (ADEC) as a method to quantify aggregated SOC and to measure aggregate strength. Sarkhot et al. (2007a) reported that 45% of the SOC (0 10 cm soil depth) supporting a young loblolly pine ( Pinus ta eda L .) plantation was contained in aggregates. The authors also found that aggregates of different sizes had different dispersion energies. A later study, Azuaje et al. (2012), following similar methodology, found 40% of SOC aggrega ted in the top 30 cm of 10 20 y r old loblolly and slash pine ( Pinus elliottii L.) forests The purpose of this study was to address three objectives that are current gaps in our understanding of SOC dynamics in Coastal Plain Spodosols. The first objective

PAGE 21

21 was to investigate th e distribution and short term changes in whole soil OC and each with the intention of better defining short term (2.5 y ears ) SOC dynamics in a mid rotation, intensively managed loblolly pine stand. We hypothesized that, based on previ ous studies on similar soils, the larger size classes closest to the would be the most sensitive to short term changes under uniform management since it is presumably receiving the most detritus. The second objective was to study the speci fic mineralization potential of the SOC in each size fraction. The goal was to discover which size fractions showed the greatest vulnerability to short term microbial decomposition. Since no data currently exist on the mineralizability of SOC in soil size fractions for these sandy Coastal Plain Spodosols, we hypothesized that the largest size fraction would be expected to receive the freshest SOC and have the highest specific mineralization rate. These data would be useful in interpreting results in subsequ ent chapters. The final and third objective was to investigate the amount and strength of SOC residing in the large sand size fraction (2000 250 m) aggregates. Aggregates in this size fraction contain large portions of the total SOC and appear to be sensi tive to management changes (Sarkhot et al. 2007a and b). This objective defines the characteristic aggregation in a size fraction sensitive to management changes based on aggregate dispersive energy, also useful in interpreting results in subsequent chapte rs. Methodology Experimental Site The study site was located within a larger field study owned by R a yonier Inc. and managed by the Forest Biology Research Cooperative at the University of Florida. The

PAGE 22

22 the study site was provided by Roth et al. (2007). The research was accomplished in a flatwoods ecosystem of the l ower Coastal Plain region th at was planted to loblolly pine The area is a low lying ridge with deep, po orly drained, acidic soils. The soil was a Pomona fine sand (sandy, siliceous, hyperthermic Ultic Alaquods). A typical soil profile consists of a sandy, acidic surface soil with a spodic horizon from 60 to 90 cm and an argillic horizon from 130 to 180 cm. The water table is within 25 to 100 cm from the surface for 6 months or more during most years (Soil Survey Staff, 2011). During the study, the monthly mean temperature ranged between an average low of 13 C during January and February and a high of 27 C during June, July, and August (NCDC, 2011). At the initiation of the study, May 2007, the region wa s experiencing extreme drought and although these conditions improved over the duration of the study it was often considered to be abnormally dry conditions (U.S. Drought Monitor website, 2011). Annual mean precipitation documented at the Gainesville Regional airport, 13 k m from the study site, was 890, 1170, 1008, and 1191 mm for years 2006, 2007, 2008 and 2009, respectively and a long term annual mean of 12 42 mm for the years 1960 2010 (NCDC, 2011). The plantation was managed under an intensive regime of fertilizer and weed control for optimal growth. The stand was bedded following harvest and it received understory competition control and 660 kg ha 1 of 10 10 10 plus micronutrients at time of the planting. Understory control was managed for the first two years with applications of imazapyr and sulfometuron methyl at labeled rates and included follow up treatments as needed through age three. The stan tilization regime was prescribed based on

PAGE 23

23 annual foliar analysis. At the time of and for the duration of the study the total fertilizer additions were (kg ha 1 ) : N (745), P (180), K (161), Mg (56), S (56), Ca (50), Fe (38), Mn (5.4), Cu (3.5), Zn (2.2), a nd B (1.5). The planting density was 2990 stems ha 1 giving 128 trees per 19.5 m x 21.9 m plot The trees were planted at a narrower than normal tree spacing of 1.22 m x 2.75 m. Genetics of loblolly pine, i ntensive fertilization and narrow spacing was use d in the original Roth et al. (2007) to study the tree response and changes in soil quality when stands are managed for maximum production (see Roth et al. (2007) for the mixed loblolly plots) We selected these sites to study SOC changes because loblolly pine grown under these conditions is a fast growing species which produces large amounts of biomass and will be most likely to show changes in SOC and treatment effects with in the shortest amount of time. The experimental design was a random ized complete block design in stands of genetically elite full sib loblolly pine families planted in January 2000. Our study was replicated in three blocks and this experiment was designed to address changes in the untreated control plots. However, for litter input mea surements, it was necessary to also use the exclusion treatment plots, discussed in Chapter 3, where litterfall was measured periodically from these plots. The study plots were established in the interbed region of three center rows of the above described stands. In each interbed, untreated control and exclusion treatment plots, 0.5 m 2 area, were permanently marked with PVC tubes. For this experiment, all soil samples were taken from the untreated control plots and within this defined area. There were three sample replications from each block. The soils sampled from these plots represent the soil condition under the loblolly pine plantati on management described above. Canopy closure in these stands had been

PAGE 24

24 reach after approximately five years of growth At month 0 the trees were in their 7 th year of growth and by the end of the study, month 31, they were entering their 10 th year. Soil and Litterfall Sampling Soil samples at month 0 (May 2007) were taken from the control plot to establish the C content at the beginning of the study. Eight sample cores per treatment plot were the forest floor, separated into soil depths of 0 to 10 cm and 10 to 20 c m and combined for eac h block. Soil samples were subsequently collected from the control plot at 31 months. Three soil samples were collected with a 6.5 cm diameter aluminum cylinder to a depth of 20 cm after removing the forest floor. Each sample was separated into soil from t he 0 to 10 cm and 10 to 20 cm soil depths and soil from each depth was combined for each plot. Forest floor biomass was measured on the aboveground and above plus belowground exclusion treatment plots. At the initiation of the study the forest floor was removed from all of the 0.5 m 2 exclusion plots, oven dried at 70C and weighed. The plot area was then covered with shade cloth upon which subsequent litter fell. Litter was collected from all of the 0.5 m 2 exclusion plots periodically, oven dried and we ighed. An unanticipated litter harvest took place in the fall of 2008 by a commercial company that did not have permission to be operating the area. All litter was removed from the study areas including the untreated control plots. In order to recreate th e forest floor, forest floor material was gathered from an unraked loblolly stand of the same age and within a few hundred meters from the study plots. The initial forest floor mass

PAGE 25

25 measured in May, 2007 was used to reestablish the forest floor on the untr eated control plots. Laboratory Methods The same day the soil samples were collected, fresh soil weight was measured and sampled for moisture content. Bulk density was calculated from the soil volume and the weight of soil corrected for moisture content. The samples were subsequently air dried and passed through a 2 mm mesh sieve and the coarse fraction (>2 mm) was weighed and stored. One hundred g of air dry fine earth fraction soil was then dry sieved through a column of sieves for 5 minutes on a horizon tal shaker following the method of Sarkhot et al. (2007a). The sieves fractionated the soil into 2000 to 250 m, 250 to 150 m, 150 to 53 m and <53 m size fractions (Sarkhot et al., 2007a). Total organic carbon in the coarse size fraction, fine earth fra ction and in the four soil size fractions was measured by Loss on ignition (LOI) and then converted to organic C based with a pedotransfer function developed from 133 soil samples taken from a nearby watershed study and described in Azuaje et al (2012). Th e model created between LOI and SOC, where SOC% = 0.4922*LOI% + 0.065, had a correlation coefficient of 0.84. The LOI was accomplished by combusting 25 g soil sample at made LOI an appropriate method for estimating organic matter given that carbonates and loss of structural water were not i ssues (Nelson and Sommers, 1996). Mineralization of SOC in each size fraction less than 2 mm was accomplished by using the soils sampled at 31 months. Twenty grams of air dried soil from the three

PAGE 26

26 larger size fractions and five g of the <53 m size fract ion were brought to field capacity (0.33 bar tension) and placed in a microcosm made of high density polyethylene. The samples were brought to field capacity by first calibrating moisture content at field capacity for a subset of samples. This was done b y choosing samples that represented the range of organic matter found within size fractions. The soil size constant suction of 0.033 Mpa. The suction was applied until no more water was removed. A linear relationship was developed for each size fraction based on the LOI and the weight of water present in the soil at 0.033 Mpa: 0.033 Mpa water weight (g) = m*LOI (g) + b. This model was used to calculate the amount of water each m icrocosm required to maintain 0.033 Mpa. The soil fractions and water were well mixed and a liquid base trap of 0.25 M NaOH was securely placed inside the microcosm. They were then placed in a t the study by respiration (Anderson, 1982) was used to determine the rate of C mineralization. The alkaline traps were exchanged for fresh traps after weeks 1, 2, 4, 6, 10, 14, 18, and 23. It was at these times the moisture content was adjusted and ambient air refreshed the microcosm. The alkaline traps were titrated with 0.1 N hydrochloric acid (HCl) and the following equation was used to calculate C respired while the trap was confined in the microcosm: C (mg) respired = [(B T) x M x E]/ DF

PAGE 27

27 W here B is the volume in mL of HCl to titrate the blank aliquot, T is the mL necessary to titrate a sample aliquot, M is the normality of the titrant, E represents the equivalent weight of C (Anderson, 1982). Aggregate dispersion energy curves were developed on the 2000 to 250 m size fraction from soil sampled at 31 months. One curve was made for each of the three control plot replications in the three study blocks and two soil depths, 0 10 cm and 10 20 cm. The method used to create the ADECs was based on previous studies conducted on similar forested Spodosols as described by Sarkhot et al. (2007a) and Azuaje et al. (2012) usi ng a sonic dismembrator (Fisher Scientific, Model 500, Hampton, NH). Ten sub samples, five grams each were weighed into 250 mL beakers, then placed in a thermally insulated container and 100 mL of deionized water were added. The insulated container was pla ced in a sound minimizing enclosure and a sonic probe was submerged at a constant depth of 12 mm in the center of the container. One sub sample was exposed to a set energy applications, each subsequent subsample was subjected to increasing amounts of energ y. A total of 10 energy levels were used on each subsample, the energy ranged from 0 to 153 J mL 1 A pervious study by Azuaje et al. (2012) into the sonication method verified 153 J mL 1 was sufficient energy to completely break apart aggregates. To contr ol rising temperature caused by the increasing energy, energy was pulsed with 60 s econds on and 30 s econds off. The actual energy applied to the water was calculated using a calibration factor developed by Sarkhot et al. (2007a), following the recommendati ons of Schmidt et al. (1999)

PAGE 28

2 8 Upon completion of sonication, the sub sample was wet sieved through a 250 m mesh sieve. The portion that moved through the sieve was released by the applied energy and the portion that remained was AOC and SOC unbroken by the applied energy. The C contents of these two portions were measured by LOI and the quantity of SOC liberated at that specific energy level was calculated. Statistical Methods The population distributions of the data were analyzed and the SOC measurements for the coarse fraction, soil size class fractions, and CO 2 mineralization values were found to be l ognormally distributed. Data were log transformed prior to analysis. The log transformed data sets were evaluated for outliers and any value beyond three s tandard deviations from the mean was removed. For the purposes of tables and figures, the values were back transformed. A SAS PROC MIXED model (Littell et al., 2006) was used for statistical analysis of the soil bulk density, SOC, soil size class fraction C mineralization, and ADEC studies. Results are reported as LS Means and standard errors. In the analysis of the SOC data, soil depth and months of treatment were fixed effects and block was a random effect. In the mineralization study, soil size fraction soil depth, and mineralization days were fixed effects and block was a random effect. In the ADEC study, soil depth and dispersion energy were used as fixed effects and block was a random effect. A p value equal or less than 0.1 was considered a signific ant effect in the analysis of variance. When significant main effects and/or interactions were found, multiple mean comparisons and separations were identified using the ADJUST =TUKEY option of the LSMEANS statement.

PAGE 29

29 Results At age 7, the forest floor had accumulated a total of 12.7 Mg ha 1 in dry mass. The rate of litterfall statistically increased (p=0.019) as the stand aged from 7 to 10 years old. The total mean dry litter weight collected during the first 12 months was 5.7 Mg ha 1 and 9.6 Mg ha 1 during the last 12 months. This was a 68% increase in dry mass litterfall. Total litterfall over the entire 31 month study was 22.2 Mg ha 1 Objective 1. Determine Short Term Changes of SOC and its Distribution a mong Size Fractions Soil BD was significantly dif ferent (p=0.004) among depths where it was 0.93 g cm 3 in the 0 10 cm soil depth and increased with soil depth to 1.26 g cm 3 at 10 20 cm (Figure 2 1). The soil BD was used to express SOC on a soil volume basis. The SOC in the coarse soil fraction (>2 mm ) had a small but significant interaction between time and soil depth (p=0.023). Measurements taken at time zero were different at the two sampling depths. The 10 20 cm soil depth contained 3.1 mg cm 3 SOC and there was 2.1 mg cm 3 in the 0 10 cm depth. T he significant depth difference was lost a t 31 months, when a reported 1.0 mg cm 3 gain in SOC occurred at the 0 10 cm depth and a loss of 0.7 mg cm 3 SOC was found in the 10 20 cm depth. The same month by soil depth interaction was observed when SOC was e xpressed on a soil weight basis (mg g 1 ), with a 39% gain at 0 10 cm depth and a 52% loss at the 10 20 cm soil depth. At the end of the study there was 3.1 mg SOC cm 3 and 3.3 mg SOC g 1 at the 0 10 cm soil depth and 2.3 mg cm 3 and 2.0 mg g 1 at 10 20 cm. The SOC (mg cm 3 ) in the fine earth fraction (<2 mm) at the 0 10 cm soil depth experienced a significant 80% increase over the 31 month study period, while no change in SOC occurred at the 10 20 cm soil depth (Figure 2 2). The significant

PAGE 30

30 interaction bet ween soil depth and months in field treatment had a probably value of 0.088 in explaining the changes observed in mg SOC cm 3 soil The mean SOC content at the start of the study was 9.8 mg SOC cm 3 soil or 9.5 mg SOC g 1 on a soil weight basis. At the end of the study the mean SOC was 15.7 mg cm 3 or 17.5 mg g 1 Significantly more SOC was located in the 0 10 cm soil depth than the 10 20 cm depth (p<0.001). Means were 17.1 and 8.5 mg cm 3 by respective depths; and 20.0 and 7.0 mg g 1 Distribution of SOC among size fractions averaged across both sampling times generally decreased with decreasing size class and 80% more was found at the 0 10 cm soil depth than the 10 20 cm depth (Figure 2 3). The significant interaction between soil depth and fraction siz e had a p value of 0.008. Soil C was highest in the 2 000 250 m of the 0 10 cm depth and smallest in the <53 m of the 10 20 cm depth (Figure 2 3). There wa s a 79% difference in SOC between the highest and lowest C containing fractions, with the rest of th e fractions ranging between 51% and 70% lower than 2000 250 m of the 0 10 cm depth. The 2000 250 m of the 0 10 cm depth contained 44% of the total soil C in that depth and 30% at the 10 20 cm soil depth. Significant short term changes were observed acro ss all size fractions and soil depths during the 31 month study, months in field treatment main effect had a p=<0.001 (Figure 2 4). The largest SOC increase of 2.5 mg cm 3 was located in the 2000 250 m of the 0 10 cm depth and the lowest was 0.1 mg cm 3 i n the <53 m in the 0 10 cm depth (Table 2 1). The second highest gain in SOC was 1.6 mg cm 3 in t he 150 53 m fraction in the 0 10 cm depth. Apart from the <53 m fraction, significantly more total SOC was added to the in the 0 10 cm depth. The <53 m fra ction in the 10 20 cm depth

PAGE 31

31 added 12.7 times more SOC than in the 0 10 cm depth. At the end of the study it contained 21% of the total SOC at the 10 20 cm depth and 5% at the 0 10 cm soil depth. Objective 2. Assess the Potential Mineralizablity of SOC with in and a mong Size Fractions from Soil Collected d uring the 31 Month Sampling The 250 150 m fraction in the 0 to 10 cm depth expressed significantly higher specific mineralization ra tes than did the smallest two 0 10 cm fractions. All size fractions in the 10 20 cm depth were not different from each other or from the higher 0 10 cm depth mineralization rates (Figure 2 5). The SOC least vulnerable to microbial attack resided in the <53 m fraction of the 0 10 cm depth (Figure 2 5). The difference between the highest mineralization rate (250 150, 0 to 10 cm) and the lowest (<53, 0 to 10 cm) was 40%. When specific mineralization rates are applied to the SOC in each size fraction one can see the importance of mineralization from each fraction that applies to the whole soil (Figure 2 6). In this case the 0 10 cm depth had the most mineralizable SOC. Objective 3. Quantify the AOC and Evaluate the Stability of Aggregates in the Large Sand Size Fraction (2000 250 m) from Soil Collected during the 31 Month Sampling Aggregate OC in the 2000 250 m fraction was measured and expressed as a percent of the total SOC in that fraction (%AOC) (Figure 2 7). A small, but significant larger percentage of aggregated SOC was found in the 10 20 cm soil depth and it was concentrate d in the weak, water stable aggregates. The difference between soil depths was 1.5%, (soil depth p value = 0.072). Of the carbon present in this size fraction, 11% to 15% was released by wet sieving, depending on soil depth, leaving approximately 20% as wa ter stable aggregates. In total 35% of the SOC was incorporated into

PAGE 32

32 aggregates. The shape of the aggregate dispersion curve showed a stepwise release of aggregated SOC. There were two threshol ds of C release, one after 41 J ml 1 and the other after 57 J m l 1 There was no difference in aggregation between the two soil depths. Discussion Objective 1. Short Term Changes and Distribution of SOC Changes in the surface SOC were measured during this short window of stand development. As a whole, this ecosystem w as quickly aggrading C which was reflected in the increasing litterfall rate. Intensive management of genetically improved loblolly pines in the l ower Coastal Plain produce considerable quantities of biomass through its rotation (Martin and Jokela, 2004 ; V ogel et al. 2011 ) and the rate of net primary production is thought to be peaking during the developmental age observed in this study. In nearby stands of similar genetic quality, receiving similar management inputs, aboveground net primary production was reported by Martin and Jokela (2004) to be 1 yr 1 ) between ages eight and ten. In addition, belowground biomass accretion increases with age, first developing at and then expanding l aterally and vertically. Loblolly pine root development occupies a majority of the first 1 00 cm of soil within the first five y ears (Adegbidi et al. 2004). Coa rse root development is thought to progressively increase at y (Gholz et al. 1986) and the root system of a mature southern pine contributes approximately 20 2001; Vogel et al. 2011).

PAGE 33

33 In this current study SOC accretion was observed as well. Primarily located in the fine earth fraction at the 0 10 cm of soil ( Figure 2 2) these increases were substantial with an average annual rate of SOC accretion at 2.3 mg cm 3 yr 1 in the 0 20 cm soil depth In very similar soils with similar stand management of loblolly pines of the same ge netic quality, the annual rate of SOC accumulation was estimated at the end of the 5 mg cm 3 yr 1 (Vogel et al. 2011). The planting density used in our study was approximately two times as dense as the aforementioned study Althoug h still higher, these measurements were taken during a unique developmental phase and could represent a significant portion of the long term average. The SOC distribution with soil depth and among fractions was similar to other studies o n similar soils, b increasing with increasing soil fraction size (Figure 2 3) (Sar khot et al. 2007b; Haile et al. 2007). This exemplifies the concentration of fine roots and biological activity at the n Rees and Comerford 198 6; Gholz et al. 1986; Sword 1998; Fierer et reactive clays and oxides. In this short term observation, SOC additions were made across all siz e fractions and indicated that all fractions were capable of receiving and maintaining C additions in this short time frame (Figure 2 4 and Table 2 1) Although a major site disturbance occurred due to the unplanned commercial needle harvesting event, the harvesting process focuses on removing the fresh needle fall and not the more decomposed under layers These older, more decomposed layers were likely the primary source of C entering the mineral soil (see further discussion in the subsequent chapters ) F or this reason and the quick replacement of fresh needles,

PAGE 34

34 the unplanned litter raking event that occurred in this study likely had an insignificant effect on SOC development in the untreated control plots Objective 2. Assess the Potential Mineralizabili ty of SOC within and among Soil Size Fractions As an index of maximum potential SOC loss, the specific mineralization of C among soil size fractions indicated that the C contained in these fractions was generally of the same mineralizable quality, but it also appeared that some C may be influenced by other factors (Figure 2 5) Sarkhot et al. (2007b) found the 2000 250 m fraction at the 0 10 cm depth to be C rich containing large amounts of recently added OM, but our mineralization study indicated tha t t he C quality contained in this fraction at this depth was not different than nearly all of the other fractions at both depths. This was contrary to our hypothesis predicting the largest fraction, which likely receives the largest additions of fresh SOC, to have the highest specific mineralization rate. Portions of SOC in this fraction have also been identified as being contained in soil aggregates, which may be offering some C protection. The least mineralizable C was found in the <53 m fraction at the 0 10 cm soil (2007b) work. They found higher amounts of more decomposed and recalcitrant material along with high amounts of freshly added organic matter as well. In our stud y, this particular fraction at this depth experienced the smallest amount of net SOC addition which i ndicated that its sources of C were not abundantly available and wa s possibly being incorporated into other size class fractions. The soil depth differenc es of SOC accrual and the mineralizability of SOC in this fraction indicate that the processes occurring within these depths were different. Recent studies on SOC and aggregation

PAGE 35

35 suggest that the activity of microorganisms, fine roots, mycorrhizal roots, and hyphae are important factors in building and sustaining SOC pools (Rasse et al. 2006). The decreasing root density and microbial population often reported with soil depth is one explanation of the differences seen in this study. Bulk soil mineralizatio n rates of SOC respired from the various fractions were reflections of the total amount of SOC contained in the fractions and relatively homogeneous quality of the C located in the fractions (Figure 2 6) The sum of the bulk soil mineralization rates for a ll the fractions in the 0 20 cm depth, 9.5 mg kg 1 d 1 was at the upper end of the range reported by Ahn et al. (2009) for SOC located in the top 30 cm of whole soil in Coastal Plain Spodosols. The authors estimated the maximum potential SOC loss in bulk soil mineraliz ation ranged from 3.4 to 10.6 mg C kg 1 d 1 Objective 3. Quantify the AOC and Evaluate the Stability of Aggregates in the Large Sand Size Fraction (2000 250 m) The total amount of aggregated SOC in the 2000 250 m fraction was similar to the amounts reported by Sarkhot et al. (2007a) and Azuaje et al. (2012) in north Florida pine stands in similar soils and displayed the same evidence of hierarchal structure (Figure 2 7) The increasing AOC in this fraction with increasing soil depth may be an indication of a lower proportion of particulate organic carbon ( POC ) being added to the subsurface soil. It may also suggest th at POC at this depth wa s being preferentially mineralized over the AOC, thus indicating some SOC protection being provided by aggregation. Azuaje et al. (2012) investigation into AOC determined no protection was being provided by soil aggregation ; however that study was on a whole soil basis and not by s ize fraction where protection occurring in these defined SOC pools c ould be masked. If

PAGE 36

36 aggregation wa s offering protection to SOC, it wa s in small quantities and will require investigations into the mineralizability of AOC within physically defined fractions to reveal its influence. Conclusion During this short term study sig nificant SOC accretion was measured in a quickly aggrading, mid rotation southern pine plantation. The SOC additions primarily occurred in the fine earth fraction at the 0 10 cm soil depth, which experienced an average annual SOC accretion rate of 2.3 mg cm 3 yr 1 A net increase of SOC was made in all of the size fractions and indicated that all of the fractions were capable of receiving and maintaining short term C additions. Soil depth was an important factor in many of the SOC characteristics measured in this study. This exemplifies the significant effects of SOC dynamics occurring at the interface of the forest floor and the surface soil where fine root and microbial activities are concentrated. The distribution of SOC was typical of a sandy soil, dec reasing with soil depth and fraction size. The mineralizable quality of the C among size fractions was lowest in the smallest size fraction (<53 m) in the 0 10 cm soil depth, while C mineralizability was unaffected by fraction size in the 10 20 cm soil de pth. Increasing homogeneity of SOC mineralizability with soil depth may be a result of the decreasing influence of fine root and microorganism activity. Sand size macroaggregates (2000 250 m) were slightly more concentrated in the 10 20 cm soil depth but the reason for this wa s not clear. Particulate OC may be added at a lower proportion in the subsurfa ce soil, increasing the percentage towar d AOC. An alternate hypothesis wa s that POC at this depth is being preferentially

PAGE 37

37 mineralized over the AOC. Soil a ggregation in sandy soil is not thought to offer much protection against mineralization, but additional studies are required to fully appreciate the scope of protection or lack thereof provided by aggregation.

PAGE 38

38 Figure 2 1. Differences in soil bulk dens ity in the untreated control plots at two soil depth s of an Ultic Alaquod supporting an intensively managed loblolly pine stand in north central Florida Soil bulk density in the untreated control plot did not significantly change during the study which sp ans the 7 th through 9 th year of tree growth Values are LS means and bars represent standard error. Unequal letters indicate significant differences with a p value=<0.1 Figure 2 2. Differences in SOC in the fine earth fraction over time for two soil depths in an Ultic Alaquod supporting an intensively managed loblolly pine stand in north central Florida Changes were observed during the 7 th through 9 th years of tree growth. Values are LS means and bars represent standard error. Unequal letters indicate significant differences with a p value=<0.1 a b bc bc

PAGE 39

39 Figure 2 3. Distribution of SOC among soil size fractions averaged across time for two soil depths in an Ultic Alaquod supporting an intensively managed loblolly pine stand in north central Florida Changes were observed during the 7 th through 9 th years of tree growth. Values are LS means and bars represent standard error. Unequal letters indicate significant differences with a p value=<0.1 F igure 2 4 Differences in soil size fr action SOC ov er time in an Ultic Alaquod supporting an intensively managed loblolly pine stand in north central Florida Changes were observed during the 7 th through 9 th years of tree growth. Each size fraction significantly increased with time. Values are LS means and bars represent standard error with a p value=<0.1 Size Fractions Size Fractions

PAGE 40

40 Figure 2 5 Differences in s pecific mineralization rates of cumulative C respired from size fraction SOC at two soil depths, after 162 d laboratory incubation at 30 C The s oil is in an U ltic Alaquod supporting a ten yr old intensively managed loblolly pine stand in north central Florida Values reported are LS means and standard error. Unequal letters indicate significant differences with a p value=<0.1 Size Fractions

PAGE 41

41 Soil size fraction Figure 2 6. Differences in bulk soil C respired from four soil size fractions, at two soil depths, after 162 d laboratory incubation at 30 C. The soil is in an Ultic Alaquod supporting a ten yr old intensively managed loblolly pine stand in north central Florida Values reported are LS means and standard error. Unequal letters indicate significant differences with a p value=<0.1 b c b a e d de de Size Fractions

PAGE 42

42 Figure 2 7. Dispersion of AOC as % of the 2000 at two soil depths, as a function of increasing applied sonic energy The soil is in an Ultic Alaquod supporting a ten yr old intensively managed loblolly pine stand in north central Flo rida Values reported are LS means and standard error. Unequal letters indicate significant differences with a p value=<0.1 Table 2 1. Mean total SOC increase a mong the soil size fractions at two soil depths from an Ultic Alaquod supporting an intensely ma naged loblolly pine stand in north central Florida Data presented as total increase amount in mg cm 3 and as relative percent change that occurred over the 31 month study period which span the 7 th through 9 th yrs of growth Size Fraction 2000 250 m 250 150 m 150 53 m <53 m Soil Depth (cm) 0 10 10 20 0 10 10 20 0 10 10 20 0 10 10 20 SOC (mg cm 3 ) 2.46 1.28 1.15 0.16 1.62 0.61 0.11 1.27 % Change 36% 57% 39% 7% 51% 27% 4% 110% b b c c c

PAGE 43

43 T able 2 2. ANOVA table of the specific minera lization rates of cumulative C respired from size fraction SOC, at two soil depths, after 162 d laboratory incubation at 30 C. The soil is in an Ultic Alaquod supporting a ten yr old intensively managed loblolly pine stand in north central Florida Signifi cant effects were determined with a p value=<0.1 Effects df F value p Soil Depth 1 0 0.2676 Size Fraction 3 1.35 0.955 Soil Depth x Size Fraction 3 2.99 0.038 Block 2 2.07 0.135 Error 61 Table 2 3. ANOVA table of soil size fraction bulk mineral ization rates of cumulative C respired after 162 d laboratory incubation at 30 C. The soil is in an Ultic Alaquod supporting a ten yr old intensively managed loblolly pine stand in north central Florida Significant effects were determined with a p value=<0.1 Effects df F value p Soil Depth 1 99.59 <.001 Size Fraction 3 29.07 <.00 1 Soil Depth x Size Fraction 3 9.03 <.001 Block 2 21.67 <.001 Error 62

PAGE 44

44 CHAPTER 3 THE CONTRIBUTIONS OF ABOVE AND BELOWGROUND SOUR CES OF C SUSTAINING SURFACE S OIL ORGANIC C IN AN INTENSIVELY MANAGED LOBLOLLY PINE STAND I ntensively managed southern p ine plantations in the United States have most studies show soil organic carbon (SOC) remains unaffected or negatively affected by these practices (Shan et al. 2001; Legge tt and Kelting 2006). Sources of organic matter that maintain and/or increase SOC pools in these ecosystems originate from both above and belowground sources and are primarily plant detritus. However, few studies have examined the mechanisms behind changes in soil C in these ecosystems. In particular, it is unclear what the magn itude of C input to SOC results from litterfall inputs or fine root inputs Understanding the processes of SOC development and the dynamics of turnover are critical in order to antic ipate effects of environmental and land management changes. Fer tilized southern pines planted o n nutrient poor soil s have prod uced considerable increases in tree biomass (Martin and Jokela 2004; Jokela et al. 2010 ; Vogel et al. 2011 ). This includes annual litterfall rate and coarse root production. By the end of the rotation a well fertilized loblolly pine ( Pinus taeda L .) can accrue 40 50 Mg of C ha 1 in the forest floor depending on management inputs (Voge l et al. 2011). The accumulation of mass in the fo rest floor is influenced by the long residence time characteristic of pine litter and minimal biotic mixing of litter into the soil (Thomas 1968; McBrayer et al. 1977; Piatek and Allen 2001). The forest floor also provides habitat for soil fauna that migr ate between the litter layer and the surface soil. Their activities accelerate the decomposition of forest floor

PAGE 45

45 and roots, they add and may mix organic material into the soil ecosystem, and their waste and remains stimulate soil flora (Abbott 1989; Lussen hop 1992; Knoepp et al. 2000). Most aboveground litter is thought to be utilized by microorganisms that produce enzymes which digest this recalcitrant material (Jorgensen et al. 1980; Berg and McClaugherty 2003; Bardgett et al. 2005). As the trees age and the forest floor develops to more advanced stages of decomposition, it becomes increasingly populated by fine roots and mycorrhizal fungi (Gholz et al. 1986; Ponge 1991). Southern pines s uch as loblolly and slash pine ( Pinus elliottii L) are tap rooting species, yet the majority of nutrient absorbing fine roots are found within the A horizon (Gholz et al. 1986 ; Retzlaff et al. 2001 ; Adegbidi et al. 2004 ). Root biomass develops first in the surface soil and quickly expands horizontally and vertically. A st udy on early loblolly pine root development determined that by age 4 years, roots had occupied 60% of the soil to a 100 cm depth (Adegbidi et al. 2004). Throughout the life of the stand, roots constitute a significant portion of the total tree biomass. Coa rse root development is thought to progressively increase at a lin ea 1986). The lateral root system of a mature loblolly pine can contribute 15 20% of a ff et al. 2001; Shan et al. 2001 ). In early stages of development an expanding root biomass. In maturity, mu ch of the belowground primary production is dedicated to sustaining the large root system (Gholz et al. 1985; Wiseman and Se i ler, 2004). In a study of 17 y r old slash pine receiving fertilization and understory control, 10% of the total annual primary production was devoted to fine root production, which only constituted about 2% of the total tree biomass (Shan et al. 2001). On

PAGE 46

46 average fine roots have the quickest turnover rates within the root system but studies have determined that not all fine roots of the same diameter have the same structure, function and tur nover rate (Pregitzer 2002 ; Baddeley and Watson 2005 ). The use of radiocarbon dating of fine root s suggest that many small fine roots live longer than a year and for pine species several years or more (Gaudinski et al. 2001 Matamala et al. 2003 ). It has also been shown that fine roots colonized by mycorrhizal fungi hav e increased root longevity and a decreased decomposition rate. This may be occurring because colonized roots are less sensitive to desiccation and their structures are encased in chitin, a recalcitrant compound (Eissenstat et al. 2000; Fan and Guo 2010). E ctomycorrhizal root tips of pine are often surrounded by aggregated organic matter enmeshed by fungal hyphae (Burke et al. 2006). Organic matter concentrates around sources of rhizodeposition. Exudates produced by fine roots and associated mycorrhizal hyph ae along with their delicate tissues are continually being consumed by soil macro and microfauna and soil flora (Knoepp et al. 2000). Microbial biomass residue has been identified as an important source of material found in humus and an agent of soil or ganic matter stabilization (Simpson et al. 2007 ; Miltner et al. 2011.) Contributions of the total C input from various sources into the surface soils supporting loblolly pine plantations are influenced by stand conditions (i.e. genetic quality, stand age, planting density, site fertility, and light and water availability) (Albaugh et al 1998; King et al. 1999 ; Burkes et al. 2003 ). The contributions can be segregated into C sources originating from aboveground and belowground sources Reported annual litte rfall rate at mid rotation loblolly pine stand range is between 3 and 7 Mg ha 1 depending on management inputs (Jokela and Martin, 2000). With little soil

PAGE 47

47 mixing occurring in these soils, the primary C contribution made by the forest floor is presumed to be through DOC leaching. By mid rotation, the forest floor develops distinct Oi and Oe horizons. Polglase et al. (1992 a ) reported total labile C concentration in the Oi and Oe layers as 14% and 6% by weight, respectively. Fresh needlefall composing the Oi layer is a significant source of inorganic phosphorus and may accelerate the decomposition process particularly in fertilized stands (Polglase et al. 1992b ; Sanchez 2001 ). Loblolly needle decomposition studies have found approximately 25% of the C is rele ased within the first year, followed by a long period of slow release (Jorgensen et al. 1980). They are also In a mature southeastern loblolly forest, annual dissolved OC inputs from forest floor to the surface 15 cm have been reported to add 3.2 mg C cm 2 (Richter et al., 1999). For the same soil dep th, the same authors estimate d an annual contri bution of 6.3 mg C cm 2 from rhizo deposition of roots with <2mm diameter, (authors assumed a 50% turnover rate in biomass and root C content of 50%). The largest C input belowground comes from fine root turnover ; and as southern pines mature, the production of fine root biomass reaches a steady state with ephemeral patterns of seasonal growth (Gholz et al. 1986; King et al. 2002). Fine root biomass estimates (diam eter <0.5 mm) in young, densely planted loblolly have been estimated (Burkes et al. 2003) to c ontribute 8.3 mg C cm 2 year 1 to the first 15 cm of surface soil in a 4 y r old stand. Currently it is unknown how the various sources of C contribute to the net formation and maintenance of SOC in these forest ecosystems. In particular, it is unclear what the magnitude of C input to SOC is as a result of aboveground litter inputs

PAGE 48

48 or root turnover and belowground processes. The objective of this study is to estimat e the relative contributions of above and belowground sources of C formation or enhancing the SOC in an in tensively managed mid rotation loblolly pine plantation growing in sandy soils of the l ower Coastal Plain. We hypothesized that belowground sources of C dominate over aboveground inputs for the maintenance and development of surface SOC. Belowground sources of C have several advantages for developing into SOC. Specifically the aggregating effects created around fine roots are seen as an important mech anism f or stabilizing C in these sandy acidic soils. Also, the large amounts of C input by belowground biomass and its long turnover times suggest it plays a dominating role in creating and sustaining SOC. Methodology Field Treatments This portion of the study was performed in the same stands of replicated blocks described in Chapter 2. A randomized complete block study design comprised of three study blocks with three replications of each field treatment per block. Within each block, the study was carried out in the middle of three center rows and utilized the interbed region to randomly place one replication of each of the three treatments. Each treatment plot was a 0.5 m 2 and permanently marked with PVC tubes. The two treatments beyond the untreated cont rol were designed to eliminate sources of C inputs and prohibit them from entering the soil. The first treatment excluded aboveground litter and prohibited any future C addition from this source. This was accomplished by removing the entire O horizon

PAGE 49

49 pres ent at the initiation of the study and keeping this bare soil condition for the entirety of the study. In addition to removing the O horizon from the designated treatment plot area an extra 0.5 m surrounding the plot vicinity was also removed in order to reduce All understory vegetation present was cut at the soil surface and removed. The total removal area was 2 m 2 and OC was removed leaving only the bare mineral soil surface. Shade cloth, 1 m by 2 m, was placed over the bare soil and held in place with metal stakes. Monthly to bimonthly the aboveground litter on the surface of the shade cloth was removed and any sprouting vegetation, fungi or moss present under the shade cloth was also removed. The second treatment excluded above plus belowground litter additions. This treatment used the same method as the first treatment to eliminate C inputs from above with the addition of trenches that surround ed the treatment plot and prohibited new root growth from entering the plot. This was accomplished by digging a 40 cm deep trench 0.5 m from the outline of the treatment plot, lining it with aluminum sheet metal and back filling with soil. The treatment plot was then covered with shade cloth tacked in place with metal stakes The aboveground litter on the surface of the shade cloth was removed monthly to bimonthly and any sprouting vegetation, fungi or moss present under the shade cloth was also removed. The final treatment was an untreated control. A 0.5 m 2 area was marked i n the interbed and left to represent the natural soil state experiencing the full influence from the O horizon and belowground root growth.

PAGE 50

50 All soil samples were taken from the 0.5 m 2 plot area. There we re three replications of each tre atment in each of t hree blocks. In total, there were nine replicates of each field treatment across all three blocks. Soil Sampling Soil samples at time 0 were taken to establish the C content at the beginning of the study. After removing the forest floor, eight sample core s per treatment plot were into soil depths of 0 to 10 cm and 10 to 20 cm. The soils were combined for each treatment by block. Soil samples were subsequently collecte d at 31 months from each plot. Three soil samples were collected with a 6.5 cm diameter aluminum cylinder to a depth of 20 cm after removing the forest floor. Each sample was separated into soil from the 0 to 10 cm and 10 to 20 cm soil depths and combine d for each plot. Live Root Observations To ensure the root exclusions were successful at eliminating new root growth, the distribution of fine roots was evaluated at the end of the study in all of the treatment plots. A field method described by Escamilla et al (1991) for similar soil s was used. The method wa s based on the relationship between fine root biomass, root length density, and the number of roots crossing a two dimensional plane. A vertical plane of soil was exposed in each treatment plot by tre nching. Fine roots protruding from the soil face were counted in a 30 cm horizontal area at 5 cm vertical increments to 20 cm. T he root exclusion trenches were effective S tatistical analysis was run on the live root counts taken at the end of the study showed a significant interaction between field treatment and soil depth, p=0.001(Figure 3 1). At both soil depths the above plus

PAGE 51

51 belowground exclusion plots significantly reduced the presence of live roots relative to the control with a reduction of 75% a nd 92% at the 0 10 cm and 10 20 cm soil depths, respectively. Laboratory Methods The same day the soil samples were collected, weight was measured and sampled for moisture content. Bulk density was calculated from the soil volume and the weight of soil was corrected for moisture content. The samples were subsequently air dried and passed through a 2 mm mesh sieve and the >2 mm fraction was weighed and stored. Total organic carbon in the whole soil and in the >2 mm size fraction was measured by Loss on igni tion (LOI) and then converted to organic C based on a pedotransfer function develo ped from 133 soil samples from a nearby watershed study (Azuaje et al. 2012; see Chapter 2). The LOI was accomplished by combusting 25 g t least 15 hours). The low soil pH and sandy nature (>95% sand) made LOI an appropriate method for estimating organic C given that carbonates and loss of structural water were not issues (Nelson and Sommers 1996). Statistical Methods The relative percent change over time was used to analyze the short term changes in bulk density, SOC in the coarse fraction, fine earth fraction, and soil size class fractions, minimizing the uneven variability among treatment plots (Bonate 2000 ). The relative percent change wa s defined as the difference between the variable at the end of the study, 31 months, and time zero. The population distributions of the data were analyzed and the relative percent changes in SOC measurements for the coarse

PAGE 52

52 fraction and fine earth fractio n values were found to be lognormally distributed. Prior to analysis these data were log transformed, in order to make the log transformation a constant value of one was added to the relative change in SOC to create all positive data points (McDonald 2009 ). All data sets were evaluated for outliers and any value beyond three standard deviations from the mean was removed. For the purposes of tables and figures, the data were back transformed. SAS PROC MIXED model (Littell et al. 2006) was used for statisti cal analysis of soil bulk density and SOC. Results are reported as LS Means and standard errors. In the analysis of the root counts, bulk density, and SOC data soil depth and months in treatment were fixed effects and block was a random effect. A p value e qual or less than 0.1 was considered a significant effect in the analysis of variance. When significant main effects and/or interactions were found, multiple mean comparisons and separations were identified using the ADJUST =TUKEY option of the LSMEANS sta tement. Results At the end of the 31 month study period the field treatments had significantly was highest in treatment plots that excluded above plus belowground sou rces of C and lowest in the u ntreated control plots, which wer e open to all sources of C inputs (Figure 3 2). Although the field treatments significantly affected bulk density, the changes were not significant increases or decreases, as they all contain ed 0% change in the 90% confidence interval and significant change cannot be determined ( Appendix B 1) Soil depth was also a main effect in determining the relative percent change in bulk density

PAGE 53

53 (p=<0.001). Across all treatments, soil bulk density significa ntly increased in the 0 10 cm soil depth, while it significantly decre ased in the 10 20 cm soil depth ( Appendix B 1 ). High variability created a large margin of error re ported in the relative change in soil bulk density. Among treatments, the effect of soi l depth on the relative change in the coarse fraction SOC ( mg cm 3 ) fraction after 31 months was significant (p=0.001). The analysis revealed that SOC accumulated in the 0 10 cm soil depth, which increased by 26%. Soil OC in the 10 20 cm depth decreased; t he mean relative change was 34% (data not shown). Treatment was also a significant main effect, p=0.073, as the aboveground exclusion mean was significantly higher than the above plus belowground exclusion treatment The above plus belowground exclusion significantly decreased during the study but no t significant change occurred in the untreated control or aboveground exclusion plots. Both exclusion treatments were no different than the untreated control (Figure 3 3 an d Appendix B 2 ). The relative perc ent change of SOC (mg cm 3 ) contained in the fine earth soil fraction was significantly affected by excluding C inputs (p=0.014). The above plus belowground exclusion treatment plots experienced no increase in SOC content and were significantly different than the untreated control plots (Figure 3 4). The untreated control plots experienced a mean increase of 42% over the 31 month study period. Soil OC for the aboveground exclusion plots increased by 14% and w as not statistically different than the untreate d control. The 14% increase in SOC was not a significant increase as the 90% confidence interval contained zero percent change and was also no t different than the above plus belowground exclusion treatment plots (Figure 3 4

PAGE 54

54 and App endix B 3 ). Soil depth was a significant effect in predicting the relative change of SOC in the fine earth fraction (p=<0.001). The majority of SOC accrued during the study was in the 0 10 cm soil depth. The mean relative percent increase at this depth was 4 2%, while the SOC in the 10 20 cm soil depth went relatively unchanged ( Appendix B 3). Discussion The larg e inherent spatial variability commonly found in the SOC distribution of forest soils and can make precise measurements and detection of change difficult. A coefficient o f variation of 20% in forest soil C measurements is common ( Haines and Cleveland 1982 ; Richter et al. 1999 ; Conant et al. 2003 ). By repeatedly sampling soil within small sample plots, the ability to detect changes in SOC over short periods of time is incre ased (Conant et al. 2003). Within the short time frame of this study, evidence of SOC accretion occurring in the untreated surface soil was detectable in the fine earth fraction. This was a product of the fast growth and biomass production taking place at this stage of stand development and the high planting density A mid rotation loblolly pine stand aggrades C at a higher rate than during any other phase of its development. Similar stands with half the planting density have been measured to produce 22 Mg ha 1 yr 1 of aboveground net primary production on very similar soils during this phase (Martin and Jokela 2004). A form of a substantial forest floor and as fine root production in the surface soil. At the beginning of the study, 13 Mg ha 1 of dry mass had accumulated i n the forest floor. In all approximately 22.2 Mg ha 1 fresh litter was added to the forest floor over the 31 month

PAGE 55

55 study period. The mean annual additio n was 8.6 Mg ha 1 but it was not evenly distributed throughout the study, as the litterfall rate increased with time. Aboveground litter included needles and some coarse woody debris. Loblolly needles have been measured to release approximately 25% of the C within the first year, followed by a long period of slow release (Jorgensen et al. 1980). In this study, the maximum ann ual C release from fresh needle fall would have been approximately 10.8 mg C cm 2 This assumes that the en tire aboveground litter mas s was needles, which is an overestimation as a small portion of coarse woody debris was collected with the needlefall. Mature loblolly and mixed hardwood coniferous forest C studies ha ve reported approximately 2.5 mg C cm 2 of DOC leaching annually from th e bottom of th e O horizon and approximately < 2 mg C cm 3 in the 0 15 cm soil depth, which then sharply decreases with soil depth (Richter et al. 1995, 1999; Dosskey and Bertsch 1997). Belowground inputs are more difficult to measure. Shan et al. (2001) es timated 10% of the total annual primary production dedicated to fine root production in southern pine managed with intensive fertilization and understory control. If the aboveground biomass accumulation rate in this current study followed the same trend as those described by Martin and Jokela (2004), annual fine root net primary production would be estimated at 28.8 mg cm 2 During this phase of development, if half of the fine root production turns over, 14.4 mg C cm 2 would be added annually to the soil. With fine root biomass concentrated in the surface 20 cm of soil it is reasonable to a ssume 0.7 mg C cm 3 annual C additions occurred from fine root turnover. This is assuming root C content of approximately 50% of the biomass and 22% more fine root total biomass in a

PAGE 56

56 1.6 times more densely planted loblolly pine stand (Burkes et al. 2003). As a result we estimated a 43% increase in SOC in the fine earth fraction, prim arily at the 0 10 cm soil depth (Figure 3 4) The annual mean SOC a ccretion at this site w as 2.3 mg C cm 3 yr 1 which falls within the range estimated in a comparable stand (Chapter 2; Volgel et al. 2011). When the forest floor was excluded from adding C to the soil, the SOC was not significantly different than the untreated control plots. Othe r forest soil studies that used prolonged O horizon removal reported similar results for several temperate deciduous forest ecosystems (Nadelhoffer et al. 2004 ; Garten 2009) However, the soil trenching method was very effective in preventing new fine root growth. By excluding belowground inputs these treatment plots did not experience the SOC accretion that occurred in the control plots, indicating a primary dependence on subsurface processes. Short term effects of eliminating above and belowground inpu ts of C resulted in higher soil bulk density and a loss of SOC in the coarse fraction. Related studies that compared management strategies of southern pines have suggested that roots play a dominant role in SOC content. The absence of understory roots in s tands managed with sustained understory competition control is often given as the probable reason for the lower SOC content (Shun et al. 2001; Echeverra et al. 2004; Rifai et al 2010; Vogel et al. 2011). However, there has been no research to directly ad dress this issue in loblolly pine ecosystems until now. Although it was determined that short term SOC accretion and maintenance in this forest was principally dependent on subsurface processes, it cannot be ignored that the aboveground exclusion plots we re not statistically different than the above plus

PAGE 57

57 belowground exclusion either. Many litter removal and pine litter raki ng studies report SOC content was not affected by these treatments ( Ross et al. 1994 ; Nadelhoffer et al. 2004 ; Blazier et al. 2008 ; Ga rten 2009 ). Blazier et al. ( 2008 ) reported a small but sign ificant loss of SOC in the surface five cm of soil after seven years of annual litter raking and application of inorganic fertilizers in a loblolly pine, while the unfertilized stands reported no l oss in SOC with annual litter raking The SOC loss was possibly a result of increased decomposition due to repeated applications of N (Khan et al. 2007). However, the forest floor may not be supplying the soil with as large a quantity of C as once thought particularly in younger forests. A disconnection between the C of recently added forest li t ter and the mineral soil has been observed in Tennessee pine hardwood forests (Froberg et al. 2007b; Kramer et al 2010). Tracing the decay of C 14 released into the atmosphere from a nearby incinerator, the authors determined after four yea rs of various manipulations of C 14 enriched C additions that 14% of the DOC in the top 15 cm originated from the C 14 forest floor litter (Froberg et al. 2007b). Further studie s revealed <6% of the recent litter C had accumulated into the soil microbial biomass. Si milar results were found with C 13 isotope labeling to trace DOC from fresh C into the soil (Hagedorn et al. 2004). Hagedorn et al. (2004) found that 5 10% of the labe led DOC at the 5 10 cm soil depth had originated from fresh litter and recent rhizodepo sition. In addition, they reported that recently deposited DOC was preferentially mineralized in a soil incubation study. The majority of DOC that reached the mineral so il surface is thought to originate from the more decomposed Oe horizon, while more recently released DOC is mineralized in the forest floor (Froberg et al. 2007a). Results of these findings correspond with radiocarbon data that identified the

PAGE 58

58 age of DOC in foreste d surface soils around 20 30 y rs and its presence in the soil related to microbial processes (Trumbore et al. 1992; Guggenberger et al. 1994 ; Tegen & Drr 1996 ) It is possible, to some degree that nutrient cycling in the forest floor is di scoupled from the mineral soil in southern pine ecosystems. By mid rotation, the forest floor in loblolly pine has been reported to im mobilize nitrogen and phosphoru s (Piatek and Allen 2000). This may be an indication of a microbial population in the forest floor utilizing the labile C released from the fresh litter inputs. In addition, litter manipulation studies have found fine roots to be unaffected by litter removal in tropical and temperate coniferous forests (Sayer et al. 2007; Okada et al. 2011) but it has been found to decrease in the presence of soil fauna and ectomycorrhizal hyphae in the surface soil (Ponge et al. 1993; Okada et al. 2011). Soil fauna fecal pellets have been observed in these soils as a common constituent of the SOC content (Azuaje et al. 2012). Further investigation s will be required to determine if the absence of the forest floor continues to impede SOC formation and maintenance in this ecosystem. Conclusion Annually large amounts of C are released from the forest floor as DOC and fine root t urnover in aggrading loblolly pine ecosystem s The purpose of this study was to determine the significance of the C contributions from above and belowground sources to SOC development. It was concluded that SOC formation and maintenance was signific antly dependent on belowground sources of C, but the large inherent spatial variability of forest SOC made it difficult to establish the role of the forest floor. The forest floor provides several conditions and opportunities to have an influence on SOC

PAGE 59

59 de velopment in the surface soil. It provides a habitat for soil fauna, supports ectomycorrhiz al hyphae growth, and leaches significant amounts of DOC. How much of the DOC from the forest floor is reaching the soil matrix and influencing SOC development is in question and requires further investigation to determine it role in these forest ecosystems. Individually or in combination, these factors could be having a short term impact on the SOC content. Long term observation will be required to determine if this impact continues, although litter raking and litter removal studies suggest it will not reduce SOC in the next 5 years.

PAGE 60

60 Figure 3 1. Evaluation of the effectiveness of root exclusion treatment through live root counts. A vertical profile inside ea ch of the treatment plots was exposed and the numbers of live roots were counted and recorded in 30 cm across x 10 cm depth area increments. Counts were made January 2010, in a ten yr old intensively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida after the last soil sampling of the study Figure 3 2. Differences in % change in soil b ulk density after to 31 months of exclusion treatment in an intensively managed loblolly pine stand growing in a n Ultic Alaquod in north central Florida The study spans the 7 th yr through the 9 th yr of growth. Values are LS means of relative percent change with standard error bars. Different letters signify mean separations with a p value=<0.1 a a b b b c Untreated control Above ground exclusion Above plus below ground exclusion Untreated control Above ground exclusion Above plus belowground exclusion

PAGE 61

61 Figure 3 3. Diffe rences in % change of the coarse fraction (>2 mm) after 31 months of exclusion treatment in an intensively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida The study spans the 7 th yr through the 9 th yr of growth. Values are LS means of relative percent change with standard error bars. Different letters signify mean separations with a p value=<0.1 Untreated control Above ground exclusion Above plus belowground exclusion

PAGE 62

62 Figure 3 4. Differences in % change of the fine earth fraction (<2 mm) after 31 months of exclusion treatment in an inte nsively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida The study spans the 7 th yr through the 9 th yr of growth. Values are LS means of relative % change with standard error bars. Different letters signify mean separation s with a p value=<0.1 Untreated control Above ground exclusion Above plus b elow ground exclusion

PAGE 63

63 CHAPTER 4 THE RELATIVE IMPROTA NCE OF CARBON SOURCE S CONTIRBUTING TO SO IL ORGANIC CARBON POOLS IN AN INTENSIVLY MAN AGED LOBLOLLY PINE STAND Intensively managed loblolly pine ( Pinus taeda L .) forests are prod uctive and commonly planted o n southeastern U.S. Coastal Plain Spodosols. Despite frequent fresh carbon (C) additions of aboveground leaf litter and subsurface root turnover, soil organic carbon (SOC) storage has been considered limited and not a significant factor for C sequestrati on management (Shan et al. 2001). The warm/humid climate and very acidic nature has weathered the soil, creating a sandy, weak structured surface horizon (Soil Survey Staff 2011) and a clay fraction primarily consisting of quartz and some kaolinite (Harris and Carlisle 1986). Ultimately, large portions of SOC are vulnerable to persistent microbial consumption. The two major sources o f C entering soil in a loblolly pine forest are from aboveground plant biomass C and belowground root/hyphae C. Each has its own physiochemical and distribution properties that influence its potential to form SOC pools order for this material to contribute to SOC it must be reduced in size th rough decomposition and incorporated into the soil. Soil m ixing and incorporation of aboveground C residues is a function of soil fauna (Brussaard et al 1997). In the absence of appropriate soil fauna, aboveground litter has little opportunity to participa te in SOC forming processes (Phillips and Fitz p atrick 1999). Carbon from the forest floor can also enter soil as dissolved organic C, although C cycling between the forest floor and mineral soil may not always be significantly coupled ( Piatek and Allen 200 0 ; Froberg e t. al. 2007; Kramer et al. 2010 ).

PAGE 64

64 In contrast, belowground sources of C are produced and enter directly into the soil matrix primarily as residues of root and mycorrhizal hyphae turnover, root and microflora exudations, root cap cells, and roo t epidermis (Farrar et al. 2003; Nguyen 2003 ; Godbold et al. 2006 ). Root residues persist longer in soils than shoot residues (Bird et al. 2007; Kong et al. 2010) and root turnover has been seen to be intimately involved in the creation of sand size soil aggregates (Gale et al. 2000). Several forest soil studies have used litter exclusion to study SOC formation and concluded that SOC depends more on belowground C inputs than the aboveground forest floor (Nadelhoffer et al. 2004 ; Garten 2009 ). A previous l itter exclusion study in a managed southern pine Spodosol found belowground C to be the dominant source for the formation of SOC (Chapter 3), but the influence of C sources on different SOC pools in Coastal Plain Spodosols was not investigated. The purpose of this study was to identify the importance of C sources to the short term assimilation, maintenance and stability of SOC pools in a Spodosol supporting an intensi vely managed loblolly pine stand This land cover and soil type define large areas in the s outheastern U.S.A. In this region l oblolly pine an important commercial species, covers approximately 12 million ha in planted and natural stands (Baker and Langdon 1990) and Spodosols a dominate soil order occupies an estimated 5.7 million ha (Adegbid i et al. 2002) Yet, until the importance of the different C sources is defined our capacity to predict SOC responses to management and climate change will be limited. Physical size fractionation of these sandy soils has been successful in separating SOC pools of different physical and chemical characteristics (Sarkhot et al. 2007a, b, and 2008; Haile et al. 2008; Chapter 2). Thus the first objective was to observe the

PAGE 65

65 changes over time in SOC within these size fractions and evaluate the importance of the C source to the formation and maintenance of SOC. We hypothesized that SOC formation in these soils were primarily dependent on belowground sources of C, mainly in the form of fine root turnover These C sources have the advantage of being directly incorpo rated into the soil matrix, as opposed to aboveground source ; which must go through some process of initial entry into the soil ( i.e. physical soil mixing or leaching ) The large sand size fractions are described as containing substantial amounts of fresh ly added particulate organic carbon ( POC ) and ag gregates of varying strength (S arkhot et al. 2007a) We expect these pools to be the most dynamic in gains and losses of SOC through the field treatments, i n part due to new additions of POC in treatments rec eiving new C additions but also to the role that fine roots and hyphae play in aggregate stabilization. The smaller size fractions define the very fine root size classes which are characterized by ephemeral growth and rapid turnover rates ( King et al. 2002 ) and we hypothesized belowground exclusion of C will have a negative effect on these C pools. The smaller size fractions, specifically those in the 150 53 m range, appear to contain more stable aggregate organic carbon (AOC) (Sarkhot et al. 2007a; A zuaje et al. 2012) Formation of these smaller ag gregates has been linked to formation process occurring within macroaggregates (Gale et al. 2000; Six et al. 2000; Rasse 2005; Kong et al. 2010) Destabilization of larger aggregates may enrich the smaller s ize fractions with POC and AOC. The second objective of this study was to evaluate the effect that C sources have on the formation and stability of the large sand size microaggregates. We hypothesize d

PAGE 66

66 that aggregates in the 2000 250 m fraction would requ ire continuous inputs to maintain aggregated OC, particularly for the low energy aggreg ates. Fresh inputs have been reported to enrich aggregates in this size class and are cited as a probable reason for their stability (Angers and Giroux 1996; Puget et al 1995). Methods Field Treatments This portion of the study was performed in the same stands of replicated blocks described in Chapter 2. A randomized complete block study design comprised of three study blocks with three replications of each field treatme nt per block. Within each block, the study was carried out in the middle of three center rows and utilized the interbed region to randomly place one replication of each of the three treatments. Each treatment plot was a 0.5 m 2 and permanently marked with P VC tubes. The two treatments beyond the untreated control were designed to eliminate sources of C inputs and prohibit them from entering the soil. The first treatment excluded aboveground litter and prohibited any future C addition from this source. This was accomplished by removing the entire O horizon present at the initiation of the study and keeping this bare soil condition for the entirety of the study. In addition to removing the O horizon from the designated treatment plot area an extra 0.5 m surro unding the plot vicinity was also removed in order to reduce All understory vegetation present was cut at the soil surface and removed. The total removal area was 2 m 2 and organic material was removed leaving only the bare mineral soil surface. Shade cloth, 1 m by 2 m, was placed over the bare soil and held in place with metal stakes. Monthly to bimonthly the

PAGE 67

67 aboveground litter on the surface of the shade cloth was removed and any sprouting vegetation, fungi or moss pr esent under the shade cloth was also removed. The second treatment excluded above plus belowground litter additions. This treatment used the same method as the first treatment to eliminate C inputs from above with the addition of trenches that surround ed the treatment plot and prohibited new root growth from entering the plot. This was accomplished by digging a 40 cm deep trench 0.5 m from the outline of the treatment plot, lining it with aluminum sheet metal and back filling with soil. The treatment plot was then covered with shade cloth tacked in place with metal stakes. The aboveground litter on the surface of the shade cloth was removed monthly to bimonthly and any sprouting vegetation, fungi or moss present under the shade cloth was also removed. The final treatment was an untreated control. A 0.5 m 2 area was marked in the interbed and left to represent the natural soil state experiencing the full influence from the O horizon and belowground root growth. All soil samples were taken from the 0.5 m 2 plo t area. There we re three replications of each tre atment in each of three blocks. In total, there were nine replicates of each field treatment across all three blocks. Soil Sampling Soil samples at time 0 were taken to establish the C content at the beginn ing of the study. Eight sample cores per treatment plot were taken with a 1.3 cm diameter soil into soil depths of 0 to 10 cm and 10 to 20 cm and bulked. Soil samples w ere subsequently collected after 31 months from each plot. Three soil samples were

PAGE 68

68 collected with a 6.5 cm diameter aluminum cylinder to a depth of 20 cm after removing the forest floor. Each sample was separated into soil from the 0 to 10 cm and 10 to 20 cm soil depths and combined for each plot. Laboratory Methods On the same day that the soil samples were collected, they were measured for fresh weight and moisture content. Bulk density was calculated from the soil volume and the weight of soil corrected for moisture content. The samples were subsequently air dried and passed through a 2 mm mesh sieve. The coarse and fine earth fractions were weighed and stored. One hundred g of air dry so il was then dry sieved through different sized sieves for 5 minute s on a horizontal shaker following the method of Sarkhot et al. (2007a). The sieves fractionated the soil into 2000 to 250 m, 250 to 150 m, 150 to 53 m and < 53m size fractions. Total organic carbon in the four fine earth fractions were measured by los s on ignition (LOI) and then converted to organic C based on a pedotransfer function developed from 133 soil samples taken from a nearby watershed study (Azuaje et al. 2012). The model created between LOI and SOC, where SOC% = 0.4922*LOI% + 0.065, had a c orrelation coefficient of 0.84. The LOI was accomplished and sandy nature (>95% sand) made LOI and conversion by pedotransfer function an appropriate method for estima ting organic C given that carbonates and loss of structural water were not issues (Nelson and Sommers 1996) and the strong correlation of the pedotransfer function (Azuaje et al. 2012). Aggregate dispersion energy curves (ADEC) were developed on the 2000 to 250 m size fraction from soil sampled after 31 months of field treatment. One curve was

PAGE 69

69 made for each of the three field treatment replicat ions per study block at two soil depths, 0 10 cm and 10 20 cm. In total there were nine replications for each fie ld treatment at each soil depth. The method used to create the ADECs was based on previous studies conducted on forested Spodosols as described by Sarkhot et al. (2007a) and Azuaje et al. (2012) using a sonic dismembrator (Fisher Scientific, Model 500, Ham pton, NH). Ten sub samples, five grams each, were weighed into 250 mL beakers, then placed in a thermally insulated container and 100 mL of deionized water were added. The insulated container was placed in a sound minimizing enclosure and a sonic probe was submerged at a constant depth of 12 mm in the center of the container. One sub sample was exposed to a set energy application. Each subsequent subsample was then subjected to increasing amounts of energy. A total of 10 energy levels were used on each sub sample, with the energy ranging from 0 to153 J mL 1 A pervious study by Azuaje et al. (2012) into the sonication method suggested that 153 J mL 1 was a useful maximum energy level. To control increasing temperature caused by the increasing energy inputs the energy was pulsed with 60s on and 30s off. The actual energy applied to the water was calculated using a calibration factor developed by Sarkhot et al. (2007a), following the recommendations of Schmidt et al. (1999). Upon completion of sonication, the sub sample was wet sieved through a 250 m mesh sieve. The portion that moved through the sieve was released by the applied energy and the portion that remained was AOC and SOC unbroken by the applied energy. The C contents of these two portions were measu red by LOI and the quantity of SOC liberated at that specific energy level was calculated as described above.

PAGE 70

70 Statistical Methods The population distributions of the data were analyzed. Evaluation of the field treatment effects on SOC focused on the relat ive change of SOC occurring between sampling times and between two soil depths. The effect of the field treatments on the size fractions was evaluated on each individual fraction. The change occurring over time was evaluated relative to the initial SOC mea sured at month 0. The population distribution was found to be lognormally distributed. In order to use the log value, a constant value of one was added to the relative change in SOC to create all positive data points (McDonald 2009). The log values were us ed in the statistical analysis but for the purposes of tables and figures, the values were back transformed. The ADEC data was found to be normally distributed. The data sets were evaluated for outliers and any value beyond three standard deviations from t he mean was removed. A SAS PROC MIXED model (Littell et al. 2006) was used for statistical analysis of the size fraction SOC and ADEC studies. Results are reported as LS Means and standard errors. In the analysis of the size fraction SOC data, soil depth a nd months of treatment were fixed effects and block was a random effect. In the ADEC, soil depth and dispersion energy were used as fixed effects and block was a random effect. A p value equal to or less than 0.1 was considered a significant effect. When s ignificant main effects and/or interactions were found, multiple mean comparisons and separations were identified using the ADJUST =TUKEY option of the LSMEANS statement.

PAGE 71

71 Results Objective 1. Assess the Contribution of Above and Belowground Sources to SOC Pools Defined by Size Fraction The field treatments had no significant effect on the net change of SOC in the 2000 250 m size fraction. Positive LSmeans were reported for each of the field treatments. Although all treatments were not significantly differ ent, the untreated control was the only one to have a mean different fro m zero (Figure 4 1 ). The untreated control exp erienced a mean net gain of 3.0 mg C cm 3 ; the a boveground exclusion treatment gained 1.2 mg C cm 3 and the above plus b elowground exclu sion treatment gained 1.8 mg C cm 3 (Table 4 1 ). There was not a significant soil depth main effect. The SOC in the 250 150 m was also not affected by the field treatments. In this size fraction all of the field treatment LSmeans were positive but contai ned zero in the 90% confidence interval and cannot be defined as increasing in SOC. The LSmeans wer e very simlar in value among treatments and range between 21.1% and 29.8% (Figure 4 1 ). The untreated contro l experienced a net gain of 0.2 mg C cm 3 a boveg round exclusion gained 0.8 mg C cm 3 and above plus belowground exclusion gained 0.7 mg C cm 3 (Table 4 1 ). The net change in SOC was significantly affected by soil depth, p=0.006), w ith a significant 40% increase in SOC at the 0 10 cm soil depth and no net change at the 10 20 cm soil depth. The f ield treatment was highly significant main effect in the 150 53 m size fraction (p value=0.004). Soil OC was lost in the treatment plots that excluded above plus belowground sources of C but the change in SOC was not different than zero in the untreated control and aboveground exclusion treatment plots (Figure 4 1 ). The untreated contro l experienced a net gain of 1.2 mg C cm 3 abo veground exclusion a

PAGE 72

72 loss of 0.1 mg C cm 3 and above plus bel owground exclusio n a loss of 0.4 mg C cm 3 (T able 4 1 ). There was not a significant soil depth main effect. Net change in fraction, experienced a significant treatment by soil depth interaction (p=0.025). Th ere was no significant treatment effect at the 0 10 cm soil depth and all of the treatments were no t different than zero. The untreated contro l experienced a net loss of 0.1 mg C cm 3 aboveground exclusion gained 0.9 mg C cm 3 and the above plus b elowg round exclusion gained 0.2 mg C cm 3 (Table 4 1 ). In the 10 20 cm soil depth, the untreated control plot more than double d its SOC content over the 31 month study period (Figure 4 1 ). The two exclusion treatments were no t different than zero but the means were positive. The untreated contr ol experienced a net gain of 1.3 mg C cm 3 the aboveground exclusio n gained 0.5 mg C cm 3 and the above plus b elowground exclusion gained 0.5 mg C cm 3 (Table 4 1 ). The >2mm data were discussed in Chapter 3; however in summary treatment was also a significant main effect, p=0.073. The aboveground exclusion mean was significantly higher than the above plus belowground exclusion. The above plus belowground exclusion significantly decreased during the study but no signi ficant change occurred in the untreated control or the aboveground exclusion plots. Both exclusion treatments were no t different than the untreated control. The untreated control experienced a net loss of 0.2 mg C cm 3 aboveground exclusion gained 1.1 mg C cm 3 and the above plus belowground exclusion lost 1.3 mg C cm 3 (Table 4 1 ).

PAGE 73

73 Objective 2. Determining the Effect of Limiting C Sources on the Stability of Macroaggregates Statistical analysis of the AOC as a % of total SOC in the 2000 250 m fractio n revealed a significant treatment by soil depth interaction, p=0.002 (Figure 4 2). Excluding aboveground C inputs significantly decreased the overall proportion of AOC as a % of tota l SOC in the 0 10 cm soil depth The above plus belowground exclusion t reatment had the second lowest amount of AOC but was no t differe nt than the untreated control at t he 0 10 cm soil dept h There was a difference of 4.1% AOC as a % of total fraction SOC between the untreated control and aboveground exclusion plot in this so il depth. A 4.1% loss of AOC from the untreated control plots would equate to 0.23 mg g 1 of tota l fraction SOC. Aggregated OC at the 10 20 cm soil depth was unaffected by treatment. Aggregated OC measurement as a result of the energy level applied was al so a significant main effect (p value=<0.001). The effect of treatment was uniform across the energy curve. The stepwise pattern of AOC being released with increasing energy level illustrates the progression of AOC of similar strength, being broken apart b y a range of energies At some point of increasing energy, a thr eshold wa s reached and high er strength AOC were destabilized by a range of energies After a continual liberation of AOC starting at zero through 8 J mL 1 the first order of stable aggregat es was obse rved in the range of 25 to 29 J mL 1 Then soil aggregates wer e continually released u ntil the 94 J mL 1 energy level was reached, beyond which no higher strength aggregates we re thought to remain.

PAGE 74

74 Discussion Objective 1. Assess the Contributio n of Above and Belowg round Sources to SOC Pools Defined by Size Fraction Aside from the coarse fraction, SOC generally increased across all of the fractions in the untreated control plots but significant increases were located in the 2000 250 m in both soil depths and the10 20 cm soil depth >53 m fraction. This forest system is in a developmental phase characterized by a climatic rate of biomass accretion (Martin and Jokela 2004). During the short term observation period of this study, approximately 26. 6 Mg ha 1 in dry weight of new litter was added to the forest floor and fine root primary production was estimated to contribute approximately 28.8 g C m 2 annually (Chapter 3). The increases of SOC among the 2000 250 m and 250 150 m sand size fractions were expected in the control and aboveground litter exclusion plots as indication of new C inputs from both sources and as an indication of the importance of subsurface influences in the aboveground exclusion treatment. These fractions have been characte rized as containing large amounts of recently added organic material, significant amounts of particulate matter and aggregates combined with fine roots and fungal hyphae (Oades 1988 ; King et al. 2002 ; Sarkhot et al. 2007a, b ). The diameter of loblolly pin e ephemeral fine roots o ccurred within the range of 2000 400 m, whose turnover rate has been estimated to be 166 300 days depending on root diameter and water availability (King et al. 2002). Mycorrhizal associated fine roots f all within this range, 600 2 00 m, but obser ved turnover rates where a much slower at 507d ays (King et al 2002).

PAGE 75

75 The c hemical composition of the >53 m fraction was described by Sarkhot et al. (2007a) as containing large amounts of aromatics, esters, and amides. The aromatic compoun ds indicate more decomposed material, while the esters and amides may originate from microbial, fine root, and root tip cells. The specific mineralization rate of this frac tion in the 10 20 cm soil depth indicated that it was more mineralizable than the 0 10 cm depth (Chapter 2). The source of the net gain of SOC in the untreated plots was unknown, but based on the high mineralizability it was likely a flux in root and/or microbial activity. Eliminating aboveground sources of C had a marginal effect on SOC ; it wa s never significantly different than the untreated control but its mean was typically lower. Many prolonged forest floor removal and repeated litter raking studies report ed that SOC content was not affected by these treatments after 4 7 years ( Ross et al. 1994 ; Nadelhoffer et al. 2004; Blazier et al. 2008 Garten 2009 ). Fine root growth has also been observed to be unaffected by litter removal in tropical and temperate coniferous forests (Sayer et al. 2007; Okada et al. 2011). Research into newly li berated DOC leaching fro m the forest floor suggest that it is preferentially mineralized and constitutes only 5% to 10% of the total D OC in the surface soil ( Trumbore et al. 1992 ; Guggenberger et al. 1994 ; Tegen & Drr 1996; Hagedorn et al. 2002, 2003 ). Th e forest floor removal has been found to have a negative effect on soil fauna and ectomycorrhizal hyp hae in the surface soil and represents a strong hypothesis to expl ain the depressed SOC content among these treatment plots (Ponge et al. 1993; Okada et al 2011).

PAGE 76

76 The significant loss of SOC observed in the above pl us belowground treatment plots of the 150 53 m fraction revealed the dependence of SOC development on subsoil processes. T he shortest turnover rate occurred in small fine roots <1 mm diameter (King et al. 2002). The loss of SOC in this size fracti on wa s likely a reflection of the ephemeral fine root decomposition which was not replenished by new growth. In OC was aggregated (Sarkhot et al. 2007a, Azuaje et al. 2012). The formation of t hese sand size microaggregates wa s thought to be a process occurring inside larger aggregates, a process stimulated by addition of fresh organic material and bound together by the associated microbial activity ( Oades 1984; Golchin et al. 1994; Angers et al. 1997 ; Si x et al. 2000; Gale et al. 2000 ). Fine root turnover has been identified as an important source of fresh organic material stimulating the development aggregates (Gale et al. 2000; Angers et al. 1997). Aggregate formation in this size frac tion class should be impacted by the absence of fresh root material. Objective 2. Determining the Effect of Limiting C Sources on the Stability of Macroaggregates In the absence of th e forest floor, the proportion of AOC in the 0 10 cm soil suffered a signif icant loss of 4.1%, indicating the role played by the aboveground C in AOC formation and/or stability. In terms of SOC, this would impact the AOC in the unt reated control by a loss of 0.2 mg of AOC g 1 of the total fraction SOC. If the loss of AOC continued this could become detrimental to this C pool in the long term. Reasons for this require more investigation. Microbial activity stimulated by the labile DOC from the forest floor may be an influential factor in sand size macroaggregate stability. As stated before, the forest floor provides habitat for soil fauna, which should play a role in

PAGE 77

77 stimulating AOC processes. Forest floor removal has also negative ly affect ed fungal biomass and ectomycorrhizal hyphae (Nadelhoffer et al. 2004; Okada et al. 2011) and the link between fungi and AOC in la rge sand size soil aggregates has been well documented ( Tisdall and Oades 1982; Oades and Waters 1991; Oades 1993; Sarkhot et al. 2007a; Azuaje et al 2012; ) A recent study by Azuaje et al. (2012) determined th at C protection offered by so il aggregation in these soils was poor and insignificant. Azuaje et al. (2012) observed C mineralization in whole soil after a range of aggregate disrupting ene rgy was applied. The complex nature of the whole soil, large whole soil POM content and shorter incubation time may have obscured protection being offered by aggregation. The in situ mineralization that occurred in the above plus belowground exclusion tre atment plots contradicts these findings. The AOC as a % of t otal SOC maintained their integrity and did not suffer SOC loss es after 31 months of receiving no new C inputs. Protection of SOC with aggregates may be a reality to some degree in the soils. Conc lusion Eliminating aboveground sources of C had insignificant effects on short term SOC processes had a significant negative affect on the surface SOC. In particular it wa s the SOC pools in the >2 mm and 150 53 m size fractions t hat we re most dependent on belowground sources of C in the short term. We hypothesize d that fine root activity was the significant loss in the >2 mm and 150 53 m size fractions. We concluded th at subsurface processes are the primary driver of short term SOC formation and rotation loblolly pine forest. To a small

PAGE 78

78 degree soil aggregation was impacted by the absence of a forest floor yet it wa s questi onable if the impact would become substantial in the long term. Dissolved OC supplied from the litter layer and the presence of soil fauna or ectomycorrhizal hyph ae may be significant factors affecting aggregation stability in thes e sandy soils and furthe r study is warranted

PAGE 79

79 Figure 4 1. Differences in % change of physical size fractions after 31 months of exclusion treatment in an intensively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida The study spans the 7 th yr through the 9 th yr of growth. Values are LS means of relative percent change with standard error bars. Different letters signify mean separations with a p value=<0.1 Untreated control Above ground exclusion Above p lus belowground exclusion a b b b

PAGE 80

80 Figure 4 2. Differences in AD EC of AOC as % of total SOC in the 2000 250 m fraction after 31 months of exclusion treatment in a 10 yr old intensively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida Values are LS mea ns of AOC as % of total SOC w ith standard error bars. Energy curves followed by different letters are significantly different at p<0.1 but graphically the data has been segregated by soil depth, (A) 0 10 cm depth, (B) 10 20 cm depth, for clarity (A) ab ab (B) Untreated control Above ground exclusion Above p lus belowground exclusion Untreated control Above groun d exclusion Above p lus belowground exclusion a bc c d

PAGE 81

81 Table 4 1 Sample m eans and standard deviations of changes in SOC ( mg cm 3 ) among soil size fractions after 31 months of exclusion treatment in an intensively managed loblolly pine stand growing in an Ultic Alaquod The study spans the 7 th yr through the 9 th yr of g rowth of a stand in north central Florida Untreated control Aboveground exclusion Above plus belowground excl. Size Fraction mean stdv mean stdv mean stdv >2 mm* 0.16 2.77 1.06 3.01 1.34 3.08 2000 250 m 2.96 4.66 1.23 5.65 1.76 4.04 250 150 m 0.19 1.96 0.83 1.12 0.69 1.04 150 53 m 1.21 1.19 0.11 2.81 0.44 1.59 <53 m 0 10 cm 0.06 1.49 0.88 1.39 0.24 0.67 <53 m 10 20 cm 1.27 0.47 0.54 1.65 0.49 1.14 <2mm data from Chapter 3

PAGE 82

82 CHAPTER 5 SYNTHESIS The overa ll objective of this thesis was to explore the short term formation, maintenance, and mineralizability o f SOC at the surface of a sandy Spodosol supporting intensively managed planted loblolly pine This is a dominate soil order i n the so utheastern United States occupying an estimated 5.7 million ha and commonly supports loblolly pine an important commercial species, covering approximately 12 million ha in this region as planted and natural stands (Baker and Langdon 1990; Adegbidi et al. 2002). This was a ddressed by three specific objectives: (1) investigate the natural short term whole soil changes in SOC to set a baseline for treatment changes while also determining the distribution and changes of SOC by size fraction including SOC specific mineralizat ion rates and aggregate stability; (2) investigate the importance of above and belowground C sources to the development of SOC through a sequential exclusion experiment; and (3) investigate the influence of the above and belowground sources C on the size fraction SOC pools and aggregation. The study was designed to determine the relative importance of litter and root turnover on the maintenance of SOC and soil aggregates in this specific yet important soil/vegetation type. The high net primary production rate of this rapidly growing mid rotation loblolly pine ( Pinus taeda L .) stand provided the opportunity to measure significant short term changes in surface SOC. This study measured a mean annual accretion rate of 2.3 mg SOC cm 3 soil yr 1 The annual SOC accretion rates estimated in this study were similar to estimations based on the cumulative rotation accretion rate of a mature loblolly stand (Vogel et al. 2011). The majority of net gain occurred in the 0 10 cm soil depth (Figure

PAGE 83

83 2 2 ) and more specifica lly, short term changes were most significant in the 2000 250 m and 150 53 m size fractions (Figure 2 3 and Figure 4 1 ) The combined accretion measured in these two frac tions at the 0 10 cm soil depth accounted for almost half of the total net SOC addit ions. R egardless, almost all size fractions received and maintained SOC additions and indicated that these soils wer e capable of accruing and sustaining SOC on a short term scale. The exceptions to the SOC increase s experienced by size fractions were the > 2 mm and the <53 m fra ction in the 0 10 cm soil depth Over 31 m onths, both fractions showed a lack of SOC accretion the in control plots (Figure 4 1) The lack of change in the >2 mm fraction may reflect more intense partitioning to fine root development as a response to the alleviation of drought stress which occurred during the first 12 months of the study. The chemical composition of the <53 m fraction has been defined as a combination of labile and recalcitrant C (Sarkhot et al. 2007b). This fractio n may be incorporated into larger size fraction aggregates and although an increase was not measured during thi s study, the long term changes may be more uniform Specific mineralization of SOC in the 0 10 cm soil depth was affected by size fraction. Miner alization rates tended to be higher in the larger fraction sizes, with as much as a 40% difference between the highest and lowest rates; suggesting another potential reason for lower net accumulation in the <53 m fraction 0 10 cm soil depth When the fra ction SOC content was examined, the largest size fraction (2000 250 m) in the 0 10 cm soil depth contained the greatest quantity of SOC. This, in combination with its labile mineralization quality, indicated the greatest potential quantity of SO C to parti cipate in change. It wa s not clear to what degree mineralizability play ed a role in the

PAGE 84

84 overall understanding of short term changes since SOC enhancement, not loss, represented this stage of stand/SOC development. The size classes used in this study are k nown to have different chemical compositions and varying degrees of soil aggregation stability (Sarkhot et al. 200 7a, b; Azuaje et al. 2012). It wa s still unclear how much SOC protection wa s being offered by aggregation. In this study, evaluation of aggreg ation in the 2000 250 m fraction revealed that the proportion of AOC was not diminished when new C inputs were eliminated for 31 months. Azuaje et al. (2012) found whole soil mineralization to be unaffected by aggregate dispersion. Due to the large propor tion of labile SOC, these questions can only be answered by focusing on the C pools that contain the largest portions of aggregated C. It was clear on a whole soil basis that C derived from belowground sources provided the main contributions to increasing and maintaining SOC pools during t his short term study conducted in a loblolly plantation. Several studies in hardwood forests have suggested the same dependence on belowground C sources for SOC development and maintenance (Nadelhoffer et al. 2004; Frober g et al. 2007; Garten 2009; Kramer et al. 2010). In addition, southern pine litter raking studies have reported surface SOC to be unaffected after 5 7 consecutive years of litter removal (Ross et al. 1994; Blazier et al. 2008). Eliminating new sources of C for an extended period of time revealed a SOC dependence of root inputs to the >2 mm and 150 53 m fractions. Lateral coarse roots and associated root branches are the primary constituent of the >2 mm size class in these sandy soils that developed from ma rine deposit parent material. Roots are also

PAGE 85

85 suspected as the principle sour ce of C in the 150 53 m fraction. Very fine roots (<1 mm) have been identified as ephemeral with seasonal patterns of growth. They are dynamic but not uniform in function and morp hology (King et al. 2002). First order roots, the smallest fine roots, experience the highest rate of turnover and high concentrations of nitrogen at the root tip (Pregitzer et al. 2002b). First order root activity appeared to be significant in maintaining SOC in the 150 53 m fraction, possibly as these small roots turnover. total SOC and litter fall were meas ured; but because loblolly pine is one of the most studied trees spe cies, a wealth of information is available on diverse aspects of its C cycle. Utilizing literature and data from this current study, a flow diagram was constructed with estimations of annual C inputs and outputs to the surface soil (Figure 5 1). Fine root turnover, the primary belowground source of C to SOC development, was esti mated to contribute 14.4 mg C cm 2 soil annually to the surface 20 cm of soil. This was based on estimated fine root biomass production at this developmental stage management, and s tems ha 1 It also assumed h alf of the fine root biomass would turnover in that time and half of the mass is C. Annual C released from fresh needl e fall was approximately 10.8 mg C cm 2 assuming an initia l 25% C loss in the first year followed by a slow g radual release thereafter. It wa s suspected that C liberated from freshly fall en needles wa s not a large C source entering the soil. A growing collection of research suggests that it is primarily integrated into C cycling confined to the forest floor, DOC leaching from the forest floor and into the surface soil has been linked to more decomposed needle litter. The annual

PAGE 86

86 contribution of DOC leaching from the bottom of the O ho rizon was estimated to be 2.5 mg C cm 2 of DOC from limited available literature f rom mature loblolly pine stands. By mid rotation, the forest floor of southern pines is only begi nning to develop discrete levels of decomposition If indeed the source of forest floor DOC is from more decomposed material, this may be an over estimation of C input. The maj ority of the SOC increase occurred in the soil closest to the surface (Figure 2 2) and primarily in two size fractions, 2000 250 m and 150 53 m (Figure 4 1) Although it was not validated, fine root activity is thought to be concentrate d in these two size fractions. Carbon mineralization is primarily determined by the C content of the fraction and losses were concentrated in the largest fraction, 2000 250 m. Soil OC content decreasing with fraction size is the normal distribution in san dy soils. The estimate of total annual soil microbial C respiration was an average of several studies of soil respiration in mid rotation Florida pine plantations ( Ewel et al. 1986; Clark et al. 2004; Samuelson et al. 2009). The 30% portion attributed to s oil microbial respiration, was based on root respiration measurements in mid rotation, fertilized loblolly pine study by Maier and Kress (2000). At this phase in ecosystem development, C is aggrading not only in tree and forest floor biomass, but SOC as w ell. Soil OC accretion in this study was twice as high as the estimated annual accretion of SOC in an end of rotation lobloll y pine stand, under similar management, soil type, and genetic quality (Vogel et al. 2010). The estima ted long term accretion rate reported by Vogel e t al. (2010) was approximately 1.5 mg cm 3 yr 1 in the 0 20 cm of soil. The phase of development the pine stand studied in this thesis is described as one of rapid growth and biomass development and it is probable

PAGE 87

87 that more SOC yr 1 wou ld be added during this phase as opposed to any other phase of development. The highest annual needle fall rate occurs during this mid rotation phase and the production rate measured in this study is similar to loblolly pine stands at this age (Jokela and Martin 2000). The needle fall rate levels off around age 10, with a correlation between needle production and fine root growth, it too eventually enters into steady state production. Data from this study suggest SOC development is primarily dependent on fi ne root turnover and therefore may anticipate the long term SOC growth, needle fall and fine root growth (Figure 5 2). This would explain the higher SOC accretion rate a t the mid ration stage than the long term average and indicates this as a critical time in SOC development. The litterfall and SOC accretion rates measured in this study also indicate that if SOC development is entirely dependent on fine root turnover, as suspected, annual fine root turnover at th is phase of stand development was higher than annual needlefall. This study observed short term changes in SOC development through a brief time and forest SOC development in general requires further study. Climate variations, specifically periods of significant drought like those observed immediately prior to the initiation of this study likely had a significant impact on the C dynamics and SOC changes measured during the study. What portions of the SOC accretion measured was due to the alleviation of water stress and normal SOC development is not entirely clear. The effects of the s evere and prolonged drought from 2006 through 2007 on C dyna mics in slash pine stands growing in close proximity to this study were reported in 9 year study

PAGE 88

88 by B racho et al. (201 2 ). The authors observed significant neg ative effects on net ecosystem C accumulation compared to previous wet years, where drought reduce d net ecosystem C exchange by 25%. Water stress negatively impacted needle growth and induced early needle drop. It also affected radiation use efficiency by closing stomata opening and decreasing the uptake of C Although belowground growth responses were not measured, it is probable that it too was affected considering the intensive C investment required by fine root respiration and maintenance As the drought conditions were alleviated during our study, net primary production would convert from a contrac ting state into a biomass production state. In addition to the change in climate, the SOC accretion measured in this study would be augmented through stand fertilization management and through the combination of increased belowground biomass and decreased soil respiration (Johnsen et al. 2001). The inherent spatial varia bly of SOC in forested soils mak e it difficult to detect short term changes. The employment of resampling microplots and measurement of initial SOC state greatly increase s the capacity to d etect small divergence of SOC contents over short periods of time (Conant et al. 2003). Statistical analysis of detecting SOC change in a forested soil by Conant et al. (2003) would suggest that several more microplots and another year of field treatment w ould have increased the power of detecting SOC change. I t i s recommend that this study could be repeated with the use of larger plots and more replicates to better contrast the role of above and below ground inputs on SOC. Studies on the effect of manageme nt and land use changes on SOC have identified various C pools more sensitive to change than others. This study began to

PAGE 89

89 narrow the focus on changes of SOC in physical fractions but it is unclear what C pools within those fractions are most affected by C d evelopment and input restriction. Analysis of AOC in all the fractions and how it was affected by the field treatments would add insight into the development and protective capacity of AOC in forested, sandy soils. It would also add depth to the questions surrounding the effects of prolonged litter raking activities on SOC pools. Future studies of this nature would benefit by measuring the actual C inputs and outputs entering and exiting the soil instead of relying on literature sources Most specifically focusing on capturing input estimates of DOC leaching in from the forest floor and C additions from fine root and mycorrhizal hyphae turnover, and the C outputs of in situ soil microbial respiration and soil DOC leaching away from the surface soil. Additio consideration in future studies are t he role of soil fauna, fine root/ectom ycorrhizal biomass residing in the litter layer and possibly most importantly, the role soil moisture pla y s in connecting C cycling in the forest floor to the mineral soil through elevated microbial activity Hass et al. (2010) found drainage class to be a significant factor of forest floor biomass in loblolly pine stands, where forest floor biomass was signi ficantly higher in the excessively drained sites and decomposition was inhibited by lack of consistent moisture. In more poorly drained landscapes, where the forest floor litter is more consistently moist under persistent decomposition, aboveground litter may have a greater opportunity as a C source in SOC development. Although data were not presented by block

PAGE 90

90 block showed the largest negative affect on SOC development in the aboveground e xclusion treatment plots. T h e results of this simple short term study revealed its methods were effective in the investigation of SOC development processes. It has also identified opportunities in future studies to illuminate more definitive insights into the complex processes involved in SOC development. The applications of these methods are not only useful in studies of monoculture plantings of intensively managed pine forests, but may be used to investigate naturally forested soils in the so utheast US and across other regions.

PAGE 91

91 Figure 5 1. Estimated annual C inputs in intensively managed mid rotation loblolly pine stand growing in an Ultic Alaquod in north central Florida. I nputs include d fine roo t turnover and forest floor DOC Estimated inputs were compared to the measured net change of SOC in the fine earth fraction with the percent net con tribut ions by size fraction s at two soil depth s The estimated annual C outputs from microbial mineralization with p rojected percentages of C mineralized from four size fra ctions, at two soil depths, were estimated from a 162 d laboratory incubation Note: Ann ual soil microbial respiration wa s without root respiration and was estimated from literature sources. Soil microbial C respiration* 27.5 mg C cm 2 Net SOC Change +4 6 mg C cm 2 0 20 cm soil 8 % 6 % 7 % 4 % 2000 250 m 250 150 m 150 53 m Projected C min eralization by size f raction & s oil depth <53 m Soil Size Fractions 28% 13 % 19% 1% 15 % 2 % 7% 15% Net SOC change by fraction 0 10 cm depth Net SOC change by fr action 10 20 cm depth Fresh Needle C loss 10.8 mg C cm 2 Forest Floor DOC 2.5 mg C cm 2 Needle fall 43 mg C cm 2 Fine Root Turnover 14.4 mg C cm 2 0 20 cm Soil 37% 18 % 15 % 6 % 10 cm 20 cm

PAGE 92

92 Figure 5 of biomass accumulation concentrated in between the young and mature stages of growth. Soil OC development in an intensively m anaged loblolly pine stand may resemble a similar trend Time Accretion Rate

PAGE 93

93 APPENDIX A STATISTICAL OUTPUT F ROM SAS Abbreviations Definition D Soil Depth; 0 10 cm, 10 20 cm M Months in Field Treatment; 0, 31 F Physical Size Fraction; 2000 250 m, 250 150 m, 150 53 m, <53 m DE Dispersion Energy; 0 ,8, 25, 29, 41 ,49, 57, 94, 119, 153 T Field Treatment a 2000 250 m Fraction Size b 250 150 m Fraction Size c 150 53 m Fraction Size d <53 m Fraction Size Control Untreated Control Field Treatment Abv Aboveground E xclusion Field Treatment Abv+Blw Above plus Below ground Exclusion Field Treatment Adj P Tukey Kramer mean saturations Table A 1. Chapter 2 Statistical output f rom SAS Control Plot Soil Bulk Density g cm 3 soil n=12 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F D 1 6 19.69 0.0044 M 1 6 0.99 0.3581 D*M 1 6 2.18 0.1907 Least Squares Means Eff ect Depth Months Estimate s.e. M 0 1.14 0.07 M 31 1.06 0.07 D 0 10 0.93 0.07

PAGE 94

94 Table A 1. Continued Effect Depth Months Estimate s.e. D 10 20 1.26 0.07 D*M 0 10 0 0.9145 0.08727 D*M 0 10 31 0.9504 0.08727 D*M 10 20 0 1.3564 0.08727 D*M 10 20 31 1.1718 0.08727 Differences of Least Squares Means Effect Depth _Depth t Value Pr > |t| Adj P D 0 10 10 20 4.44 0.0044 0.0044 Control Plot on Coarse Fraction ( >2 mm ) mg C cm 3 soil n=36 Log Transformed data Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F M 1 30 0.09 0.7659 D 1 30 0.7 0.4094 D*M 1 30 5.73 0.0231 Least Squares Means Effect Depth Months Es timate s.e. M 12 0.4502 0.1448 M 31 0.4279 0.1448 D 0 10 0.408 0.1448 D 10 20 0.4701 0.1448 D*M 0 10 12 0.3303 0.154 D*M 0 10 31 0.4857 0.154 D*M 10 20 12 0.5701 0.154 D*M 10 20 31 0.3701 0.154 Diff erences of Least Squares Means Effect Depth Months _Depth _Months t Value Pr > |t| Adj P D*M 0 10 12 0 10 31 1.48 0.1492 0.4614

PAGE 95

95 Table A 1. Continued Effect Depth Months _Depth _Months t Value Pr > |t| Adj P D*M 0 10 12 10 20 12 2.28 0.0296 0.1004 D*M 0 10 12 10 20 31 0.38 0.7072 0.9811 D*M 0 10 31 10 20 12 0.8 0.4277 0.852 D*M 0 10 31 10 20 31 1.1 0.2795 0.6915 D*M 10 20 12 10 20 31 1.91 0.0664 0.2473 Control Plot >2 mm mg C g 1 soil n=36 Log Transformed data Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F M 1 30 0.08 0.7764 D 1 30 0.57 0.455 D*M 1 30 4.41 0.0443 Least Squares Means Effect Depth Months Estimate s.e. M 12 0.4354 0.1799 M 31 0.4135 0.1799 D 0 10 0.4535 0.1799 D 10 20 0.3954 0.1799 D*M 0 10 12 0.384 0.1879 D*M 0 10 31 0.523 0.1879 D*M 10 20 12 0.4869 0.1879 D*M 10 20 31 0.304 0.1879 Differences of Least Squares Means Effect Dep th Months _Depth _Months t Value Pr > |t| Adj P D*M 0 10 12 0 10 31 1.28 0.2098 0.5814 D*M 0 10 12 10 20 12 0.95 0.3502 0.7789 D*M 0 10 12 10 20 31 0.74 0.4663 0.881 D*M 0 10 31 10 20 12 0.33 0.7417 0.9871

PAGE 96

96 Table A 1. Continued Effect Dep th Months _Depth _Months t Value Pr > |t| Adj P D*M 0 10 31 10 20 31 2.02 0.0525 0.2035 D*M 10 20 12 10 20 31 1.69 0.102 0.3481 Control Plot Fine Earth Fraction (<2 mm) mg C cm 3 soil n=12 Log Transformed data Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F D 1 6 20.09 0.0042 M 1 6 9.37 0.0222 D*M 1 6 4.14 0.0881 Least Squares Means Effect Months Depth Estimate s.e. D 0 10 17.0633 2.5004 D 10 20 8.485 9 2.5004 M 0 9.8456 2.5004 M 31 15.7036 2.5004 D*M 0 0 10 12.1875 2.8431 D*M 0 10 20 7.5037 2.8431 D*M 31 0 10 21.9391 2.8431 D*M 31 10 20 9.468 2.8431 Differences of Least Squares Means Effect Months Depth _M onths _Depth t Value Pr > |t| Adj P D*M 0 0 10 0 10 20 1.73 0.1342 0.3862 D*M 0 0 10 31 0 10 3.6 0.0113 0.0425 D*M 0 0 10 31 10 20 1 0.3537 0.7529 D*M 0 10 20 31 0 10 5.33 0.0018 0.0071 D*M 0 10 20 31 10 20 0.73 0.4953 0.8835 D*M 31 0 10 31 10 20 4.61 0.0037 0.0144

PAGE 97

97 Table A 1. Continued Control Plot Fine Earth Fraction (<2 mm) mg C g 1 soil n=12 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F D 1 6 8.83 0.0249 M 1 6 3.33 0.11 78 D*M 1 6 1.43 0.2769 Least Squares Means Effect Months Depth Estimate s.e. D 0 10 19.9944 4.4841 D 10 20 6.9646 4.4841 M 0 9.4788 4.4841 M 31 17.4801 4.4841 D*M 0 0 10 13.3729 5.4512 D*M 0 10 20 5 .5848 5.4512 D*M 31 0 10 26.6158 5.4512 D*M 31 10 20 8.3444 5.4512 Differences of Least Squares Means Effect Depth _Depth t Value Pr > |t| Adj P D 0 10 10 20 2.97 0.0249 0.0249 Control Plot Physical Size Fractions mg C cm 3 soil Log Transformed data n=96 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F M 1 78 17.74 <.0001 D 1 78 59.16 <.0001 F 3 78 18.27 <.0001 D*M 1 78 0.47 0.4964

PAGE 98

98 Table A 1. Conti nued Effect Num DF Den DF F Value Pr > F F*M 3 78 0.33 0.8053 D*F 3 78 4.19 0.0083 D*F*M 3 78 2.03 0.1169 Least Squares Means Effect Depth Fraction Months Estimate s.e. M 0 0.4218 0.07074 M 31 0.5622 0.06 67 D 0 10 0.6202 0.06875 D 10 20 0.3638 0.06875 F a 0.6762 0.07268 F b 0.4579 0.07268 F c 0.5024 0.07268 F d 0.3317 0.07268 D*M 0 10 0 0.5614 0.07641 D*M 0 10 31 0.679 0.06875 D*M 10 20 0 0.2823 0 .07641 D*M 10 20 31 0.4454 0.06875 F*M a 0 0.5939 0.08663 F*M b 0 0.4147 0.08663 F*M c 0 0.4321 0.08663 F*M d 0 0.2467 0.08663 F*M a 31 0.7585 0.07268 F*M b 31 0.5011 0.07268 F*M c 31 0.5728 0.07268 F*M d 31 0.4166 0.07268 D*F 0 10 a 0.9052 0.07996 D*F 0 10 b 0.5388 0.07996 D*F 0 10 c 0.5956 0.07996 D*F 0 10 d 0.4414 0.07996 D*F 10 20 a 0.4471 0.07996

PAGE 99

99 Table A 1. Continued Effect Depth Fraction Months Estimate s .e. D*F 10 20 b 0.377 0.07996 D*F 10 20 c 0.4093 0.07996 D*F 10 20 d 0.2219 0.07996 D*F*M 0 10 a 0 0.8389 0.1041 D*F*M 0 10 b 0 0.4672 0.1041 D*F*M 0 10 c 0 0.5067 0.1041 D*F*M 0 10 d 0 0.433 0.1041 D*F*M 0 10 a 3 1 0.9715 0.07996 D*F*M 0 10 b 31 0.6104 0.07996 D*F*M 0 10 c 31 0.6845 0.07996 D*F*M 0 10 d 31 0.4497 0.07996 D*F*M 10 20 a 0 0.3489 0.1041 D*F*M 10 20 b 0 0.3623 0.1041 D*F*M 10 20 c 0 0.3575 0.1041 D*F*M 10 20 d 0 0.060 4 0.1041 D*F*M 10 20 a 31 0.5454 0.07996 D*F*M 10 20 b 31 0.3918 0.07996 D*F*M 10 20 c 31 0.461 0.07996 D*F*M 10 20 d 31 0.3835 0.07996 Differences of Least Squares Means Effect Depth Fraction Months _Depth _Fraction _Months t Val ue Pr > |t| Adj P M 0 31 4.21 <.0001 <.0001 D*F 0 10 a 0 10 b 5.5 <.0001 <.0001 D*F 0 10 a 0 10 c 4.64 <.0001 0.0004 D*F 0 10 a 0 10 d 6.96 <.0001 <.0001 D*F 0 10 a 10 20 b 6.87 <.0001 <.0001 D*F 0 10 a 10 20 c 7.92 <.0001 <.0001 D*F 0 10 a 10 20 d 7.44 <.0001 <.0001 D*F 0 10 a 10 20 d 10.25 <.0001 <.0001 D*F 0 10 b 0 10 c 0.85 0.3971 0.9893

PAGE 100

100 Table A 1. Continued Effect Depth Fraction Months _Depth _Fraction _Months t Value Pr > |t| Adj P D*F 0 10 b 0 10 d 1.46 0. 1479 0.8252 D*F 0 10 b 10 20 a 1.38 0.1731 0.8657 D*F 0 10 b 10 20 b 2.43 0.0176 0.2436 D*F 0 10 b 10 20 c 1.94 0.0556 0.5265 D*F 0 10 b 10 20 d 4.75 <.0001 0.0002 D*F 0 10 c 0 10 d 2.31 0.0234 0.3001 D*F 0 10 c 10 20 a 2.23 0.0289 0.3478 D*F 0 10 c 10 20 b 3.28 0.0016 0.032 D*F 0 10 c 10 20 c 2.79 0.0065 0.1114 D*F 0 10 c 10 20 d 5.6 <.0001 <.0001 D*F 0 10 d 10 20 a 0.09 0.9315 1 D*F 0 10 d 10 20 b 0.96 0.3375 0.978 D*F 0 10 d 10 20 c 0.48 0.6315 0.9997 D*F 0 10 d 10 20 d 3.29 0.0015 0.0308 D*F 10 20 a 10 20 b 1.05 0.2964 0.9647 D*F 10 20 a 10 20 c 0.57 0.5718 0.9992 D*F 10 20 a 10 20 d 3.38 0.0011 0.0241 D*F 10 20 b 10 20 c 0.48 0.6301 0.9997 D*F 10 20 b 10 20 d 2.33 0.0226 0.293 D*F 10 20 c 10 20 d 2.81 0.0063 0.1074 Control Plot Physical Size Fraction mg C g 1 soil Log Transformed data n=96 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F M 1 78 21.9 <.0001 D 1 78 117.88 <.0001 F 3 78 15.57 <.0001 D*M 1 78 2.57 0.1127

PAGE 101

101 Table A 1. Continued Effect Num DF Den DF F Value Pr > F F*M 3 78 0.28 0.8401 D*F 3 78 3.57 0.0177 D*F*M 3 78 1.73 0.1681 Least Squares Means Effect Depth Fractio n Months Estimate s.e. M 0 0.3759 0.1012 M 31 0.5448 0.09795 D 0 10 0.6564 0.0996 D 10 20 0.2643 0.0996 F a 0.6445 0.1028 F b 0.4262 0.1028 F c 0.4707 0.1028 F d 0.3 0.1028 D*M 0 10 0 0.6008 0.1 059 D*M 0 10 31 0.7119 0.0996 D*M 10 20 0 0.1509 0.1059 D*M 10 20 31 0.3778 0.0996 F*M a 0 0.5479 0.1148 F*M b 0 0.3687 0.1148 F*M c 0 0.3861 0.1148 F*M d 0 0.2007 0.1148 F*M a 31 0.7411 0.1028 F*M b 31 0.4837 0.1028 F*M c 31 0.5554 0.1028 F*M d 31 0.3992 0.1028 D*F 0 10 a 0.9413 0.109 D*F 0 10 b 0.5749 0.109 D*F 0 10 c 0.6317 0.109 D*F 0 10 d 0.4775 0.109 D*F 10 20 a 0.3476 0.109

PAGE 102

1 02 Table A 1. Continued Effect Depth Fraction Months Estimate s.e. D*F 10 20 b 0.2775 0.109 D*F 10 20 c 0.3098 0.109 D*F 10 20 d 0.1224 0.109 D*F*M 0 10 a 0 0.8783 0.1307 D*F*M 0 10 b 0 0.5066 0.1307 D*F*M 0 10 c 0 0.5461 0.1307 D*F*M 0 10 d 0 0.4724 0.1307 D*F*M 0 10 a 31 1.0044 0.109 D*F*M 0 10 b 31 0.6433 0.109 D*F*M 0 10 c 31 0.7173 0.109 D*F*M 0 10 d 31 0.4826 0.109 D*F*M 10 20 a 0 0.2175 0.1307 D*F*M 10 20 b 0 0.2309 0.1307 D*F*M 10 20 c 0 0.2261 0.1 307 D*F*M 10 20 d 0 0.07097 0.1307 D*F*M 10 20 a 31 0.4778 0.109 D*F*M 10 20 b 31 0.3242 0.109 D*F*M 10 20 c 31 0.3934 0.109 D*F*M 10 20 d 31 0.3159 0.109 Differences of Least Squares Means Effect Depth Fraction Months _Dept h _Fraction _Months t Value Pr > |t| Adj P M 0 31 4.68 <.0001 <.0001 D*F 0 10 a 0 10 b 5.07 <.0001 <.0001 D*F 0 10 a 0 10 c 4.29 <.0001 0.0013 D*F 0 10 a 0 10 d 6.42 <.0001 <.0001 D*F 0 10 a 10 20 a 8.22 <.0001 <.0001 D*F 0 10 a 10 20 b 9.19 <.0001 <.0001 D*F 0 10 a 10 20 c 8.75 <.0001 <.0001 D*F 0 10 a 10 20 d 11.34 <.0001 <.0001 D*F 0 10 b 0 10 c 0.79 0.4342 0.9934

PAGE 103

103 Table A 1. Continued Effect Depth Fraction Months _Depth _Fraction _Months t Value Pr > |t| Adj P D* F 0 10 b 0 10 d 1.35 0.1812 0.8768 D*F 0 10 b 10 20 a 3.15 0.0023 0.0458 D*F 0 10 b 10 20 b 4.12 <.0001 0.0023 D*F 0 10 b 10 20 c 3.67 0.0004 0.0099 D*F 0 10 b 10 20 d 6.27 <.0001 <.0001 D*F 0 10 c 0 10 d 2.14 0.0359 0.4022 D*F 0 10 c 10 20 a 3.93 0.0002 0.0043 D*F 0 10 c 10 20 b 4.9 <.0001 0.0001 D*F 0 10 c 10 20 c 4.46 <.0001 0.0007 D*F 0 10 c 10 20 d 7.05 <.0001 <.0001 D*F 0 10 d 10 20 a 1.8 0.076 0.6229 D*F 0 10 d 10 20 b 2.77 0.007 0.1181 D*F 0 10 d 10 20 c 2.32 0. 0228 0.2949 D*F 0 10 d 10 20 d 4.92 <.0001 0.0001 D*F 10 20 a 10 20 b 0.97 0.3347 0.9773 D*F 10 20 a 10 20 c 0.52 0.6016 0.9995 D*F 10 20 a 10 20 d 3.12 0.0025 0.0495 D*F 10 20 b 10 20 c 0.45 0.6566 0.9998 D*F 10 20 b 10 20 d 2.15 0.0348 0.3946 D*F 10 20 c 10 20 d 2.59 0.0113 0.1739 Total Cumulative Fraction Mineralization (162d) Respired mg C g 1 Fraction SOC Inverse Transformed data n=71; 1 outlier removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F F 3 61 1.35 0.2676 D 1 61 0 0.955 D*F 3 61 2.99 0.0378

PAGE 104

104 Table A 1. Continued Least Squares Means Effect Fraction Depth Estimate s.e. F a 0.01792 0.001676 F b 0.0159 0.001636 F c 0.01883 0.001636 F d 0.01987 0.001636 D 10 0.01817 0.001287 D 20 0.01809 0.001274 D*F a 10 0.01575 0.002306 D*F b 10 0.01426 0.002187 D*F c 10 0.01943 0.002187 D*F d 10 0.02325 0.002187 D*F a 20 0.02008 0.002187 D*F b 20 0.01755 0.002187 D*F c 20 0.01824 0.002187 D*F d 20 0.01648 0.002187 Differences of Least Squares Means Effect Fraction Depth _Fraction _Depth t Value Pr > |t| Adj P D*F a 10 b 10 0.5 0.6198 0.6198 D*F a 10 c 10 1.23 0.2238 0.2238 D*F a 10 d 10 2.51 0.0149 0.0149 D*F a 10 a 20 1.45 0.153 0.153 D*F a 10 b 20 0.6 0.55 0.55 D*F a 10 c 20 0.83 0.4089 0.4089 D*F a 10 d 20 0.24 0.8073 0.8073 D*F b 10 c 10 1.78 0.0798 0.0798 D*F b 10 d 10 3.1 0.0029 0.0029 D*F b 10 a 20 2.01 0.0492 0.0492 D*F b 10 b 20 1.13 0.2611 0.2611 D*F b 10 c 20 1.37 0.1751 0.1751 D*F b 10 d 20 0.77 0.4461 0.4461

PAGE 105

105 Table A 1. Continued Effect Fraction Depth _Fraction _Depth t Val ue Pr > |t| Adj P D*F c 10 d 10 1.32 0.1931 0.1931 D*F c 10 a 20 0.22 0.8228 0.8228 D*F c 10 b 20 0.65 0.5198 0.5198 D*F c 10 c 20 0.41 0.6834 0.6834 D*F c 10 d 20 1.01 0.3142 0.3142 D*F d 10 a 20 1.09 0.2795 0.2795 D*F d 10 b 20 1.9 6 0.0542 0.0542 D*F d 10 c 20 1.73 0.0894 0.0894 D*F d 10 d 20 2.33 0.0231 0.0231 D*F a 20 b 20 0.87 0.3865 0.3865 D*F a 20 c 20 0.63 0.528 0.528 D*F a 20 d 20 1.24 0.2199 0.2199 D*F b 20 c 20 0.24 0.813 0.813 D*F b 20 d 20 0.37 0.714 6 0.7146 D*F c 20 d 20 0.6 0.5475 0.5475 Total Cumulative Soil Fraction Mineralization (162d) Respired C mg g 1 Fine Earth Fraction (<2mm) Log Transformed data n=72 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F F 3 62 29.07 <.0001 D 1 62 99.59 <.0001 D*F 3 62 9.03 <.0001 Least Squares Means Effect Fraction Depth Estimate s.e. F a 0.5799 0.1003 F b 0.813 0.1003 F c 0.81 0.1003 F d 1.1 11 0.1003

PAGE 106

106 Table A 1. Continued Effect Fraction Depth Estimate s.e. D 0 10 0.627 0.09612 D 10 20 1.0299 0.09612 D*F 0 10 a 0.2437 0.1081 D*F 0 10 b 0.568 0.1081 D*F 0 10 c 0.6336 0.1081 D*F 0 10 d 1.0628 0.1081 D*F 10 20 a 0.9161 0.1081 D*F 10 20 b 1.058 0.1081 D*F 10 20 c 0.9865 0.1081 D*F 10 20 d 1.1592 0.1081 Differences of Least Squares Means Effect Depth Fraction _Depth _Fraction t Value Pr > |t| Adj P D* F 0 10 a 0 10 b 4.02 0.0002 0.0038 D*F 0 10 a 0 10 d 10.14 <.0001 <.0001 D*F 0 10 a 10 20 c 9.2 <.0001 <.0001 D*F 0 10 a 10 20 a 8.33 <.0001 <.0001 D*F 0 10 a 10 20 b 10.08 <.0001 <.0001 D*F 0 10 a 10 20 d 11.34 <.0001 <.0001 D*F 10 20 a 10 20 b 1.76 0.0837 0.6499 D*F 10 20 a 10 20 d 3.01 0.0038 0.0689 D*F 0 10 b 0 10 d 6.13 <.0001 <.0001 D*F 0 10 b 10 20 c 5.18 <.0001 <.0001 D*F 0 10 b 10 20 a 4.31 <.0001 0.0015 D*F 0 10 b 10 20 b 6.07 <.0001 <.0001 D*F 0 10 b 10 20 d 7. 32 <.0001 <.0001 D*F 10 20 b 10 20 d 1.25 0.2151 0.9124 D*F 0 10 c 0 10 a 4.83 <.0001 0.0002 D*F 0 10 c 0 10 b 0.81 0.4199 0.9918 D*F 0 10 c 0 10 d 5.32 <.0001 <.0001 D*F 0 10 c 10 20 c 4.37 <.0001 0.0012

PAGE 107

107 Table A 1. Continued Effect Depth Fraction _Depth _Fraction t Value Pr > |t| Adj P D*F 0 10 c 10 20 a 3.5 0.0009 0.0185 D*F 0 10 c 10 20 b 5.26 <.0001 <.0001 D*F 0 10 c 10 20 d 6.51 <.0001 <.0001 D*F 10 20 c 10 20 a 0.87 0.3865 0.9875 D*F 10 20 c 10 20 b 0.89 0 .3792 0.9863 D*F 10 20 c 10 20 d 2.14 0.0365 0.4029 D*F 0 10 d 10 20 c 0.94 0.3485 0.9802 D*F 0 10 d 10 20 a 1.82 0.0741 0.6113 D*F 0 10 d 10 20 b 0.06 0.9533 1 D*F 0 10 d 10 20 e 1.19 0.2372 0.9308 AOC as % of Total 2000 25 0 m Fraction SOC n=173; 7 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F D 1 151 3.28 0.0721 DE 9 151 56.28 <.0001 D*DE 9 151 0.79 0.6272 Least Squares Means Effect Depth DispEngy Estimate s.e. D 0 10 22.5191 2.516 D 10 20 23.9783 2.5221 DE 0 13.5089 2.7505 DE 8 14.0883 2.7505 DE 25 15.442 2.7683 DE 29 16.6749 2.7906 DE 41 17.2878 2.7906 DE 49 23.0934 2.7906 DE 57 25.0567 2.7505 DE 94 34.6667 2.7505

PAGE 108

108 Table A 1. Continued Effect Depth DispEngy Estimate s.e. DE 119 35.4 2.7505 DE 153 37.2683 2.7505 D*DE 0 10 0 11.6567 3.0182 D*DE 0 10 8 13.0233 3.0182 D*DE 0 10 25 13.5733 3.0182 D*DE 0 10 29 14.6611 3.0182 D*DE 0 10 41 16.4856 3.0182 D*DE 0 10 49 21.5111 3.0182 D*DE 0 10 57 25.6344 3.0182 D*DE 0 10 94 34.5211 3.0182 D*DE 0 10 119 36.09 3.0182 D*DE 0 10 153 38.0344 3.0182 D*DE 10 20 0 15.3611 3.0182 D*DE 10 20 8 15.1533 3.0182 D*DE 10 20 25 17.3106 3.0823 D*DE 10 20 29 18.6886 3.162 D*DE 10 20 41 18.09 3.162 D*DE 10 20 49 24.6757 3.162 D*DE 10 20 57 24.4789 3.0182 D*DE 10 20 9 4 34.8122 3.0182 D*DE 10 20 119 34.71 3.0182 D*DE 10 20 153 36.5022 3.0182 Differences of Least Squares Means Effect Depth DispEngy _Depth _DispEngy t Value Pr > |t| Adj P D 0 10 10 20 1.81 0.0721 0.0721 DE 0 8 0.33 0.7421 1 DE 0 25 1.08 0.2806 0.9857 DE 0 29 1.74 0.0839 0.7708 DE 0 41 2.08 0.0395 0.5467 DE 0 49 5.27 <.0001 <.0001

PAGE 109

109 Table A 1. Continued Effect Depth DispEngy _Depth _DispEngy t Value Pr > |t| Adj P DE 0 57 6.57 <.0001 < .0001 DE 0 94 12.04 <.0001 <.0001 DE 0 119 12.46 <.0001 <.0001 DE 0 153 13.52 <.0001 <.0001 DE 8 25 0.76 0.4494 0.999 DE 8 29 1.42 0.1572 0.9187 DE 8 41 1.76 0.0807 0.7598 DE 8 49 4.95 <.0001 <.0001 DE 8 57 6 .24 <.0001 <.0001 DE 8 94 11.71 <.0001 <.0001 DE 8 119 12.13 <.0001 <.0001 DE 8 153 13.19 <.0001 <.0001 DE 25 29 0.67 0.5052 0.9996 DE 25 41 1 0.3189 0.9919 DE 25 49 4.15 <.0001 0.0022 DE 25 57 5.39 <.0001 <.0001 DE 25 94 10.77 <.0001 <.0001 DE 25 119 11.18 <.0001 <.0001 DE 25 153 12.23 <.0001 <.0001 DE 29 41 0.33 0.7447 1 DE 29 49 3.42 0.0008 0.0272 DE 29 57 4.61 <.0001 0.0004 DE 29 94 9.89 <.0001 <.0001 DE 29 119 10. 29 <.0001 <.0001 DE 29 153 11.32 <.0001 <.0001 DE 41 49 3.09 0.0024 0.0701 DE 41 57 4.27 <.0001 0.0014 DE 41 94 9.55 <.0001 <.0001 DE 41 119 9.95 <.0001 <.0001 DE 41 153 10.98 <.0001 <.0001

PAGE 110

110 Table A 1. Continued Effect Depth DispEngy _Depth _DispEngy t Value Pr > |t| Adj P DE 49 57 1.08 0.2823 0.9861 DE 49 94 6.36 <.0001 <.0001 DE 49 119 6.76 <.0001 <.0001 DE 49 153 7.79 <.0001 <.0001 DE 57 94 5.47 <.0001 <.0001 DE 57 119 5.8 9 <.0001 <.0001 DE 57 153 6.95 <.0001 <.0001 DE 94 119 0.42 0.6771 1 DE 94 153 1.48 0.1409 0.8979 DE 119 153 1.06 0.2894 0.9874 Table A 2. Chapter 3 Statistical output from SAS Live Root Counts Number of Roots c m 2 n=54 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 46 37.29 <.0001 D 1 46 108.21 <.0001 T*D 2 46 13.19 <.0001 Least Squares Means Effect Treatment Depth Estimate s.e T Control 0.05111 0.006146 T Abv 0.05426 0.006146 T Abv+Blw 0.009072 0.006146 D 0 10 cm 0.06296 0.005664 D 10 20 cm 0.01333 0.005664 T*D Control 0 10 0.08297 0.007405 T*D Control 10 20 0.01926 0.007405 T*D Abv 0 10 0.0 8926 0.007405

PAGE 111

111 Table A 2. Continued Effect Treatment Depth Estimate s.e T*D Abv 10 20 0.01926 0.007405 T*D Abv+Blw 0 10 0.01666 0.007405 T*D Abv+Blw 10 20 0.001489 0.007405 Differences of Least Squares Means Effec t Treatment Depth _Treatment _Depth t Value Pr > |t| T*D Control 0 10 Control 10 20 7.71 <.0001 T*D Control 0 10 Abv 0 10 0.76 0.4505 T*D Control 0 10 Abv 10 20 7.71 <.0001 T*D Control 0 10 Abv+Blw 0 10 8.03 <.0001 T*D Control 0 10 Abv +Blw 10 20 9.86 <.0001 T*D Control 10 20 Abv 0 10 8.47 <.0001 T*D Control 10 20 Abv 10 20 0 1 T*D Control 10 20 Abv+Blw 0 10 0.31 0.7544 T*D Control 10 20 Abv+Blw 10 20 2.15 0.0368 T*D Abv 0 10 Abv 10 20 8.47 <.0001 T*D Abv 0 10 Ab v+Blw 0 10 8.79 <.0001 T*D Abv 0 10 Abv+Blw 10 20 10.62 <.0001 T*D Abv 10 20 Abv+Blw 0 10 0.31 0.7544 T*D Abv 10 20 Abv+Blw 10 20 2.15 0.0368 T*D Abv+Blw 0 10 Abv+Blw 10 20 1.84 0.0729 Field Treatment Ef fect on Soil Bulk Density g soil cm as Relative % Change n=54 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 46 3.02 0.0584 D 1 46 20.22 <.0001 T*D 2 46 2.48 0.095

PAGE 112

112 Table A 2. Continued Least Squares Means Effect Treatment Depth Estimate s.e C Control 0.04509 0.05575 T Abv 0.04743 0.05575 T Abv+Blw 0.07296 0.05575 D 0 10 0.1178 0.05179 D 10 20 0.06763 0.05179 T*D Control 0 10 0.042 0.0662 1 T*D Control 10 20 0.132 0.06621 T*D Abv 0 10 0.087 0.06621 T*D Abv 10 20 0.008 0.06621 T*D Abv+Blw 0 10 0.225 0.06621 T*D Abv+Blw 10 20 0.079 0.06621 Differences of Least Squares Means Effect Treatment _Treatmen t t Value Pr > |t| Adj P T Control Abv 1.83 0.0735 0.1708 T Control Abv+Blw 2.34 0.0239 0.0607 T Abv Abv+Blw 0.51 0.6158 0.8691 Field Treatment Effect on Coarse Fraction ( >2 mm ) mg C cm soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=53; 1 outlier removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 45 2.78 0.0729 D 1 45 12.43 0.001 T*D 2 45 0.46 0.6363 Least Squares Means Effect Treatment Depth Estimate s.e T Control 0.02197 0.06809 T Abv 0.06382 0.07015

PAGE 113

113 Table A 2. Continued Effect Treatment Depth Estimate s.e T Abv+Blw 0.1619 0.06809 D 0 10 0.09883 0.05582 D 10 20 0.1789 0.05695 T*D Control 0 10 0.1389 0.09589 T*D Control 10 20 0.1829 0.09589 T*D Abv 0 10 0.2334 0.09589 T*D Abv 10 20 0.1058 0.1017 T*D Abv+Blw 0 10 0.07589 0.09589 T*D Abv+Blw 10 20 0.248 0.09589 Differen ces of Least Squares Means Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P T Control Abv 0.88 0.3811 0.6527 T Control Abv+Blw 1.47 0.1497 0.3169 T Abv Abv+Blw 2.33 0.0245 0.0621 D 0 10 10 20 3.53 0.001 0.001 Field Treatment Effect on Coarse Fraction ( >2 mm ) mg C g 1 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=52; 2 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 44 2.43 0.1001 D 1 44 18.12 0.0001 T*D 2 44 0.25 0.783 Least Squares Means Effect Treatment Depth Estimate s.e. T Control 0.1334 0.162 T Abv 0.09597 0.1673 T Abv+Blw 0.4274 0.162 D 0 10 0.0692 0.149

PAGE 114

114 Table A 2. Continued Effect Treatment Depth Estimate s.e. D 10 20 0.5071 0.149 T*D Control 0 10 0.195 0.1986 T*D Control 10 20 0.4619 0.1986 T*D Abv 0 10 0.2174 0.207 T*D Abv 10 20 0.4094 0.207 T *D Abv+Blw 0 10 0.2048 0.1986 T*D Abv+Blw 10 20 0.6499 0.1986 Differences of Least Squares Means Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P T Control Abv 0.22 0.8243 0.9729 T Control Abv+Blw 1.81 0.0772 0.17 83 T Abv Abv+Blw 1.98 0.0545 0.1303 D 0 10 10 20 4.26 0.0001 0.0001 Field Treatment Effect on Fine Earth Fraction (<2mm) mg C cm soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=53; 1 outlier re moved Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 45 4.71 0.0139 D 1 45 15.44 0.0003 T*D 2 45 0.47 0.6266 Least Squares Means Effect depth Treatment Estimate s.e. T Control 0.153 8 0.04572 T Abv 0.05821 0.04471 T Abv+Blw 0.01221 0.04471 D 0 10 0.1531 0.03956 D 10 20 0.01992 0.03904 T*D Control 0 10 0.244 0.06155

PAGE 115

115 Table A 2. Continued Effect depth Treatment Estimate s.e. T*D Co ntrol 10 20 0.06362 0.05852 T*D Abv 0 10 0.1687 0.05852 T*D Abv 10 20 0.05223 0.05852 T*D Abv+Blw 0 10 0.04673 0.05852 T*D Abv+Blw 10 20 0.07115 0.05852 Differences of Least Squares Means Effect Treatment _Treatment t V alue Pr > |t| Adj P T Control Abv 1.76 0.0847 0.1936 T Control Abv+Blw 3.06 0.0037 0.0102 T Abv Abv+Blw 1.32 0.1938 0.3921 Field Treatment Effect on Fine Earth Fraction (<2 mm) mg C g 1 soil as Relative % Change Log(1+ Relat ive % Change in mg SOC) Transformed n=53; 1 outlier removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 45 5.94 0.0051 D 1 45 3.36 0.0735 T*D 2 45 1.36 0.2665 Least Squares Means Effect Treatment Depth Estimate s.e. T Control 0.1713 0.051 T Abv 0.04149 0.04988 T Abv+Blw 0.03513 0.04988 D 0 10 0.1042 0.04415 D 10 20 0.01427 0.04358 T*D Control 0 10 0.2137 0.06862 T*D Control 10 20 0. 1289 0.06524 T*D Abv 0 10 0.1367 0.06524 T*D Abv 10 20 0.05376 0.06524

PAGE 116

116 Table A 2. Continued Effect Treatment Depth Estimate s.e. T*D Abv+Blw 0 10 0.03792 0.06524 T*D Abv+Blw 10 20 0.03234 0.06524 Differences of Least Squares Means Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P T Control Abv 2.15 0.037 0.0914 T Control Abv+Blw 3.42 0.0014 0.0038 T Abv Abv+Blw 1.29 0.2042 0.409 D 0 10 10 20 1.83 0.0735 0.0735 Table A 3. Chapter 4 Statistical output from SAS Field Treatment Effect on 2000 250 m Size Fraction mg C cm soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=52; 2 outliers removed Type 3 Tests of Fix ed Effects Effect Num DF Den DF F Value Pr > F T 2 44 1.87 0.1667 D 1 44 0.15 0.6966 T*D 2 44 0.62 0.5437 Least Squares Means Effect Treatment Depth Estimate s.e. T Control 0.1646 0.07524 T Abv 0.04133 0.07 622 T Abv+Blw 0.06235 0.07622 D 0 10 0.1005 0.07002 D 10 20 0.07837 0.07098 T*D Control 0 10 0.1327 0.0891 T*D Control 10 20 0.1965 0.0891 T*D Abv 0 10 0.06641 0.0891 T*D Abv 10 20 0.01625 0.09235 T*D Abv+ Blw 0 10 0.1023 0.0891

PAGE 117

117 Table A 3. Continued Effect Treatment Depth Estimate s.e. T*D Abv+Blw 10 20 0.02236 0.09235 Field Treatment Effect on 2000 250 m Size Fraction mg C g 1 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=52; 2 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 44 3.19 0.0509 D 1 44 0.85 0.3608 T*D 2 44 0.5 0.611 Least Squares Means Ef fect Treatment Depth Estimate s.e. T Control 0.1932 0.09447 T Abv 0.01988 0.09541 T Abv+Blw 0.04136 0.09541 D 0 10 0.05631 0.08951 D 10 20 0.1133 0.09042 T*D Control 0 10 0.1261 0.108 T*D Control 10 20 0.2603 0.108 T*D Abv 0 10 0.02768 0.108 T*D Abv 10 20 0.01208 0.1112 T*D Abv+Blw 0 10 0.0151 0.108 T*D Abv+Blw 10 20 0.06763 0.1112 Differences of Least Squares Means Effect Treatment _Treatment t Value Pr > |t| Adj P T Con trol Abv 2.31 0.0259 0.0655 T Control Abv+Blw 2.02 0.0495 0.1193 T Abv Abv+Blw 0.28 0.7795 0.9572

PAGE 118

118 Table A 3. Continued Field Treatment Effect on 250 150 m Size Fraction mg C cm 3 soil as Relative % Change Log(1+ Relative % C hange in mg SOC) Transformed n=51; 3 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 43 0.26 0.7691 D 1 43 8.24 0.0063 T*D 2 43 0.25 0.7808 Least Squares Means Effe ct Treatment Depth Estimate s.e. T Control 0.08639 0.08402 T Abv 0.1134 0.0849 T Abv+Blw 0.08327 0.08439 D 0 10 0.147 0.08215 D 10 20 0.04173 0.08274 T*D Control 0 10 0.1433 0.08941 T*D Control 10 20 0.0295 0. 08941 T*D Abv 0 10 0.1483 0.08941 T*D Abv 10 20 0.07858 0.09265 T*D Abv+Blw 0 10 0.1494 0.08941 T*D Abv+Blw 10 20 0.01711 0.09077 Differences of Least Squares Means Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P Depth 0 10 10 20 2.87 0.0063 0.0063

PAGE 119

119 Table A 3. Continued Field Treatment Effect on 250 150 m Size Fraction mg C g 1 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=51; 3 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 43 0.7 0.5042 D 1 43 0.83 0.3661 T*D 2 43 0.5 0.6089 Least Squares Means Effect Treatment Depth Estimate s.e. T Contr ol 0.115 0.06996 T Abv 0.07522 0.07173 T Abv+Blw 0.06996 0.06996 D 0 10 0.1028 0.06791 D 10 20 0.07063 0.06872 T*D Control 0 10 0.1367 0.07577 T*D Control 10 20 0.09325 0.07577 T*D Abv 0 10 0.1095 0.07577 T*D Abv 10 20 0.04091 0.08213 T*D Abv+Blw 0 10 0.06219 0.07577 T*D Abv+Blw 10 20 0.07773 0.07577 Field Treatment Effect on 150 53 m Size Fraction mg C cm 3 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Trans formed n=53; 1 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 45 6.23 0.0041 D 1 45 0.54 0.4683 T*D 2 45 1.35 0.2704

PAGE 120

120 Table A 3. Continued Least Squares M eans Effect Treatment Depth Estimate s.e. T Control 0.1407 0.1402 T Abv 0.02216 0.1402 T Abv+Blw 0.2694 0.1417 D 0 10 0.01561 0.1328 D 10 20 0.08497 0.1321 T*D Control 0 10 0.1778 0.162 T*D Control 1 0 20 0.1036 0.162 T*D Abv 0 10 0.1071 0.162 T*D Abv 10 20 0.1514 0.162 T*D Abv+Blw 0 10 0.3317 0.1672 T*D Abv+Blw 10 20 0.2071 0.162 Differences of Least Squares Means Effect Treatment _Treatment t Value Pr > |t| Adj P T Control Abv 1.42 0.1632 0.3406 T Control Abv+Blw 3.51 0.001 0.0029 T Abv Abv+Blw 2.12 0.0397 0.0975 Field Treatment Effect o n 150 53 m Size Fraction mg C g 1 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=54 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 46 8 0.001 D 1 46 1.66 0.2035 T*D 2 46 3.56 0.0366 Least Squares Means Effect Treatment Depth Estimate s.e. T Control 0.1693 0.11 T Abv 0.01488 0.11

PAGE 121

121 Table A 3. Continued Effect Treatment Depth Estimate s.e. T Abv+Blw 0.2977 0.11 D 0 10 0.1005 0.09866 D 10 20 0.02479 0.09866 T*D Control 0 10 0.1713 0.1384 T*D Control 10 20 0.1673 0.1384 T*D Abv 0 10 0.06835 0.1384 T*D Abv 10 20 0.0386 0.1384 T*D Abv+Blw 0 10 0.5411 0.1384 T*D Abv+Blw 10 20 0.05434 0.1384 Differences of Least Squares Means Effect Treatment Depth _Treatmen t _Depth t Value Pr > |t| Adj P T*D Control 0 10 Control 10 20 0.02 0.9813 1 T*D Control 0 10 Abv 0 10 0.61 0.5436 0.9896 T*D Control 0 10 Abv 10 20 1.25 0.2185 0.811 T*D Control 0 10 Abv+Blw 0 10 4.24 0.0001 0.0014 T*D Control 0 10 Abv+Blw 10 20 1.34 0.1864 0.7607 T*D Control 10 20 Abv 0 10 0.59 0.5592 0.9913 T*D Control 10 20 Abv 10 20 1.22 0.2271 0.8228 T*D Control 10 20 Abv+Blw 0 10 4.21 0.0001 0.0015 T*D Control 10 20 Abv+Blw 10 20 1.32 0.1941 0.7738 T*D Abv 0 10 Abv 10 20 0.64 0.5281 0.9877 T*D Abv 0 10 Abv+Blw 0 10 3.62 0.0007 0.0089 T*D Abv 0 10 Abv+Blw 10 20 0.73 0.4695 0.9773 T*D Abv 10 20 Abv+Blw 0 10 2.99 0.0045 0.0483 T*D Abv 10 20 Abv+Blw 10 20 0.09 0.9259 1 T*D Abv+Blw 0 10 Abv+Blw 10 20 2.89 0.005 8 0.0605

PAGE 122

122 Table A 3. Continued Field Treatment Effec t on <53 m Size Fraction mg C cm 3 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=52; 2 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 44 0.88 0.4235 D 1 44 10.92 0.0019 T*D 2 44 4.01 0.0252 Least Squares Means Effect Treatment Depth Estimate s.e. T Control 0.1699 0.1053 T Abv 0.1192 0.1057 T Abv+B lw 0.1033 0.1057 D 0 10 0.0593 0.1031 D 10 20 0.2023 0.1035 T*D Control 0 10 0.01665 0.1115 T*D Control 10 20 0.3231 0.1115 T*D Abv 0 10 0.1114 0.1115 T*D Abv 10 20 0.1271 0.113 T*D Abv+Blw 0 10 0.04985 0.1115 T*D Abv+Blw 10 20 0.1568 0.113 Differences of Least Squares Means Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P T*D Control 0 10 Control 10 20 4.18 0.0001 0.0018 T*D Control 0 10 Abv 0 10 1.29 0.2032 0.7878 T*D Co ntrol 0 10 Abv 10 20 1.46 0.1517 0.6914 T*D Control 0 10 Abv+Blw 0 10 0.45 0.653 0.9974 T*D Control 0 10 Abv+Blw 10 20 1.85 0.0708 0.4445 T*D Control 10 20 Abv 0 10 2.89 0.006 0.0624 T*D Control 10 20 Abv 10 20 2.59 0.013 0.1214

PAGE 123

123 Table A 3 Continued Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P T*D Control 10 20 Abv+Blw 0 10 3.73 0.0006 0.0069 T*D Control 10 20 Abv+Blw 10 20 2.2 0.0334 0.26 T*D Abv 0 10 Abv 10 20 0.21 0.8369 0.9999 T*D Abv 0 10 Abv+ Blw 0 10 0.84 0.4059 0.9584 T*D Abv 0 10 Abv+Blw 10 20 0.6 0.5514 0.9905 T*D Abv 10 20 Abv+Blw 0 10 1.02 0.3132 0.9087 T*D Abv 10 20 Abv+Blw 10 20 0.38 0.7039 0.9989 T*D Abv+Blw 0 10 Abv+Blw 10 20 1.41 0.1646 0.7189 Field Treat ment Effec t on <53 m Size Fraction mg C cm 3 soil as Relative % Change Log(1+ Relative % Change in mg SOC) Transformed n=54 Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 46 3.04 0.0576 D 1 46 25 .2 <.0001 T*D 2 46 5.11 0.0099 Least Squares Means Effect Treatment Depth Estimate s.e. T Control 0.1985 0.09353 T Abv 0.09258 0.09353 T Abv+Blw 0.0796 0.09353 D 0 10 0.01513 0.091 D 10 20 0.232 0.091 T*D Control 0 10 0.01011 0.1007 T*D Control 10 20 0.3868 0.1007 T*D Abv 0 10 0.07266 0.1007 T*D Abv 10 20 0.1125 0.1007 T*D Abv+Blw 0 10 0.03739 0.1007 T*D Abv+Blw 10 20 0.1966 0.1007

PAGE 124

124 Table A 3. Continued Differences of Least Squares Means Effect Treatment Depth _Treatment _Depth t Value Pr > |t| Adj P T*D Control 0 10 Control 10 20 5.03 <.0001 0.0001 T*D Control 0 10 Abv 0 10 0.84 0.4075 0.9591 T*D Control 0 10 Abv 10 20 1.37 0.1778 0.7452 T*D Control 0 10 Abv+Blw 0 10 0.63 0.5287 0.9877 T*D Control 0 10 Abv+Blw 10 20 2.49 0.0164 0.1476 T*D Control 10 20 Abv 0 10 4.2 0.0001 0.0016 T*D Control 10 20 Abv 10 20 3.67 0.0006 0.0079 T*D Control 10 20 Abv+Blw 0 10 5.67 <.0001 <.0001 T*D Control 10 20 Abv+Blw 10 20 2.54 0.0144 0.1331 T*D Abv 0 10 Abv 10 20 0.53 0.5969 0.9945 T*D Abv 0 10 Abv+Blw 0 10 1.47 0.1481 0.684 T*D Abv 0 10 Abv+Blw 10 20 1.66 0.1045 0.5667 T*D Abv 10 20 Abv+Blw 0 10 2 0.051 0.3562 T*D Abv 10 20 Abv+Blw 10 20 1.12 0.2669 0.869 T*D Abv+Blw 0 10 Abv+Blw 10 20 3.13 0.0031 0.034 Field Treatment Effect on AOC as % of 2000 250 m Fraction SOC n=508; 32 outliers removed Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F T 2 446 3.8 0.0231 D 1 446 51.47 <.0001 DE 9 446 152.21 <.0001 T*DE 18 446 0.39 0.9889 T*D 2 446 6.18 0.0022 T*D*DE 27 446 0.65 0.9164 Least Squares Means Effect Treatment Depth EngLevel Esti mate s.e. T Control 23.2609 3.5356

PAGE 125

125 Table A 3. Continued Effect Treatment Depth EngLevel Estimate s.e. T Abv 21.8485 3.5359 T Abv+Blw 23.171 3.5369 D 0 10 21.074 3.5281 D 10 20 24.4462 3.5285 DE 0 12.3 803 3.5884 DE 8 13.6553 3.5919 DE 25 15.6384 3.5914 DE 29 15.8049 3.5918 DE 41 18.1665 3.5952 DE 49 22.0558 3.5903 DE 57 25.7525 3.5868 DE 94 33.88 3.5884 DE 119 34.1233 3.5898 DE 153 36.1443 3 .5899 T*DE Control 0 13.5089 3.7266 T*DE Control 8 14.0883 3.7266 T*DE Control 25 15.3609 3.7396 T*DE Control 29 16.7425 3.7562 T*DE Control 41 17.3555 3.7562 T*DE Control 49 23.1611 3.7562 T*DE Control 57 25.0567 3.7266 T*DE Control 94 34.6667 3.7266 T*DE Control 119 35.4 3.7266 T*DE Control 153 37.2683 3.7266 T*DE Abv 0 10.4387 3.7396 T*DE Abv 8 13.1886 3.7564 T*DE Abv 25 15.0884 3.7396 T*DE Abv 29 15.1322 3.7266 T*DE Abv 41 18.3353 3.7396 T*DE Abv 49 21.1422 3.7266

PAGE 126

126 Table A 3. Continued Effect Treatment Depth EngLevel Estimate s.e. T*DE Abv 57 25.1856 3.7266 T*DE Abv 94 32.8555 3.7396 T*DE Abv 119 32.7281 3.7396 T*DE Abv 153 34.3909 3.7527 T*DE Abv+Blw 0 13.1932 3.7396 T*DE Abv+Blw 8 13.689 3.7527 T*DE Abv+Blw 25 16.4659 3.7527 T*DE Abv+Blw 29 15.54 3.7525 T*DE Abv+Blw 41 18.8089 3.7692 T*DE Abv+Blw 49 21.8641 3.7396 T*DE Abv+Blw 57 2 7.0153 3.7396 T*DE Abv+Blw 94 34.118 3.7396 T*DE Abv+Blw 119 34.2418 3.7525 T*DE Abv+Blw 153 36.7737 3.7396 T*D Control 0 10 22.5191 3.5564 T*D Control 10 20 24.0027 3.5607 T*D Abv 0 10 19.106 3.5575 T*D Abv 10 20 24.5911 3.5604 T*D Abv+Blw 0 10 21.597 3.5628 T*D Abv+Blw 10 20 24.745 3.5591 T*D*DE Control 0 10 0 11.6567 3.9289 T*D*DE Control 0 10 8 13.0233 3.9289 T*D*DE Control 0 10 25 13.5733 3.9289 T*D*DE Control 0 10 29 14.6611 3.9 289 T*D*DE Control 0 10 41 16.4856 3.9289 T*D*DE Control 0 10 49 21.5111 3.9289 T*D*DE Control 0 10 57 25.6344 3.9289 T*D*DE Control 0 10 94 34.5211 3.9289 T*D*DE Control 0 10 119 36.09 3.9289 T*D*DE Control 0 10 153 38.0344 3 .9289

PAGE 127

127 Table A 3. Continued Effect Treatment Depth EngLevel Estimate s.e. T*D*DE Control 10 20 0 15.3611 3.9289 T*D*DE Control 10 20 8 15.1533 3.9289 T*D*DE Control 10 20 25 17.1484 3.9781 T*D*DE Control 10 20 29 18.824 4.04 02 T*D*DE Control 10 20 41 18.2254 4.0402 T*D*DE Control 10 20 49 24.8111 4.0402 T*D*DE Control 10 20 57 24.4789 3.9289 T*D*DE Control 10 20 94 34.8122 3.9289 T*D*DE Control 10 20 119 34.71 3.9289 T*D*DE Control 10 20 153 36.5 022 3.9289 T*D*DE Abv 0 10 0 7.7978 3.9289 T*D*DE Abv 0 10 8 9.1622 3.9289 T*D*DE Abv 0 10 25 11.2478 3.9289 T*D*DE Abv 0 10 29 11.9789 3.9289 T*D*DE Abv 0 10 41 14.8478 3.9289 T*D*DE Abv 0 10 49 18.1667 3.9289 T*D*DE Abv 0 10 57 22.5744 3.9289 T*D*DE Abv 0 10 94 30.3389 3.9289 T*D*DE Abv 0 10 119 30.9884 3.9781 T*D*DE Abv 0 10 153 33.9572 3.9781 T*D*DE Abv 10 20 0 13.0797 3.9781 T*D*DE Abv 10 20 8 17.2149 4.0411 T*D*DE Abv 10 20 25 18.929 3.9 781 T*D*DE Abv 10 20 29 18.2856 3.9289 T*D*DE Abv 10 20 41 21.8228 3.9781 T*D*DE Abv 10 20 49 24.1178 3.9289 T*D*DE Abv 10 20 57 27.7967 3.9289 T*D*DE Abv 10 20 94 35.3722 3.9781 T*D*DE Abv 10 20 119 34.4678 3.9289 T*D*DE Abv 10 20 153 34.8247 3.9781

PAGE 128

128 Table A 3. Continued Effect Treatment Depth EngLevel Estimate s.e. T*D*DE Abv+Blw 0 10 0 10.5997 3.9781 T*D*DE Abv+Blw 0 10 8 11.3297 3.9781 T*D*DE Abv+Blw 0 10 25 13.3622 3.9781 T*D*DE Abv+Blw 0 10 29 14.1622 3.9781 T*D*DE Abv+Blw 0 10 41 17.6456 4.0402 T*D*DE Abv+Blw 0 10 49 19.7872 3.9781 T*D*DE Abv+Blw 0 10 57 25.8684 3.9781 T*D*DE Abv+Blw 0 10 94 33.0659 3.9781 T*D*DE Abv+Blw 0 10 119 33.8597 3.9781 T*D*DE Abv+ Blw 0 10 153 36.2897 3.9781 T*D*DE Abv+Blw 10 20 0 15.7867 3.9289 T*D*DE Abv+Blw 10 20 8 16.0484 3.9781 T*D*DE Abv+Blw 10 20 25 19.5697 3.9781 T*D*DE Abv+Blw 10 20 29 16.9178 3.9781 T*D*DE Abv+Blw 10 20 41 19.9722 3.9781 T*D*D E Abv+Blw 10 20 49 23.9411 3.9289 T*D*DE Abv+Blw 10 20 57 28.1622 3.9289 T*D*DE Abv+Blw 10 20 94 35.17 3.9289 T*D*DE Abv+Blw 10 20 119 34.624 3.9781 T*D*DE Abv+Blw 10 20 153 37.2578 3.9289 Differences of Least Squares Means Effect Treatment Depth EngLevel _Treatment _Depth _EngLevel t Value Pr > |t| Adj P DE 1 2 1.22 0.2248 0.9697 DE 1 3 3.11 0.002 0.0612 DE 1 4 3.26 0.0012 0.0387 DE 1 5 5.46 <.0001 <.0001 DE 1 6 9.27 <.0001 <.0001 DE 1 7 12.96 < .0001 <.0001 DE 1 8 20.73 <.0001 <.0001 DE 1 9 20.86 <.0001 <.0001

PAGE 129

129 Table A 3. Continued Effect Treatment Depth EngLevel _Treatment _Depth _EngLevel t Value Pr > |t| Adj P DE 1 10 22.8 <.0001 <.0001 DE 2 4 2.03 0.0433 0.5808 DE 2 5 4.21 <.0001 0.0013 DE 2 6 7.96 <.0001 <.0001 DE 2 7 11.59 <.0001 <.0001 DE 2 8 19.28 <.0001 <.0001 DE 2 9 19.41 <.0001 <.0001 DE 2 10 21.34 <.0001 <.0001 DE 3 4 0.16 0.8752 1 DE 3 5 2.36 0.0186 0.352 DE 3 6 6.09 <.0001 <.0001 DE 3 7 9.7 <.0001 <.0001 DE 3 8 17.41 <.0001 <.0001 DE 3 9 17.56 <.0001 <.0001 DE 3 10 19.48 <.0001 <.0001 DE 4 5 2.2 0.0281 0.4564 DE 4 6 5.92 <.0001 <.0001 DE 4 7 9.53 <.0001 <.0001 DE 4 8 17.23 <.0001 <.0001 DE 4 9 17.38 <.0001 <.0001 DE 4 10 19.29 <.0001 <.0001 DE 5 6 3.64 0.0003 0.0111 DE 5 7 7.19 <.0001 <.0001 DE 5 8 14.82 <.0001 <.0001 DE 5 9 14.97 <.0001 <.0001 DE 5 10 16.87 <.000 1 <.0001 DE 6 7 3.56 0.0004 0.0149 DE 6 8 11.33 <.0001 <.0001 DE 6 9 11.5 <.0001 <.0001 DE 6 10 13.43 <.0001 <.0001

PAGE 130

130 Table A 3. Continued Effect Treatment Depth EngLevel _Treatment _Depth _EngLevel t Value Pr > |t| Adj P DE 7 9 8.07 <.0001 <.0001 DE 7 10 10.02 <.0001 <.0001 DE 8 9 0.23 0.8156 1 DE 8 10 2.17 0.0304 0.4775 DE 9 10 1.93 0.0543 0.649 T*D Control 0 10 Control 10 20 1.84 0.0665 0.4413 T*D Control 0 10 Abv 0 10 4.31 <.0001 0.00 03 T*D Control 0 10 Abv 10 20 2.57 0.0104 0.106 T*D Control 0 10 Abv+Blw 0 10 1.13 0.2587 0.8684 T*D Control 0 10 Abv+Blw 10 20 2.78 0.0056 0.0618 T*D Control 10 20 Abv 0 10 6.04 <.0001 <.0001 T*D Control 10 20 Abv 10 20 0.71 0.4754 0.980 2 T*D Control 10 20 Abv+Blw 0 10 2.89 0.0041 0.0468 T*D Control 10 20 Abv+Blw 10 20 0.91 0.3649 0.9446 T*D Abv 0 10 Abv 10 20 6.77 <.0001 <.0001 T*D Abv 0 10 Abv+Blw 0 10 3.04 0.0025 0.0301 T*D Abv 0 10 Abv+Blw 10 20 7.01 <.0001 <.0001 T*D Abv 10 20 Abv+Blw 0 10 3.6 0.0004 0.0048 T*D Abv 10 20 Abv+Blw 10 20 0.19 0.8507 1 T*D Abv+Blw 0 10 Abv+Blw 10 20 3.81 0.0002 0.0022

PAGE 131

131 APPENDIX B CONFIDENCE INTERVALS OF RELATIVE PERCENT CHANGE ANALYSIS Table B 1. The 90% confidence limits of the relative percent change in soil bulk density after 31 months of exclusion treatments in an intensively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida Treatment Soil depth cm Lower 90%CI Upper 90% CI Untreated control 13.9% 4.8% Aboveground exclusion 4.6% 14.1% Above plus 2.1% 16.6% belowground exclusion 0 10 3.09% 20.5% 10 20 15.5% 1.9% Table B 2. The 90% confidence limits of the relative percent change in coarse fraction SOC (mg cm 3 ) after 31 mo nths of exclusion treatments in an intensively managed loblolly pine stand growing in an Ultic Alaquod in north central Florida Treatment Soil depth cm Lower 90%CI Upper 90% CI Untreated control 26.9% 23.7% Aboveground exclusion 11.7% 51.9% Above p lus 47.1% 10.4% belowground exclusion 0 10 1.2% 55.8% 10 20 46.8% 17.4% Table B 3 The 90% confidence limits of the relative percent change in fine earth fraction SOC (mg cm 3 ) after 31 months of exclusion treatments in an intensively mana ged loblolly pine stand growing in an Ultic Alaquod in north central Florida Treatment Soil depth cm Lower 90%CI Upper 90% CI Untreated control 19.4% 70.1% Aboveground exclusion 3.8% 35.9% Above plus 18.2% 15.6% belowground exclusion 0 10 22.1% 65.8% 10 20 17.9% 11.1%

PAGE 132

132 Table B 4 The 90% confidence limits of the relative percent increase in SOC (mg cm 3 ) among the soil size fractions by significant main effect or interactions after 31 months of exclusion treatments in an intensively m anaged loblolly pine stand growing in an Ultic Alaquod in north central Florida <2mm data from Chapter 3 Untreated control Aboveground exclusion Above plus belowground excl. Size Fraction Lower 90%CI Upper 90% CI Lower 90%CI Upper 90% CI Lower 90 %CI Upper 90% CI >2 mm* 26.9% 23.7% 11.7% 51.9% 47.1% 10.4% 2000 250 m 9.2% 95.4% 18.1% 47.7% 14.0% 55.0% 250 150 m 11.9% 68.9% 6.5% 80.3% 12.6% 67.9% 150 53 m 19.6% 137.7% 44.7% 63.4% 68.9% 7.0% <53 m 0 10 cm 32. 5% 59.9% 16.0% 98.9% 27.1% 72.6% <53 m 10 20 cm 36.7% 223.9% 13.5% 107.5% 7.3% 122.2%

PAGE 133

133 LIST OF REFERENCES Abbott, I., 1989. The influence of fauna on soil structure. P. 39 50 in Animals in Primary Succession: The Role of Fauna in Reclai med Lands Majer, J.D. (ed). Cambridge University Press, Cambrige. Adegbidi, H.G., E.J. Jokela, N.B. Comerford, and N.F. Barros. 2002. Biomass development of intensively managed loblolly pine plantations growing on Spodosols in the southeastern USA. For. E co. and Manag. 167: 91 102. Adegbidi, H.G., N.B. Comerford, E.J. Jokela, N.F. Barros. 2004. Root development of young loblolly pine in Spodosols in southeast Georgia. Soil Sci. Soc. Am. J. 68:596 604. Ahn, M.Y., A.R. Zimmerman, N.B Comerford., J.O. Sickman and S. Grunwald. 2009. Carbon mineralization and labile organic carbon pools in the sandy soils of a north Florida watershed. Ecosystems. 12:672 685. Albaugh, T. J., H.L. Allen, P.M. Dougherty, L.W. Kress, J.S. King. 1998. Leaf area and above and belowg round growth responses of loblolly pine to nutrient and water additions. For. Sci. 44:317 328. Alvarez, R., and R.S. Lavado. 1998. Climate, organic matter and clay content relationships in the Pampa and Chaco soils, Argentina. Geoderma. 83:127 141. Anderso n J. 1982. Soil respiration. P. 842 845 in Methods of soil analysis. Part 2. 2nd Ed., Page A.L., R.H. Miller, and D.R. Keeney (eds.). Am. Soc. of Ag., Inc., Madison. Angers, D.A., M. Giroux. 1996. Recently deposited organic matter in soil water stable aggr egates. Soil Sci. Soc. Am. J. 60:1547 1551. Angers, D.A., S. Recous, C. Aita. 1997. Fate of carbon and nitrogen in water stable aggregates during decomposition of C 13 and N 15 labeled wheat straw in situ Euro J of Soil Sci. 48:295 300. Azuaje, E.I., N .B. Comerford, W.G. Harris, J.B. Reeves III, S. Grunwald. 2012. Loblolly and slash pine control organic carbon in soil aggregates and carbon mineralization. Forest Ecol. Manag. 263:1 8. Baddeley, J .A. and C.A. Watson. 2005. Influences of root diameter, tr ee age, soil depth and season of fine root survivorship in Prunus avium. Plant Soil. 276:15 22. Baker, J.B. and O.G. Langdon, 1990. Pinus taeda L., Loblolly Pine, pp. 497 512. In R.M. Burns and B.H. Honkala (eds.), Silvics of North America, Vol. 1, Conifer s, U.S.D.A. For. Serv. Agric. Handbk. 654, Washington, D.C.

PAGE 134

134 Bardgett, R.D., Bowman, W.D., Kaufmann, R. & Schmidt, S.K. 2005. A temporal approach to linking aboveground and belowground ecology. Trends Eco l Evol. 20: 634 641. Berg, B. and C. McClaugherty. 2 003. Plant Litter: Decomposition, Humus Formation, Carbon Sequestration 1st edn. Springer, Berlin. Bird, J.A., M.S. Torn. 2006. Fine roots vs. Needles: A comparison of C 13 and N 15 dynamics in a ponderosa pine forest soil Biogeochem 79:361 382. Blazier M. A., W. B. Patterson, S.L. Hotard. 2008 Straw harvesting, fertilization, and fertilizer type alter soil microbiological and physical properties in a lob lolly pine plantation in the mid south USA. Biol. Fert. Soils. 45:145 153. Bonate, P.L. 2000. Analysis of Pretest Posttest Designs Boca Raton: Chapman & Hall/CRC. p. 75 89 B racho, R., G Starr, H L. Gholz, T A. Martin W P. Cropper, H. W. L oescher. 2012 Controls on carbon dynamics by ecosystem structure and climate for southeastern U.S slash pine plantations. Ecol. Monogr. 82: 101 128 Brussaard, L., V.M Behan Pelletier, D.E. Bignell, V.K. Brown, W. Didden, P. Folgarait, C. Fragoso, D.W. Freckman, V.V. S.R. Gupta, T. Hattori, D.L. Hawksworth, C. Klopatek, P. Lavelle, D.W. Malloch, J. Rusek, B. Soderstrom, J.M. Tiedje, R.A. Virginia. 1997. Biodiversity and ecosystem functioning in soil. Ambio. 26:563 570. Brunn, S., G.I. Agern, B.T. Christensen, L.S. Jens en. 2010. Measuring and modeling continuous quality distributions of soil organic matter. Biogeosciences. 7:27 41. Burke, D.J., A.M. Kretzer, P.T. Rygiewicz, M. A. Topa. 2006. Soil bacterial diversity in a loblolly pine plantations: influence of ectomycor rhizas and fertilization. FEMS Microbiol. Ecol. 57:409 419. Burkes, E.C., R.E. Will, G.A. Barron Gafford, R.O. Teskey, B. Shiver. 2003. Biomass partitioning and growth efficiency of intensively managed Pinus taeda and Pinus elliottii stands of different pl anting densities. For. Sci. 49:224 234. Buyanovsky, G.A., Aslam, M. and Wagner, G.H. 1994. Carbon turnover in soil physical Fractions. Soil Sci. Soc. Am. J. 58: 1167 1173. Christensen, B. T. 1992. Physical fractionation of soil and organic matter in prima ry particle size and density separates. Adv. Soil S ci 20: 1 89. Christensen, B.T. 1996. Carbon in primary and secondary organomineral complexes. P. 97 165 in Structure and Organic Matter Storage in Agricultural Soils Carter, M.R. and B.A. Stewart (eds.) CRC Press, Inc., Boca Raton, FL.

PAGE 135

135 Christensen, B. T. 2001. Physical fractionation of soil and structural and functional complexity in organic matter turnover. Eur. J. Soil Sci. 52: 345 353. Clark, K.L., H.L. Gholz, M.S. Castro. 2004. Carbon dynamics alo ng a chronosequence of slash pine plantations in north Florida. Ecol. App.14:1154 1171. Conant, R.T., G.R. Smith, K. Paustain. 2003. Spatial variability of soil carbon in forested and cultivated sites: implications for change detection. J. Environ. Qual. 3 2:278 286. Conant, R. T., J. Six, and K. Paustian. 2004. Land use effects on soil carbon fractions in the southeastern United States: II. Change in soil carbon fractions along a forest to pasture chronosequence. Biol. Fert. Soils 40 :194 200. Czarnes, S., P.D. Hallett, A. G. Bengough, I. M. Young. 2000. Root and microbial derived mucilages affect soil structure and water transport. Eur. J. Soil Sci. 51:435 443. Dosskey, M.G. and P.M. Bertsch. 1997. Transport of dissolved organic matter through a sand fores t soil. Soil Sci. Soc. Am. J. 61:920 927. Echeverra, M.E., Markewitz, D., Morris, L.A., and Hendrick, R.L. 2004. Soil organic matter fractions under managed pine plantations of the southeastern USA. Soil Sci. Soc. Am. J. 68 :950 958. Eissenstat, D.M., C.E Wells, R.D. Yanai, J.L. Whitbeck. 2000. Building roots in a changing environment: implications for root longevity. New Phytol. 147 : 33 42. Escamilla, J.A., Comerford, N.B., Neary, D.G., 1991. Soil core break method to estimate pine root distribution. Soi l Sci. Soc. Am. J. 55: 1722 1726. Ewel, K.C., W.P. Cropper, H.L. Gholz. Soil CO2 evolution in Florida slash pine plantations. II. Importance of root respiration. Can. J. For. Res. 17:330 333. Fan, P. and D. Guo. 2010. Slow decomposition of lower order root s: a key mechanism of root carbon and nutrient retention in the soil. Oecologia. 163:509 515. Farrar, J., M. Hawes, D. Jones, S. Lindow. 2003. How roots control the flux of carbon to the rhizosphere. Ecology. 84:827 837. Fierer, N., J.P. Schimel, P.A. Hold en. 2003. Variations in microbial community composition through two soil depth profiles. Soil Biol. Biochem. 35:167 176. Frberg, M. D. Berggren Kleja, F. Hagedorn. 2007. The contribution of fresh litter to dissolved organic carbon leached from a conifero us forest floor. Eur. J. Soil Sci. 58:108 114.

PAGE 136

136 Frberg, M., P.M. Jardine, P.J. Hanson, D.E. Todd, J.R. Tarver. 2007. Low dissolved organic carbon input from fresh litter to deep mineral soils. Soil Sci. Soc. Am. J. 71:347 354 Gale, W.J., C.A. Cambardella, T.B. Bailey. 2000. Root derived carbon and the formation and stabilization of aggregates. Soil Sci. Soc. Am. J. 64:201 207. Garten, C.T. 2009. A disconnect between O horizon and mineral soil carbon: Implications for soil C sequestration. Oecologica. 35:218 226. Gaudinski, J.B., S.E. Trumbore, E.A. Davidson, A.C. Cook, D. Markewitz, D.D. Richter. 2001. The age of fine root carbon in the three forests of the eastern United States measured by radiocarbon. Oecologia. 129:420 429. Gholz, H.L., L.C. Hendry, W.P. Cropper Jr, 1986. Organic matter dynamics of fine roots in plantations of slash pine (Pinus elliottii) in north Florida. Can. J. For. Res. 16:529 538. Godbold, D.L., M.R. Hoosbeek, M. Lukac, M. F. Cotrufo, I.A. Janssens, R. Ceulemans, A.Polle, E.J. Velthor st, G. Scarascia Mugnozza, P. De Angelis, F. Miglietta, A. Peressotti. 2006. Mycorrhizal hyphal turnover as a dominant process for carbon input into soil organic matter. Plant Soil. 281:15 24. Golchin, A., Oades, J. M., Skjemstad, J. O. and Clarke, P. 1994 Study of free and occluded particulate organic matter in soils by solid state 13C P/MAS NMR spectroscopy and scanning electron microscopy. Aust. J. Soil Res. 32:285 309. Guggenberger, G., W. Zech, H.R. Schulten. 1994. Formation and mobilization pathways of dissolved organic matter: evidence from chemical structural studies of organic matter fractions in acid forest floor solutions. Org. Geochem. 21:51 66. Hagedorn, F., M. Saurer, P. Blaser. 2004. A 13C tracer study to identify the origin of dissolved orga nic carbon in forested mineral soils. Eur. J. Soil Sci. 55:91 100. Haile, S.G, P.K.R. Nair, V.D Nair. 2007. Carbon storage of different soil size fractions in Florida silvopastoral systems. J. Enviro. Qual. 37:1789 1797. Haines, S.G. and G. Cleveland. 1982 Seasonal variation in properties of five forest soils in southwest Georgia. Soil Sci. Soc. Am. J. 45:139 143. Harris, W.G., V.W. Carlisle. 1987. Clay mineralogical relationships in Florida Haplaquods. Soil Sci. Soc. Am. J. 51:481 484. Hass, A., M.G. Mess ina, T.D. Rogers. 2010. Soil carbon after midrotation phosphorus fertilization of l oblolly p ine ( Pinus taeda L.) stands in east Texas Soil Sci. 175:382 389.

PAGE 137

137 Hassink, J. and Whitmore, A. P. 1997. A model of the physical protection of soil organic matter i n soils. Soil Sci. Soc. Am. J. 61: 131 139. Hassink Jan, A.P. Whitmore, J. Kubat. 1997. Size and density fractionation of soil organic matter and the physical capacity of soils to protect organic matter. Eur. J. Agron. 7:189 199. Jenkinson, D.S. and J.N. Ladd. 1981. Microbial biomass in soil: measurement and turnover. P. 415 471 in Soil Biochemistry Paul, E. A. and J. N. Ladd (eds.). Marcel Dekker, NY. Jobbgy, E.G. and R.B. Jackson. 2000. The vertical distribution of soil organic carbon and its relation to vegetation and climate. Ecol. Appl. 10:423 436. Jokela, E.J. and T.A. Martin. 2000. Effects of ontogeny and soil nutrient supply on production, allocation, and leaf area efficiency in loblolly and slash pine stands. Can. J. For. Res. 30:1511 1524. Jokel a, E.J., T.A. Martin, J.G. Vogel. 2010. Twenty five years of intensive forest management with southern pines: important lessons learned. J. For. 108:338 347. Johnston, C.A., P. Groffman, D. D. Breshears, Z. G. Cardon, W. Currie, W. Emanuel, J. Gaudinski, R B. Jackson, K. Lajtha, K. Nadelhoffer, D. Nelson Jr, W. M. Post, G. Retallack, L. Wielopski. 2004. Carbon cycling in soil. Front. Ecol. Environ. 2:522 528. Johnsen, K.H., J.R. Butnor, C. Maier, R. Oren, R. Pangle, L. Samuelson, J. Seiler, S.E. McKeand, H .L. Allen. 2001. Fertilization increases below ground carbon sequestration of loblolly pine plantations. In Proceedings: First national conference on carbon sequestration. Available online: http://www.netl.doe.gov/publications/proceedings/01/carbon_seq/carbon_seq01.ht ml Accessed 06/06/2012. Jorgensen, J.R., C.G. Wells, L.J. Metz. 1980. Nutrient changes in decomposing loblolly pine forest floor. Soil Sci. Soc. Am. J. 44 : 1307 1314. Karltun, E., A.F. Harrison, A. Alriksson, C. Bryant, M.H. Garnett, M.T. Olsson. 2005. Old organic carbon in soil solution DOC after afforestation evidence from C 14 analysis. Geoderma. 127:188 195. Khan, S.A., R.L. Mulvaney, T.R. Ells worth, C.W. Boast. 2007. The myth of nitrogen fertilization for soil carbon sequestration. J. Enviro. Qual. 36:1821 1832. Kindler R, A. Miltner, H.H. Richnow, M. Kastner. 2006. Fate of gram negative bacterial biomass in soil mineralization and contribution to SOM. Soil Biol. Biochem. 38:2860 2870.

PAGE 138

138 Kindler R, A. Miltner, M. Thullner, H. H. Richnow, M. Kastner M. 2009. Fate of bacterial biomass derived fatty acids in soil and their contribution to soil organic matter. Org. Geochem. 40:29 37. King, J.S., H.L. Allen, P. Dougherty, B.R. Strain. 1997. Decomposition of roots in loblolly pine: Effects of nutrient and water availability and root size class on mass loss and nutrient dynamics. Plant Soil. 195:171 184. King, J.S., T.J. Albaugh, H.L. Allen, L.W. Kress. 1 999. Stand level allometry in Pinus taeda as affected by irrigation and fertilization. Tree Phys. 19: 769 778. King, J.S., T.J. Albaugh, H.L. Allen, M. Buford, B.R. Strain, P. Dougherty. 2002. Below Ground Carbon Input to Soil Is Controlled by Nutrient Ava ilability and Fine Root Dynamics in Loblolly Pine. New Phyto. 154:389 398. Knoepp, J.D., D.C. Coleman, D.A. Crossley Jr., J.S. Clark. 2000. Biological indices of soil quality: an ecosystem case study of their use. For. Eco. and Man a g.138:357 368. Kong, A.Y .Y., and J. Six. 2010. Tracing root vs. residue carbon into soils from conventional and alternative cropping systems. Soil Sci. Soc. Am. J. 74 :1201 1210. Kramer, C., S. Trumbore, M. Frberg, L.M.C. Dozal, D. Zhang, X. Xu, G.M. Santos, P.J. Hanson. 2010. Re cent (<4 year old) leaf litter is not a major source of microbial carbon in a temperate forest mineral soil. Soil Biol. Biochem. 42:1028 1037. Kuzyakov, Y. 2011. How to link soil C pools with CO 2 fluxes? Biogeosciences. 8:1523 1537. Leggett, Z.H., and D.L Kelting. 2006. Fertilization effects on carbon pools in loblolly pine plantations on two upland sites. Soil Sci. Soc. Am. J. 70 : 279 286. Littell, R.C., G.A. Milliken, W.W. Stroup, R.D. Wolfinger, O. Schabenberger. 2006. SAS for mixed models 2nd ed. SAS I nstitute Inc., Cary, N.C Lussenhop, J., 1992. Mechanisms of microarthropod microbial interactions in soil. Adv. Ecol. Res. 23:1 33. Ltzow, M. v., I. Kgel Knabner, K. Ekschmitt, E. Matzner, G. Guggenberger, B. Marschner, H. Flessa. 2006. Stabilization of organic matter in temperate soils: mechanisms and their relevance under different soil conditions a review. Eur. J. Soil Sci. 57:426 445. Maier, C.A. and L.W. Kress. 2000. Soil CO2 evolution and root respiration in 11 year old loblolly pine (Pinus taeda) plantations as affected by moisture and nutrient availability. Can. J. For. Res. 30:347 359.

PAGE 139

139 Martin, T.A., E.J. Jokela. 2004. Developmental patterns and nutritional impact radiation use efficiency component in southern pine stands. Ecol. App. 41:1839 1854 Matamala, R., M.A. Gonzales Meler, J.D. Jastrow, R.J. Norby, W.H. schesinger. 2003. Impacts of fine root turnover on forest NPP and soil C sequestration potential. Science. 302:1385 1387. McBrayer, J.F., J.M. Ferris, L.J. Metz, C.S. Gist, B.W. Cornaby, Y Kitazawa, T. Kitazawa, J.G. Wernz, G.W. Krantz, H. Jensen. 1977. Decomposer invertebrate populations in the U.S. forest biomes. Pedobiologia. 17:89 96. McDonald, J.H. 2009. Handbook of biological statistics 2nd ed. Sparky House Publishing, Baltimore, Mar yland. pp. 160 164. McDowell, W.H. and G.E. Likens. 1988. Origin, composition, and flux of dissolved organic carbon in the Hubbard Brook Valley. Ecol. Monogr. 58:177 195. Miltner, A., P. Bombach, B. Schmidt Brcken, M. Kstner. 2011. SOM genesis: microbial biomass as a significant source. Biogeochemistry. Online only. http://dx.doi.org/10.1007/s10533 011 9658 z Accessed 04/04/2012 Nadelhoffer, K. J., R. D. Boone, R. D. Bowden, J. D. Canary, J. Ka ye, P. Micks, A. Ricca, J. A. Aitkenhead, K. Lajtha, W. H. McDowell. 2004. The DIRT experiment: litter and root influences on forest soil organic matter stocks and function. P. 300 315 in Forests in Time: The Environmental Consequences of 1000 Years of Cha nge in New England D. Foster and J. Aber (eds). Synthesis Volume of the Harvard Forest LTER Program. Oxford University Press. NCDC, National Climate Data Center [Internet]. Monthly surface data, Gainesville Regional Airport, Geinesville, Fl 2007 2009. Ava ilable from: http://cdo.ncdc.noaa.gov, accessed October 14, 2011. Nelson, D.W. and L.E. Sommers. 1996. Total carbon, organic carbon, and organic matter P. 961 1010. in Methods of Soil Analysis: Part 3 Chemical Methods SSSA Book Series No. 5, D.L Sparks et al., (eds). Soil Sci. Soc Am. Inc. Madison, WI. Nguyen, C. 2003 Rhizodeposition of organic C by plant: mechanisms and controls. Agronomie. 23: 375 396. Oades, J.M. 1978. Mucilages at the root surface. J. Soil Sci. 29:1 16. Oades, J.M. 1984. Soil organi c matter and structural stability: mechanisms and implications for management. Plant Soil. 76:319 337. Oades, J.M. 1988. The retention of organic matter in soils. Biogeochem. 5:35 70. Oades, J.M. 1993. The role of biology in the formation, stabilization an d degradation of soil structure. Geoderma. 56:377 400.

PAGE 140

140 Oades, J.M., A.G. Waters. 1991. Aggregate hierarchy in soils. Aust. J. Soil Res. 29:815 828. Okada K., T. Satomura A. Kinoshita T. Horikoshi K. Yasue M. Fukuda A. Yamada 2011. Difference of pine ectomycorrhizal biomass in relation to forest conditions Mycoscien ce. 52:59 64. Phillips, D.H. and E.A. Fitz p atrick. 1999. Biological influences on the morphology and micromorphology of selected Podzols (Spodosols) and Cambisols (Inceptisols) from the eastern United States and north east Scotland. Geoderma. 90:327 364. P iatek, K. B. and H.L. Allen. 2000. Are forest floors in mid rotation stands of loblolly pine (Pinus taeda) a sink for nitrogen and phosphorus? Can. J. For. Res. 31:1164 1174. Polglase, P.J., Jokela, E.J., and Comerford, N.B. 1992 a Nitrogen and phosphorus release from decomposing needles of southern pine plantations. Soil Sci. Soc. Am. J. 56 : 914 920. Polglase, P.J., Jokela, E.J., and Comerford, N.B. 1992 b. Phosphorus, nitrogen, and carbon fractions in litter and soil of southern pine plantations. Soil Sci. Soc. Am. J. 56 : 566 572 Ponge, J.F. 1991. Succession of fungi and fauna during decomposition of needles in a small area of Scots pine litter. Plant Soil 138: 99 113. Ponge, J.F., P. Arpin, G. Vannier. 1993. Collembolan response to experimental perturbatio ns of litter supply in a temperate forest ecosystem. Eur.J. Soil Bio. 29:141 153. Pregitzer, K.S. 2002 Fine roots of trees: a new perspective. New Phytol. 154:267 273. Pregitzer, K.S., J.I. Deforest, A.J. Burton, M.F. Allen, R.W. Ruess, R. l. Hendrick. 20 02b. Fine root architecture of nine North American trees. Ecol. Mono. 72:293 309. Puget, P., C. Chenu, and J. Balesdent. 1995. Total and young organic matter distributions in aggregate of silty cultivated soils. Eur. J. Soil Sci. 46:449 459. Quideau, S.A., O.A. Chadwick, S.E. Trumbore, J.L. Johnson Maynard, R.C. Graham, M.A. Anderson. 2001. Vegetation control on soil organic matter dynamics. Org. Geochemistry. 32:247 252. Rasse, D.P., C. Rumpel, M.F. Dignac. 2005. Is soil carbon mostly root carbon? Mechanis ms for a specific stabilization. Plant Soil. 269:341 356.

PAGE 141

141 tree biomass and carbon allocation of juvenile trees of loblolly pine (Pinus taeda): influence of genetics and ferti lization. Can. J. For. Res. 31:960 970. Richter, D.D., D. Markewitz, J.K. Dunsomb, P.R. Heine, C.G. Wells, A. Stuanes, H.L. Allen, B. Urrego, K. Harrison, G. Bonani, W.W McFee. 1995. Carbon cycling in a loblolly pine forest: implications for the missing ca rbon sink and for the concept of soil. P233 251 in Carbon Forms and Functions in Forest Soils, Kelly, J.M. (ed.), Soil Sci. Soc. Am Madison, WI. Richter, D.D., D. Markewitz, S.E. Trumbore, C.G. Wells. 1999. Repid accumulation and turnover of soil carbon in a re establishing forest. Nature. 400:56 58. Rifai, S.W., D. Markewitz, B. Borders. 2010. Twenty years of intensive fertilization and competing vegetation suppression in loblolly pine plantations: impacts on soil C, N, and microbial biomass. Soil Biol. Biochem. 42 : 713 723. Ross, S., W.H. McKee, M. Mims. 1994. Loblolly and longleaf pine responses to litter raking, prescribe burning and nitrogen fertilization. P. 220 224 in The Eighth Biennial Southern Silvicultural Research Conference, M. B. Edwards (ed. ), USDA Forest Service, Asheville. Roth, B.E., E.J. Jokela, T.A. Martin, D.A. Huber, T.L. White. 2007. Family x environment interactions for selected loblolly and slash pine plantations in the southeastern United States. For. Eco. and Man a g 238:175 188. S amuelson, L., R. Mathew, T. Stokes, Y. Feng, D. Aubrey, C. Coleman. 2009. Soil and microbial respiration in a loblolly pine plantation in response to seven years of irrigation a nd fertilization. For. Eco. and Man a g. 258:2431 2438. Sanchez, F.G. 2001. Loblo lly pine needle decomposition and nutrient dynamics as affected by irrigation, fertilization, and substrate quality. For. Eco. Manag. 152:85 96. Sarkhot, D.V., N.B. Comerford, E.J. Jokela, J.B. Reeves III, and W.G. Harris. 2007a. Aggregation and aggregate carbon in a forested southeastern coastal plain spodosol. Soil Sci. Soc. Am. J. 71: 1779 1787. Sarkhot, D.V., N.B. Comerford, E.J. Jokela, and J.B. Reeves III. 2007b. Effects of forest management intensity on carbon and nitrogen content in different s oil size fractions of a north Florida spodosol. Plant Soil 24:291 303. Sarkhot, D.V., E.J. Jokela, N.B. Comerford. 2008. Surface soil carbon size density fractions altered by loblolly pine families and forest management intensity for a Spodoso l in the sout heastern US. Plant Soil. 307:99 111. Sayer E.J, J.S. Powers, E.V.J. Tanner. 2007. Increased Litterfall in Tropical Forests Boosts the Transfer of Soil CO 2 to the Atmosphere. PLoS ONE 2:e1299.

PAGE 142

142 Schimel, D.S., B.H. Braswell, E.A. Holland, R. McKeown, D.S. Oj ima, T.H. Painter, W.J. Parton, A.R. Townsend. 1994. Climatic, edaphic and biotic controls over storage and turnover of carbon in soils. Global Biogeochem. Cy. 8:273 293 Schmidt, M.W.I., C. Rumpel, and I. Kgel Knabner.1999. Evaluation of an ultrasonic d ispersion procedure to isolate primary organomineral complexes from soils. Eur. J. Soil Sci. 50: 87 94. Schmidt, M.W.I., M.S. Torn, S. Abiven, T. Dittmar, G. Guggenberger, I. A. Janssens, M. Kleber, I. Kgel Knabner, J. Lehmann, D. A. C. Manning, P. Nannip ieri, D.P. Rasse, S Weiner, S.E. Trumbore. 2011. Persistence of soil organic matter as an ecosystem property. Nature. 474:49 56. Shan J., L.A. Morris, R.L. Hendrick. 2001. The effects of management on soil and plant carbon sequestration in slash pine plan tations. J. App. Ecol. 38:932 941. Simpson, A.J., M.J. Simpson, E. Smith, B.P. Kelleher. 2007. Microbial ly derived inputs to soil organic matter: are current estimates too low? Environ. Sci. Technol. 41:8070 8076. Six J., E.T. Elliott, K. Paustian. 2000. S oil macroaggregate turnover and microaggregate formation: A mechanism for C sequestration under no tillage agriculture. Soil Biol. Biochem. 32: 2099 2103. Six, J., R.T. Conant, E.A. Paul, and K. Paustian. 2002. Stabilization mechanisms of soil organic matt er: Implications for C saturation of soils. Plant Soil. 241:155 176. Soil Survey Staff, Natural Resources Conservation Service, United States Department of Agriculture. Web Soil Survey. Available online at http://websoilsurvey.nrcs.usda.gov/. Accessed 10/0 1/2011. Sword, M.A., 1998. Seasonal development of loblolly pine lateral roots in response to stand density and fertilization. Plant Soil. 200:21 25. Tegen, I. and H. Drr. 1996. 14C measurements of soil organic matter, soil CO2 and dissolved organic carbo n (1987 1992). Radiocarbon. 38:247 251. Telles, E.C.C. 2003. Influence of soil texture on carbon dynamics and storage potential in tropical forest soils of Amazonia. Global Biogeochem. Cy. 17:1040 1052. Thomas, W.A. 1968. Decomposition of loblolly pine nee dles with and without addition of dogwood leaves. Ecology. 49:568 571. Tisdall J.M. and J.M. Oades. 1982. Organic matter and water stable aggregates in soils. J. Soil Sci. 33:141 163. Torn, M.S., S.E. Trumbore, O.A. Chadwick, P.M. Vitousek, D.M. Hendricks. 1997. Mineral control of soil organic carbon storage and turnover. Nature. 389: 170 173.

PAGE 143

143 Trumbore, S.E., S.L.Schiff, R. Aravena, R. Elgood. 1992. Sources and transformation of dissolved organic carbon in the Harp Lake forested catchment: the role of soils. Radiocarbon. 34:626 635. U.S. Drought Monitor. National Drought Mitigation Center (NDMC), Center at the University of Nebraska Lincoln, contact Brian Fushs. U.S. summary maps animation 2006 2009. Available online: http://drought.unl.edu/dm Accessed 12/6/11. Van Rees, K.C.J., and N.B. Comerford. Vertical root distribution and strontium uptake of a slash pine stand on a Florida Spodosol. Soil Sci. Soc. Am. J. 50:1042 1046. Vogel, J.G., L.J. Suau, T.A. Martin, E.J Jokela. 2011. Long term effects of weed control and fertilization on the carbon and nitrogen pools of a slash and loblolly pine forest in north central Florida. Can. J. For. Res. 41:552 567. Wiseman, P.E. and J.R. Seiler. 2004. Soil CO 2 efflux across four age classes of plantation loblolly pine ( Pinus taeda L.) on the Virginia Piedmont. For. Eco. And Man a g.192:297 311.

PAGE 144

144 BIOGRAPHICAL SKETCH After graduating high school and leaving the high altitude Wet Mountain Valley of Colorado, the au thor began on a winding tour through interesting colleges, courses, and part time jobs. Finding her heart in the landscape, her formal post secondary education desire to understan d the landscape as the functioning ecosystem it is began to surface and lead her to the Soil and Water Science Department and The Soil and Water Science Department became her home for the completion of her Bachelor of Science degree which included a scholarship for undergraduate research and a part time tech position in the UF Forest Soils Laboratory. The graduation was followed by a promotion to become the Forest Soils Laboratory f ull time chemist and general manager. This was soon followed by acceptance to graduate school in the Soil and Water Science Department to study soil carbon development in forest soils. The Forest Soils Lab was a hub of collaboration between students, profe ssors and visiting scientists and offered an uncountable number of exceptional professional and personal experiences under the leadershi p of Dr. Comerford. Most notable accomplishments were an intense watershed soil nutrient characterization, a soil carbon study insulation in southern Brazil, and a state wide soil collection and soil carbon analysis of over major land use/soil suborder combinations. Upon closure of the Forest Soils Lab, her adventure continued with a move to the UF/IFAS North Florida Research and Education Center and a position assisting soil

PAGE 145

145 research/extension scientist, Dr. Mackowiak. The focus moved from forest soils to agricultural and environmental topics pertaining to the Big Bend and Panhandle areas of north Florida. The unique setting of UF/IFAS research centers fosters interaction s with a wide range of scientific disciplines an d proved to be a wonderful opportunity to encounter other areas of research. Choosing to focus on finishing the graduate p rogram, the author became a full time student with an internship with the Alachua County Environmental Protectio n ivision. This has proven to be a beneficial addition to her base of knowledge and rounds out her soil a nd water degree.