Physiological Consequences of Late Leaf Spot on Peanut (Arachis hypogaea L.) Cultivars of Differing Resistance

MISSING IMAGE

Material Information

Title:
Physiological Consequences of Late Leaf Spot on Peanut (Arachis hypogaea L.) Cultivars of Differing Resistance
Physical Description:
1 online resource (138 p.)
Language:
english
Creator:
Singh,Maninderpal
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Agronomy
Committee Chair:
Erickson, John E.
Committee Members:
van Bruggen, Ariena H.
Boote, Kenneth J
Tillman, Barry L
Jones, James W

Subjects

Subjects / Keywords:
breeding -- crop -- cropgro -- development -- disease -- growth -- late -- leaf -- leafspot -- management -- metabolism -- modeling -- pathology -- peanut -- photosynthesis -- physiology -- plant -- spot
Agronomy -- Dissertations, Academic -- UF
Genre:
Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
In the southeastern United States, late leaf spot (LLS) caused by Cercosporidium personatum is one of the most widespread foliar diseases which can substantially reduce peanut yields. Crop protection strategies based on minimizing crop losses by studying host response to the pathogen rather than minimizing disease outbreaks offer a promising way to reduce fungicide use and improve cultivar selection procedures. Field experiments were conducted over two years at Citra, FL evaluating two peanut cultivars with more (York) and less (Carver) quantitative resistance to LLS, grown under fungicide sprayed and non-sprayed conditions. In the first objective, it was observed that disease severity based on canopy lesion area was reduced by 30% in York compared to Carver. No additive effects of combining the resistant cultivar with fungicide were seen, as fungicide use increased yield by 364 kg/ha for both cultivars. Despite reduced disease severity, pod yield gain was only 6% in York compared to Carver. In the second objective, leaf photosynthesis (Asat) data was analyzed using a non-linear model, y = (1 - x)^beta, where y is relative Asat, x is measured visual lesion area, and beta represents the relationship between virtual and visual lesion area. Progression of LLS severity on leaf cohorts was slower in York compared to Carver. However, the reduction in Asat with leaf cohort age was similar across the cultivars. This paradox could be explained by a higher beta value in York (4.6) compared to Carver (3.6), indicating a greater relative reduction in Asat beyond the necrotic lesion area in York. This greater reduction in Asat in York compared to Carver was most closely related to a reduction in maximum carboxylation velocity and chlorophyll. In the third objective, the CROPGRO-Peanut model was successfully used to simulate the observed leaf, pod, and total dry biomass over time when inputs on percent necrosis and defoliation were provided. Visual disease ratings were also used to derive necrosis and defoliation values to simulate LLS-induced growth and yield reductions. Results from this study indicated that future efforts to improve LLS resistance should include sustaining Asat under LLS infection along with slower disease progress.
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 Maninderpal Singh.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Erickson, John E.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043219:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 PHYSIOLOGICAL CONSEQUENCES OF LATE LEAF SPOT ON PEANUT ( Arachis hypogaea L.) CULTIVARS OF DIFFERING RESISTANCE By MANINDERPAL SINGH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL F ULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Maninderpal Singh

PAGE 3

3 T o my m om

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my sincere appreciation to my advisor Dr. John Erickson for giving me the opportunity to study at the University of Florida. I could not have asked for a more knowledgeable, supportive, or friendly advisor. I am also thankful to Dr. Ken Boote, whose understanding of crop physiology seemed endless and his assistance alway s exceeded expectation. I would also like to extend thanks to my committee members Dr. Barry Tillman, Dr. Jim Jones, and Dr. Ari ena van Bruggen for their help. I would also li ke to thank Mr. Justin McKinney, Mr. Andrew Schreffler and Mr. Phillip Alderman for all their technical support throughout the course of my research project. Help from Ja net, Johnathan, Amy, and CJ in conducting field work and data collection is also acknowledged. I would like to thank my friends Pulkit, Shekhar, Deol, Sunil, Hardev, MD Josan Gurreet, Gurpreet, and Jugpreet who brought a lot of humor and normalcy to everyday life Special thanks to Pratibha, Gungeet, Anu, Sandeep, Namrata, and Raman for their support and delicious food. And lastly, but far from least, I would lik e to thank my parents Joginder Kaur and Kehar Singh and brother Sarbrinder Singh ( Tony ) and his wife Amaninder Kaur Their support has been unwavering, and it is a blessing to hav e such loving people in my life.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURES ........................................................................................................ 10 LIST OF ABBREVIATIONS ........................................................................................... 12 ABSTRACT ................................................................................................................... 13 CHAPTER 1 INTRODUCTION .................................................................................................... 15 2 LITERATURE REVIEW .......................................................................................... 19 Background ............................................................................................................. 19 Peanut Diseases ..................................................................................................... 21 Late Lea f Spot ........................................................................................................ 21 Symptoms ........................................................................................................ 22 Disease Cycle .................................................................................................. 22 Management Strategies .......................................................................................... 23 Cultural Practices ............................................................................................. 24 Crop rotation .............................................................................................. 24 Planting date .............................................................................................. 24 Tillage ........................................................................................................ 25 Fungicide Application ....................................................................................... 25 Cultivar Selection ............................................................................................. 26 Effects of Late Leaf Spot on Peanut Physiology ..................................................... 27 Simulations of Late Leaf Spot Damage .................................................................. 31 The CROPGRO Model ..................................................................................... 31 Model Evaluation .............................................................................................. 32 3 LATE LEAF SPOT EFFECTS ON GROWTH, PHOTOSYNTHESIS, AND YIELD IN PEANUT CUL TIVARS OF DIFFERING RESISTANCE ...................................... 35 Abstract ................................................................................................................... 35 Background ............................................................................................................. 36 Mate rials and Methods ............................................................................................ 38 Experimental Site and Design .......................................................................... 38 Measures of Disease Severity and Growth ...................................................... 40 Measures of Canopy Photosynthesis and Yield ............................................... 41 Statistical Analysis ............................................................................................ 42 Results .................................................................................................................... 43

PAGE 6

6 Growth Environment ......................................................................................... 43 Disease Assessment ........................................................................................ 44 Plant Growth and Development ........................................................................ 44 Pod Yield and Quality ....................................................................................... 46 Discussion .............................................................................................................. 46 4 PHOTOSYNTHETIC CONSEQUENCES OF LATE LEAF SPOT DIFFER BETWEEN TWO PEANUT CULTIVARS WITH VARIABLE LEVELS OF RESISTANCE ......................................................................................................... 60 Abstract ................................................................................................................... 60 Background ............................................................................................................. 60 Material and Methods ............................................................................................. 64 Experimental Site and Design .......................................................................... 64 Measures of Asa t on Tagged Leaf Cohorts over Time ....................................... 65 Relations between Photosynthesis and Disease Severity ................................ 66 Data Analysis ................................................................................................... 68 Results .................................................................................................................... 68 Disease Severity and Asat on Tagged Leaf Cohorts over Time ......................... 68 Relatio ns between Photosynthesis and Disease Severity ................................ 69 Discussion .............................................................................................................. 71 5 USING THE CROPGRO PEANUT MODEL TO SIMULATE GROWTH AND YIELD IN PEANUT CULTIVARS WITH VARIABLE RESISTANCE LEVELS TO LATE LEAF SPOT .................................................................................................. 80 Abstract ................................................................................................................... 80 Background ............................................................................................................. 81 Materials and Methods ............................................................................................ 84 Experimental Site and Design .......................................................................... 84 Measures of Growth and Yield ......................................................................... 85 Measures of Disease Injury .............................................................................. 86 Description of the CROPGRO Peanut Model ................................................... 88 CROPGRO Peanut Model Inputs ..................................................................... 90 Procedure for Calibration of Genetic Coefficients ............................................. 90 Procedure for Simulating Disease Effects ........................................................ 91 Statistical Evaluation of Model Performance .................................................... 92 Results and Discussion ........................................................................................... 93 Leaf Weight and Leaf Area Simulations ........................................................... 93 Simulations of Total Biomass and Pod Weight ................................................. 94 Simulations of Canopy Photos ynthesis ............................................................ 96 Simulations with the Modified Model ................................................................ 97 Estimating Disease Induced Percent Necrosis and Defoliation from Florida 1 10 Scale ..................................................................................................... 97 Simulations Using Estimated Necrosis and Defoliation from Florida 110 Scale ............................................................................................................. 98 Conclusions ............................................................................................................ 99

PAGE 7

7 6 SUMMARY AND CONCLUSIONS ........................................................................ 117 APPENDIX A DEFOLIATION, NECROSIS, DRY BIOMASS, AND CANOPY PHOTOSYNTHESIS VALUES FOR CARVER AND YORK .................................. 120 B LEAF SPOT RATING SCALES ............................................................................ 126 LIST OF REFERENCES ............................................................................................. 129 BIOGRAPHICAL SKETCH .......................................................................................... 138

PAGE 8

8 LIST OF TABLES Table page 3 1 Fungicide spray schedule for the field experiments at Citra, FL ......................... 51 3 2 Treatment means and analysis of variance results for measured variables for Carver and York ................................................................................................ 52 3 3 Treatment means for leaf lifespan of leaf cohorts tagged at different times thr oughout the growing season. ......................................................................... 53 5 1 Genetic coefficients of the cultivars Carver, Florunner, York, and Southern Runner used for model simulations. ................................................................. 101 5 2 Root mean square error a nd index of agreement values for l eaf dry weight for cultivars Carver and York. ................................................................................ 102 5 3 Root mean square error and index of agreement values f or total biom ass for cultivars Carver and York. ................................................................................ 103 5 4 Root mean square error a nd index of agreement values for pod weight for cultivars Carver and York. ................................................................................ 103 5 5 Virtual lesion effect o n root mean square error values for total biomass and pod yield for York .............................................................................................. 104 5 6 Root mean square error a nd index of agreement values for total canopy photosynthesis for Carver and York. ................................................................ 104 5 7 Comparison of default vs. modified model for root mean square error a nd index of agreement for total biomass ............................................................... 105 5 8 Comparison of default vs. modified model for root mean square error and ind ex of agreement for pod weight .................................................................. 105 5 9 Statistics of total biomas s using disease function derived f rom measured and estimated data ................................................................................................. 106 5 10 Statistics of pod yield using disease function derived from measured and estimated data. ................................................................................................. 106 A 1 Calculated percent canopy defoliation and necrosis values for Carver and York during 2008 ............................................................................................. 120 A 2 Calculated percent canopy defoliation and necros is values for Carver and York during 2009. ............................................................................................. 1 21

PAGE 9

9 A 3 Leaf, stem, pod, and total dry weight and leaf area index values vs. days after planting during 2008. ................................................................................ 122 A 4 Leaf, stem, pod, and total dry weight and leaf area index values vs. days after planting during 2009. ................................................................................ 123 A 5 Mid day t otal canopy photosynthesis for peanut c ultivars Carver and York during 2008 and 2009. ...................................................................................... 124 B 1 Florida 110 rating scale. .................................................................................. 126

PAGE 10

10 LIST OF FIGURES Figure page 2 1 Disease cycle of late leaf spot, caused by Cercosporidium personatum (Berk. and Curt.) Deighton. ........................................................................................... 34 3 1 Peanut leaf with leaf spot disease. ..................................................................... 54 3 2 Average daily temperature, relative humidity, and cumulative rainfall for field experiment s during the study period. .................................................................. 55 3 3 Progress of la te leaf spot as estimated with the Florida 1 to10 scale and percent canopy lesion area. ............................................................................... 56 3 4 Leaf, stem, pod, and total dry matter accumulation over time for peanut cultivars Carver and York .................................................................................. 57 3 5 Mid day total canopy photosynthesis for two peanut cultivars Carver and York. ................................................................................................................... 58 3 6 R elationships between pod yield and the standardized area under the disease progress curve. ..................................................................................... 59 4 1 Progress of late leaf spot severity (percent necrotic lesion area) on the individual leaf cohorts ........................................................................................ 76 4 2 Light saturated CO2 assimilation rate of individual leaf cohorts for the two peanut cultivars Carver and York. ...................................................................... 77 4 3 Relative light saturated leaf C O2 assimilation rate in relation to disease severity for Carver and York .............................................................................. 78 4 4 Changes in photosynthetic parameters of leaves at no, low, and high disease categories for Carver and York. .......................................................................... 79 5 1 Relationship between necrotic area and effective leaf area in the default and modified model. ............................................................................................... 107 5 2 Simulated and measured leaf dry weight over time for peanut cultivars Carver and York during 2008. ........................................................................... 108 5 3 Simulated and measured leaf dry weight over time for peanut cultivars Carver and York during 2009. ........................................................................... 109 5 4 Simulated and measured total biomass over time for peanut cultivars Carver and York .......................................................................................................... 110

PAGE 11

11 5 5 Simulated and measured pod weight over time for peanut cultivars Carver and York. .......................................................................................................... 111 5 6 Simulated and measured midday total canopy photosynthesis for peanut cultivars Carver and York ................................................................................ 112 5 7 Simulated and measured total biomass and pod weight for Carver and York with default and modified model. ...................................................................... 113 5 8 The relationship between percent necrosis and Florida 110 visual rating scale. ................................................................................................................ 114 5 9 The relationship between defoliation and Florida 110 visual rating scale for the two peanut cultivars Carver and York. ........................................................ 115 5 10 Simulated and measured total biomass and pod weight for Carver and York with measured and estimated disease function. ............................................... 116 A 1 Relationship between mi ssing nodes on the mainst em and defoliation for Carver and York. .............................................................................................. 125 B 1 ICRISAT diagrammatic scale to es timate percent leaflet necrosis. .................. 127 B 2 Determination of percent necrotic leaf area using ASSESS ver 2.0 image analysis software. ............................................................................................. 128

PAGE 12

12 LIST OF ABBREVIATION S Asat Light saturated leaf CO2 assimilation rate ANOVA Analysis of Variance DAP Days Af ter Planting ELS Early Leaf Spot (caused by Cercospora arachidicola) Fv/ Fm Maximum efficiency of PSII photochemistry after dark adaptation LLS Late Leaf Spot ( caused by Cercosporidium personatum ) PPFD Photosynthetic Photon Flux Density stAUDPC Standardized Area Under the Disease Progress Curve TCP Total Canopy Photosynthesis TSMK Total Sound Mature Kernels TSWV Tomato Spotted Wilt Virus Vc,max Maximum carboxylation velocity of Rubisco CO2 Quantum efficiency of CO2 assimilation

PAGE 13

13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PHYSIOLOGICAL CONSEQUENCES OF LATE LEAF SPOT ON PEANUT ( Arachis hypogaea L.) CULTIVARS OF DIFFE RING RESISTANCE By Maninderpal Singh August 2011 Chair: John Erickson Major: Agronomy In the southeastern Unit ed States, late leaf spot (LLS) caused by Cercosporidium personatum is one of the most widespread foliar diseases, which can substantially reduce peanut yields, unless controlled by regular and costly fungicide applications. Crop protection strategies based on minimizing crop losses by studying host response to the pathogen rather than minimizing disease outbreaks offer a promising way to reduce fungicide use and improve cultivar selection procedures. The overall goal of this study was to characterize LLS severity and progression and its impact on growth, yield and photosynthetic metabolism o f peanut cultivars with differing levels of resistance t o LLS. Field experiments were conducted over two years at Citra, FL evaluating two peanut cultivars with more (York) and less (Carver) quantitative resistance to LLS grown under fungicide sprayed and nonsprayed conditions. The first objective of this stu dy was to quantify the effects of LLS on growth and yield in peanut cultivars of differing resistance. Disease severity based on canopy lesion area was reduced by 30% in York compared to Carver. No additive effects of combining the resistant cultivar with fungicide were seen, as fungicide use increased yield by 364 kg ha1 for both cultivars. Yield was more strongly related to disease severity based on canopy lesion area than to

PAGE 14

14 the Florida scale. Despite reduced disease severity, pod yield gain was only 6% in York compared to Carver. The second objective of this study was to quantify the effects of LLS on leaf photosynthetic traits in peanut cultivars. To analyze the leaf photosynthesis ( Asat) data, a nonlinear model, y = (1 x) was used, where y is relative Asat, x is measured visual lesion area, and represents the relationship between virtual and visual lesion area. Progression of LLS severity on leaf cohorts was slower in York compared to Carver. However, the reduction in Asat with leaf cohort age was similar across the cultivars. This paradox could be explained by a higher value in York (4.6) compared to Carver (3.6), indicating a greater relative reduction in Asat beyond the necrotic lesion area in York. This greater reduction in Asat in York compared to Carver was most closely related to a reduction in maximum carboxylation velocity and chlorophyll. The third objective of this study was to simulate growt h and yield as affected by LLS on peanut cultivars of differing resi stance. The CROPGRO Peanut model was able to simulate the observed leaf, pod, and total dry biomass over time when input s on percent necrotic leaf area and defoliation were provided. Correlations among measured defoliation and necrotic leaf a rea with visual disease ratings indicated that visual disease ratings could be successfully used to estimate necrosis and defoliation and to correctly simulate LLS induced reductions in growth and yield. Results from this study indicated that future efforts to improve L LS resistance should include sustaining Asat (i.e. lower value) under LLS infection alo ng with slower disease progress.

PAGE 15

15 CHAPTER 1 INTRODUCTION Peanut ( Arachis hypogaea L.) is one of the major sources of protein and oil in the world. It is cultivated on 24 million hectares in over 100 countries, generating an annual production of nearly 37 Tg (FAO, 2011). Nevertheless, worldwide peanut production is severely hampered by the incidence of numerous diseases. Early leaf spot (caused by Cercospora arachidicola S. Hori), and late leaf spot [caused by Cercosporidium personatum (Berk. & Curt.) Deighton] are among the most widespread and damaging foliar diseases of peanut throughout the world (McDonald et al., 1985). Pod yield losses of up to 50% are common when fungicides are not applied. These losses may reach as high as 70% when disease is not controlled (Shokes and Culbreath, 1997). In Florida, late leaf spot (LLS) is the predominant foliar disease (Jackson, 1981), causing yield losses of up to 50% (Pixley et al., 1990a). Consequently, regular and costly fungicide ap plications are currently used to minimize yield losses from peanut diseases (Woodward et al., 2008; Monfort et al., 2004). Improved cultivars with moderate resistance to late leaf spot, along with other integrated disease management practices, have also be en successfully used to reduce inputs and production costs (Woodward et al., 2010; Woodward et al., 2008; Monfort et al., 2004). However, the effects of LLS on the physiological responses in cultivars of differing leaf spot resistance is not well understood and could contribute to better identification of improved cultivars in breeding programs and better modeling estimates of peanut yield loss to LLS. Moreover, economic and environmental concerns of fungicide use have increased the demand for improved management strategies based on minimizing crop losses rather than minimizing disease outbreaks. Decision support

PAGE 16

16 systems that can predict yield losses rather than controlling the outbreak of disease and hence provide improved disease management options are required to meet this demand. Late leaf spot first occurs as necrotic lesions on peanut leaflets and subsequently induces leaflet abscission. Premature loss of green leaf area (by necrotic tissue and defoliation) and reduction of leaf photosynthetic capacity due to LLS contribute to a loss of canopy carbon assimilation, and thus a loss of yield. Many older peanut cultivars such as Florunner and Georgia Green have poor resistance to LLS. Loss of leaf area due to accelerated senescence was reported to be the pr edominant m echanism of yield losses in thos e cultivars (Bourgeois and Boote, 1992; Boote et al., 1980). However, for cultivars with improved resistance to LLS that experience less defoliation, yield reduction may be affected to a greater extent by the leaf physiologic al response to disease rather than primarily to loss of leaf area. Fungal pathogens generally reduce leaf photosynthesis not only through a reduction in green leaf area, but also through an effect on photosynthetic capacity of the remaining green leaf tissue (Bastiaans 1991). In order to relate reductions in leaf photosynthesis to visual lesion area, Bastiaans (1991) proposed a relatively simple model, y = (1 x), where y is the relative net assimilation rate of a diseased leaf compared to that of an asymptomatic leaf, x is the measured visual lesion area, and describes the relationship between virtual and visual lesion area. The virtual area represents loss of photosynthetic capacity beyond the visual lesion area. Thus, indicates whether the effect of disease on photosynthesis is higher ( > 1), lower ( < 1), or equal ( = 1) to that accounted for by the measured visual lesion area. Using this

PAGE 17

17 model, several studies have shown that the reduction in photosynthesis occurred beyond the measured lesion area (Bassanezi et al., 2002; Erickson et al., 2003; Zhang et al., 2009). U nderstanding why values differ, and perhaps even more importantly, whether they differ by cultivar within species is critical for improved cultivar selection and for modeling effects of disease on carbon assimilation, growth and yield ( Bastiaans, 1993; Adomou et al., 2005; B ancal et al., 2007). Some studies have reported variation in valu e within cultivars of the same species (Erickson et al., 2003; Zhang et al., 2009). Despite the importance of LLS in peanut production, there are no published values for newly released cultivars or comparisons among peanut cultivars. Crop models are esse ntial tools to evaluate growth and yield losses due to various biotic and abiotic stresses (Boote et al., 1983 a ; Naab et al., 2004; Timsina et al., 2007). Models used to predict the impact of foliar diseases on yield have generally incorporated the disease effects on defoliation and photosynthesis ( Batchelor et al., 1993; Teng et al., 1998). Quantification of the effect of LLS on photosynthetic metabolism of peanut cultivars with variable resistance levels to LLS and its inclusion in yield loss model s is of great importance for a more complete understanding of growth and yield responses to diseases and improved accuracy of crop models The CROPGRO Peanut model (Boote et al., 1998a, 1998b) is a process oriented mechanistic crop growth model which considers cr op carbon balance, crop and soil N balance, and soil water balance at the process level. This model has coupling points and procedures for entering pest damage to simulate growth and yield reductions associated with foliar pathogens like LLS (Batchelor et al., 1993; Boote et al., 1993). The primary impacts of disease are simulated as defoliation; however the impacts of

PAGE 18

18 virtual lesion area are currently simulated by simply defoliating more leaf area rather than creating direct impact at the leaf level photos ynthesis This subroutine has not been subjected to rigorous testing as previous studies testing this subroutine did not have measured data on necrosis and/or defoliation (Naab et al., 2004; Adomou et al., 2005). Further, modification may be warranted in t h e model to directly impact the leaf level photosynthetic metabolism as observed in the field (Bastiaans, 1991; Bourgeois and Boote; 1992) rather than defoliating more leaf area. The overall goal of this study was to characterize variability in LLS severi ty and progression and its impact on growth, yield and photosynthetic metabolism of peanut cultivars with variable levels of resistance to LLS. More specifically, this study was conducted with the following objectives: To characterize LLS severity and its effects on growth and partitioning, leaf lifespan, canopy photosynthesis, and pod yield of York, a relatively resistant cultivar, compared to Carver, a cultivar with relatively poor resistance to LLS in a field environment. To quantify the effects of LLS on leaf photosynthetic metabolism in two peanut cultivars with variable levels of resistance to LLS. To evaluate the CROPGRO Peanut model for its ability to simulate the impacts of LLS on growth and yield reduct ions in peanut cultivars with differing resist ance levels with measured inputs of necrosis and defoliation.

PAGE 19

19 CHAPTER 2 LITERATURE REVIEW Background Peanut is a legume grown in warm climates throughout the world. It is cultivated ar ound the world in tropical, sub tropical, and warm temperate climates. The cultivated peanut ( Arachis hypogaea L.) originated in South America and is a self pollinating, indeterminate, annual plant that is distinguished from most other species by producing aerial flowers but fruiting below the soil surface. Peanut leaves, th e photosynthetic unit of the plant, are pinnate with two opposite pairs of leaflets. The primary center of diversity for the species is the Chaco region between southern Bolivia and northwest Argentina (Gregory and Gregory, 1979). Peanut is now cultivated on 24 million ha in more than 100 countries between 40o N and 40o S, generating an annual production of nearly 37 Tg (FAO, 2011). During 2009, China, India, Indonesia, and United States (US) accounted for almost 70% of the total world production, with Chi na leading the way at 40% of production. Average pod yield ranged from 1000 kg ha1 in India to 3800 kg ha1 in US. In the US, peanut wa s produced on about 0.52 million ha during 2010, with a total production of 1.88 Tg The US peanut production is concent rated mainly in three major geographic areas : (i) the southeast, which includes Georgia, Alabama, Florida, and Mississippi; (ii) the southwest, which includes Texas, New Mexico, and Oklahoma; and (iii) VirginiaCarolina, which includes North Carolina, Sout h Carolina, and Virginia. In 2010, Georgia had the highest area and pro duction (228,647 ha and 0. 9 0 Tg ) followed by Texas (66,773 ha and 0.27 Tg ), Alabama (76,890 ha and 0.22 Tg ), and Flori da (58 679 ha and 0.21 Tg ). The s outheastern US accounted for 77% o f total US production. Florida

PAGE 20

20 provides about 11% of US peanut production val ued at around $91 million ( USDA NASS, 2011). A h ypogaea is divided into two subspecies ( hypogaea and fastigiata ) and six botanical varieties (Krapovickas and Gregory, 1994). The subspecies hypogaea which includes botanical varieties hypogaea and hirsuta does not flower on the main stem, has alternate branching pattern, is late maturing, has a high water requirement, and produces larger seeds. The subspecies fastigiata which includ es botanical varieties fastigiata, peruviana, aequatoriana, and vulgaris produces flowers on the main stem, has sequential branching, matures earlier, has a low water requirement, and produces smaller seeds. The four market types in the US peanut trade are: Runner, Virginia, Spanish, and Valencia. Botanical variety hypogaea contains the Virginia and Runner market types, fastigiata contains the Valencia market types, and vulgaris contains the Spanish types. These market types form a rough classification syst em based on pod and seed size characteristics and to a lesser extent on center of genetic origin, growing region, and growth habit (Knauft et al., 1987). Spanish market types typically have small kernels covered with a reddishbrown skin. They are used pri marily in candies and crushed for oil. They are grown mostly in Texas and Oklahoma. Valencia market types have multikernel pod characteristics, red seed coat s and medium seed size. They are grown primarily in the southeastern US and usually used for boil ing and roasting. Runner market types tend to have larger pods and seeds compared to Spanish and Valenc ia types. They are most widely grown in the southeastern US growing region and are used for oil and peanut butter production.

PAGE 21

21 Virginia market types have a large pod and seed size. They are primarily grown in North Carolina South Carolina, and Virginia and used in snack nuts. Peanut Diseases Peanut is susceptible to a variety of biotic stresses. The development and severity of peanut diseases depends on complex interactions among the host, the pathogen, and the environment. The most prevalent pathogens of peanut include tomato spotted wilt virus (TSWV, Topsovirus vectored by thrips), Sclerotium rolfsii Sacc., the causal agent of white mold, Sclerotinia mino r Jagger, the causal agent of Sclerotinia blight, Cylindrocladium parasiticum Crous, Wingfield and Alfenas, the causal agent of Cylindrocladium black rot, Puccinia arachidis Speg., the causal agent of rust, and Cercospora arachidicola S. Hori and Cercospor idium personatum ( Berk and Curt.) Deighton, the cau sal agents of early leaf spot (ELS) and late leaf spot (LLS) Many of these diseases of peanut have a limited geographic range, but the two major foliar diseases, early and late leaf spot are prevalent in almost all peanut producing regions of the world (McDonald et al., 1985; Stalker, 1997) and result in yield reductions that may approach 70% in the absence of proper management practices (Nutter, Jr. and Shokes, 1995; Shokes and Culbreath, 1997). In Florid a, previous research has indicated that LLS is the predominant disease (Jackson 1981; Pixley et al., 1990a, 1990b; Nutter Jr. and Shokes, 1995). Late Leaf Spot Late leaf spot caused by Cercosporidium personatum (Berk. & Curt.) Deighton, (telemorph = Mycosp haerella berkeleyi Jenk.) is the most widespread foliar disease of peanut in the southeastern US (Jackson, 198 1 ). Pod yield losses can be up to 7 0% when fungicides are not applied (Shokes and Culbreath, 1997; Pixley et al., 1990a

PAGE 22

22 McDonald et al., 1985; Jackson and Bell, 1969). The anamorph C personatum is commonly observed on peanut. The lesion is amphigenous with fruiting on both sides of the leaflet, but sporulation is more common on the lower (abaxial) surface. The telemorph, M berkeleyi is rarely obs erved on peanut. Conidiophores of C personatum form in dense clusters, are pale to olivaceous brown, have conspicuous conidial scars, and vary in size from 10100 x 3.0 6.5 Conidia are medium olivaceous, cylindrical, obclavate, usually straight or sl ightly curved and have 19 (mostly 3 4) septa. They vary in size from 20 70 x 4 9 (McDonald et al., 1985). The host range of C personatum is confined to the genus Arachis Symptoms Although the disease is called LLS, the symptoms of this disease devel op on petioles, stipules, stems, and even pegs during the later stages of an epidemic. Lesions are first visible around 10 days after spore deposition as tiny, pinpoint, yellowish flecks. These flecks enlarge to form coalescing, blackishbrown lesions, enl arging to 110 mm in diameter (Shokes and Culbreath, 1997). Mature, sporulating lesions may be apparent by about 15 days after spores are deposited. Sporulation occurs mainly on the lower leaflet surface, although some spores may also be produced on the upper surface of older lesions (Jenkins 1938). Late leaf spot lesions commonly have a less conspicuous or absent chlorotic halo. Disease Cycle Con idia, produced by conidiophores on peanut residues in the soil and off season plants serve as a source of initial inoculum (Figure 21). Although the telemorph of the fungus is known, the ascospores are generally not regarded as important source of

PAGE 23

23 primary inoculum. Conidia are dispersed by wind, splashing water, and insects. Peak dispersal period for conidia occur s at dew dry off in the morning and at the onset of rainfall. Multicelled conidia land on peanut tissue and start germinating. Pathogenesis of C personatum starts with the development of spore germination tubes which enter plant cells via stomata (Leal B e rtioli et al., 2010) or directly through epidermis, allowing intercellular mycelium growth. C ercosporidium personatum does not kill the host cells prior to penetration, but instead develops into haustoria (Perfect and Green, 2001; Mims et al., 1988; Abdou et al., 1974). Lesions generally develop within 1014 days of initial infection (Shokes and Culbreath, 1997). In spite of having a longer incubation period, LLS fungus produces more spores per lesions compared to ELS fungus resulting in more severe damage over a short period of time. Variables like weather, cultivar and the effectiveness of control measures determine number of disease cycles. C ercosporidium personatum favor s warm temperature and humid conditions, which are found in the southeastern peanut growing states during summer months. Maximum LLS infection occurs when temperatures are about 20oC and relative humidity exceeds 93% for more than 12 hr or with continuous leaf wetness periods of 10 hr (Shokes and Culbreath, 1997). Prolonged periods of lea f wetness or several shorter periods of leaf wetness (10 hr or longer) may be equally favorable f or LLS development. Shew et al. ( 1988) reported that infection diminished with increase in temperature from 20oC (where maximum infection occurred) to 28 to 32oC. Leaf wetness is an important limiting factor for infection (Butler et al., 1995). Management Strategies Management of late leaf spot of peanut has been accomplished through the combination of (i) reducing initial inoculum by crop rotation, early planti ng and tillage,

PAGE 24

24 and (ii) reducing the rate of disease spread by cultivar selection and multiple applications of fungicides. Late leaf spot is much worse in warm and wet weather than in cool and dry weather, making it more difficult to control. Cultural Pr actices Most cultural practices used to control LLS are aimed at reducing the initial inoculum. These include crop rotation, early planting tillage, burial of crop residues, and removal of volunteer peanut plants. However, these methods alone will not per mit sustained peanut production (Kucharek, 2005). Crop rotation Crop rotation is one of the most effective means of managing disease in any crop. It results in increased effectiveness of all disease management programs. When peanut is planted on land which has not been planted to peanut for 34 years, the onset and progress rate of LLS are delayed, as contrasted with continuous peanut culture. This alone might provide a 23 week delay in the development of LLS epidemic. It is recommended to avoid planting peanut in the same field more than once out of every three, preferably four years. Considering all peanut diseases, it is recommended to rotate peanut with grass crops such as corn, sorghum, and bahiagrass (Kemerait et al., 2010; Mossler and Aerts, 2007). P eanut yields were reported to be 19 and 41% higher after two years of corn and two years of bahiagrass, respectively (Wright et al., 2006). Planting date Early planting dates (early to midApril) result in comparatively less exposure time for peanut canopi es to hot and humid conditions most conducive for LLS development. This result s in peanut canopies exposed to LLS for less period before harvest (Shokes et al., 1982). Because inoculum concentration increases as the season progresses, LLS

PAGE 25

25 can be managed to some degree by manipulating planting dates. Mossler and Aerts (2007) reported that peanut planted in early to mid April in Florida may not have to be sprayed with fungicides until 60 days after planting (DAP), compared to 2535 DAP for peanut planted dur ing May to early June. However, this advantage is negated by more susceptibility of earlier planted peanut to TSWV (Kemerait et al., 2010). Tillage Conventional tillage involves turning the soil in an entire field, resulting in incorporation of plant r esid ues into the soil. Enhanced residue decay after tillage favors nonpathogenic soil microorganisms over pathogens, resulting in reduction of overwintering inoculum and better disease control (Nutter, Jr. and Shokes, 1997). Recent interests in conservation t illage (e.g. strip tillage) due to increased energy and labor costs prompted studies evaluating effects of conservation tillage on disease epidemics. Although the exact mechanism is not clear, LLS epidemics in striptilled peanut fields were similar or sup pressed compared with conventional tilled plots (Monfort et al., 2004; Cantonwine et al., 2006; Wright et al. 2006; Kemerait et al., 2010). Fungicide Application Multiple applications of fungicides are usually required to keep LLS disease below damaging l evels. Fungicide application currently accounts for onethird of the total variable costs needed to produce peanut (Mossler and Aerts, 2007). Commonly used foliar fung icides include chlorothalonil, t ebuconazole, propiconazole, pyraclostrobin, azoxystrobin, trifloxystrobin, and sulfur. Less commonly used fungicides include copper, maneb, mancozeb, and thiophanate (Mossler and Aerts, 2007). Fungicides are currently applied beginning approximately 3040 DAP and continuing at 1014 day intervals

PAGE 26

26 resulting in seven or more applications during a growing season (Wright et al., 2006; Kemera it et al., 2010). Shokes et al. ( 1982) reported reduced disease severity and defoliation and higher yields with earlier initiation of fungicide application in Florida. Many other studies have shown the use of fungicide application in control of LLS (Smith and Littrell., 1980; Shokes et al., 1983; Bourgeois et al., 1991; Monfort et al., 2004; Woodward et al., 2008; Woodward et al., 2010). Indiscriminate application of fungicides for LLS control may result in undesirable effects e.g. development of fungicidetolerant strains of fungus (Smith and Littrell., 1980) and increased severity of other diseases (Shokes and Culbreath, 1997). Several diseaseforecasting systems have been developed based on relative humidity, temperature or simply number of rain events to reduce spray frequencies (e.g. AU Pnut advisory, Jacobi et al., 1995; Jacobi and Backman, 1995). However, hot and humid weather in Florida result in limited use of these weather based advisories. Moreover, implementation of these systems results in control of disease outbreak rather than minimizing yield losses. Cultivar Selection Cultivars partially resistant to C personatum may also be used to reduce the rate of LLS epidemics. The highest levels of partial resistance are found in unadapted germplasm lines and in wild species derived breeding lines (Wynne et al., 1991), resulting in slow progress in breeding for resistance to LLS. Identified components of rate reducing partial r esistance include extended latent period of the fungus, reduced sporulation, and smaller lesion diameters (Cook, 1981; Chiteka et al., 1988; Aquino et al., 1995; Dwivedi et al., 2002; Cantonwine et al., 2008). Mechanism s of resistance

PAGE 27

27 include restriction o f conidial development and penetration of fungal hyphae through stomata (Leal Bertioli et al., 2010). Breeding and selection of cultivars with partial resistance to LLS have been an important part of integrated disease management programs for reducing yiel d losses in peanut (Tillman et al., 2008; Tillman and Stalker, 2009). Several new releases have shown good resistance associated with delayed disease progress and decreased defoliation. These include Southern Runner (Gorbet et al., 1987), C 99R (Gorbet and Shokes, 2002a), Georgia01R (Branch, 2002), Florida MDR 98 (Gorbet and Shokes, 2002b), Georgia05E (Branch, 2006), Tifrunner (Holbrook and Culbreath, 2007), Hull (Gorbet, 2007), DP 1 (Gorbet and Tillman, 2008), Tifguard (Holbrook et al., 2008), Georgia07 W ( Branch and Brenneman, 2008), Georganic (Holbrook and Culbreath, 2008) and York (Gorbet and Tillman, 2011) The degree of resistance in these cultivars is partial and still allow s for significant damage under severe disease epidemics. Several of the newly released cultivars are associated with unfavorable characteristics such as poor germination (e.g. York, DP 1, C 99R, and Hull) and late maturity (require extra 1421 days). Poor seed emergence results in reduced field stands and hence lower final yields (Morton, 2007). The best management strategy for LLS should integrate several of the above tactics into a program adapted to the cultivar and cultural practices of a given area. Effects of Late Leaf Spot on Peanut Physiology Peanut economic yield is a f unction of cumulative biomass and harvest index, which is determined by partitioning of assimilates to pod and effective duration of pod fill (Phakamas et al., 2008; Duncan et al., 1978). Late leaf spot disease first occurs as necrotic lesions on peanut le aflets and subsequently induces leaflet abscission. This

PAGE 28

28 defoliation commonly lowers LAI values below the optimum value of 3.0 determined by Duncan et al. (1978) which reduces light interception and can cause significant loss of canopy carbon assimilation and yield (Bourgeois and Boote, 1992; Boote et al., 1983a ). In addition t o lowered LAI, reduction in carbon assimilation capacity of the leaves infected with LLS compared to the asymptomatic leaves has also been reported (Boote et al., 1980; Bourgeois and Boote, 1992). Thus, premature loss of green leaf area (by necrotic tissue and defoliation) and reduction of leaf photosynthetic capacity due to disease can contribute to a loss of canopy carbon assimilation, and thus a loss of yield. Many older peanut cult ivars such as Florunner and Georgia Green have poor resistance to LLS. Loss of leaf area due to accelerated senescence was reported to be the predo minant mechanism of yield loss in these cultivars (Bourgeois and Bo ote, 1992; Boote et al., 1980). Bourgeois et al. (1991) reported a reduction of 37 and 46% in pod yield of Florunner in two seasons due to the loss of green photosynthetic leaf area causing significant reduction in production of carbohydrate available for pod growth. However, for cultivars with i mproved resistance to LLS that experience less defoliation (Anderson et al., 1993; Knauft and Gorbet, 1990) yield reduction may also be related to the leaf physiological response to disease instead of to loss of leaf area alone. Late leaf spot disease is characterized by chlorotic flecks that enlarge to necrotic lesions that reduce photosynthetic capacity (Boote et al., 1983a ). Necrotic lesions are photosynthetically useless area that does intercept light However, there may be an additional effect on the photosynthetic capacity of noninfected symptomless area of the leaf.

PAGE 29

29 The concept of a virtual lesion, introduced by Bastiaans (1991), can help in the classification of pathogens according to their effect on photosynthetic efficiency of their hosts. Accord ing to Bastiaans (1991), the virtual lesion is the proportion of leaf tissue, equal to or larger than the visual lesion (proportion of leaf tissue with visible symptoms), in which photosynthesis is severely reduced. In order to relate reductions in leaf photosynthesis to visual lesion area, Bastiaans (1991) proposed a relatively simple model, y = (1 x), where y is the relative net assimilation rate of a diseased leaf compared to that of an asymptomatic leaf, x is the measured visual lesion area, and des cribes the relationship between virtual and visual lesion area. The virtual area represents loss of photosynthetic capacity beyond the visual lesion area. Thus, indicates whether the effect of disease on photosynthesis is higher ( > 1), lower ( < 1), o r equal ( = 1) to that accounted for by the measured visual lesion area. Using this model, several studies have shown that the relationship between photosynthesis and visual disease severity is related to host pathogen interactions (Bassanezi et al., 2002 ; Erickson et al., 2003; Zhang et al., 2009). In general, studies have indicated that values are relatively low (< 2.5) for biotrophic pathogens (Bassanezi et al., 2001; Lopes and Berger, 2001; Robert et al., 2005; Kumudini et al., 2010), intermediate (3 to 6) for hemibiotropic pathogens (Bassanezi et al., 2001; Erickson et al., 2003; Roloff et al., 2004) and highest for necrotrophic pathogens ( Bassanezi et al., 2001; Lopes and Berger, 2001). Values of reported for some biotrophic pathosystems include 1.3 and 2.2 for Uromyces appendiculatus (Pers.:Pers) Unger on common bean ( Phaseolus vulgaris L.) (Lopes and Berger, 2001 and Bassanezi et al., 2001, respectively) and 2.3 for

PAGE 30

30 Phakospora pachyrhizi Syd. & P. Syd. on soybean ( Glycine max L. Merr.) (Kumudini et al., 2010). Thus, values of biotrophs have generally been equal to or very close to one, indicating minimal effects of biotrophic pathogens on photosynthesis beyond the vi sual lesion areas of the leaf. However, values reported for hemibiotrophic pathosystems have generally been greater than 1.0 and often greater than 3.0. For example, Erickson et al. (2003) reported 6.1 for Marssonina brunnea f. sp. brunnea on poplar ( Populus spp.), Bassanezi et al. (2001) reported 3.8 for Phaeoisariopsis griseola (Sacc.) Ferr. on common bean, and Roloff et al. (2004) found values of 2.8 and 3.1 for Septoria albopunctata Cooke on Vaccinium spp. Bourgeois and Boote (1992) found a reduction of 65% in the photosynthesis of peanut leaflets with 15% LLS disease (corresponds to a value of around 4.0). Consequently, greater reductions in photosynthesis beyond the visual lesion area seem to be more common with hemibiotrophs compared to biotrophs. Reasons for reduced photosynthesis beyond the measured visual lesion area, and thus di fferences in are still not clear, but have been related to reductions in chlorophyll (Lopes and Berger, 2001; Moriondo et al., 2005) and carboxylation efficiency. Nogues et al. (2002) concluded that decreased maximum carboxylation velocity of Rubisco ( Vc ,max) was likely the primary determinant underlying the decline in photosynthetic rate of tomato ( Lycopersicon esculentum Mill.) leaves infected by Fusarium oxysporum f.sp. lycopersici Although little difference in values are generally observed among cultivars infected by biotrophic pathogens (Bassanezi et al., 2001; Kumudini et al., 2010),

PAGE 31

31 genotypic differences in have been observed among cultivars in response to hemibiotrophic pathogens (Erickson et al., 2003). Simulations of Late Leaf Spot D amage Cr opd isease interactions have traditionally been quantified as damage functions (Pinnschmidt et al., 1994) consisting of empirical regression equations. However, these equations were very specific to a given condition and could not be used under different e nvironment al conditions Estimating the effects of disease epidemics on crop cultivars grown under different environmental and management conditions requires the use of mechanistic crop growth models (Teng et al., 1998). A generic approach whereby diseaseinduced damage is recorded and used as input to crop model s has been successfully used by number of researchers to simulate effects of disease on growth and yield reduction of crops (Pinnschmidt et al., 1995; Teng et al., 1998; Batchelor et al., 1993; Boot e et al., 1993). This approach can also be used to pinpoint the relative importance of the damage mechanisms and to identify gaps in knowledge of disease effects. The coupling points for damage mechanisms are located at the plant process level (photosynthesis, respiration, etc.) or at the state level (tissue area, weight etc.) (Boote et al., 1983 a ) Models that have been used to predict the impact of foliar diseases like LLS on growth and yield have generally incorporated the disease effects on defoliation and photosynthesis ( Batchelor et al., 1993; Williams and Boote, 1995; Naab et al., 2004; Adomou et al., 2005). The CROPGRO M odel The CROPGRO Peanut model (Boote et al., 1998a, 1998b) is a process oriented mechanistic crop growth model which considers crop carbon balance, crop and soil N balance, and soil water balance at the process level. This model has coupling points

PAGE 32

32 and procedures for entering pest damage to simulate growth and yield reductions associated with foliar pathogens like LLS (Batchelor et al. 1993; Boote et al., 1993 ; Teng et al., 1998). The primary impacts of disease are simulated as defoliation; however the impacts of virtual lesion area are currently simulated by simply defoliating more leaf area (hence zero photosynthesis on that area) rather than creating direct impact at the leaf level photosynthesis (Adomou et al., 2005) This subroutine has been tested by some previous studies (Naab et al., 2004; Adomou et al., 2005) to simulate LLS effects on peanut growth and yield. However, these st udies did not include measured data on leaf necrosis and defoliation required by the disease subroutine in the model. Either visual ratings were linearly regressed against measured necrosis range (minimum of 0% to maximum of 9%) from other studies (Bourgeois et al., 1991) to obtain necrosis values (Adomou et al., 2005) or variable defoliation and necrosis values were used to mimic leaf weight loss (Naab et al., 2004) This signifies the need of ha ving reliable disease evaluation methods to provide accurate assessment of disease effects, which could then be used as input to crop growth models and as evaluation methods in breeding programs. Model E valuation Statistics used for model evaluation vary from simply summing the difference between predicted and measured values to calculating more complicated concordance correlation coefficients (Lin et al., 2002). Two measures that are widely reported in the literature are the root mean squared error (RMSE) and the Willmott agreement index (Willmott 1981, 1982). The R MSE reflects the magnitude of the root mean sum of square differences between the predicted ( P ) and observed ( O ) values over time and is calculated as:

PAGE 33

33 n O P RMSEni i i 1 2) ( ) The D index is a descriptive index that measures dispersion of the simulated an d observed data, calculated as: n i i i n ii iO O O P O P index D1 2 1 2| | | | 1 Where n is the total number of observations, Pi is the predicted value for the i th measurement, Oi is the observed value for the i th measurement, and O is the overall mean of the observed values. A model performs well when the RMSE approaches zero and the D index is close to 1.0.

PAGE 34

34 Figure 21. Disease cycle of late leaf spot, caused by Cercosporidium personatum (Berk. and Curt.) Deighton ( reprinted with permission from Shokes and Culbreath, 1997).

PAGE 35

35 CHAPTER 3 LATE LEAF SPOT EFFEC TS ON GROWTH, PHOTOS YNTHESIS, AND YIELD IN PEANUT CULTIVARS OF DIFFERING RESISTANCE Abstract Cercosporidium personatum (Berk. & Curt.) Deighton causes late leaf spot (LLS) in peanut ( Arachis hypogaea L.) which leads to necrotic lesions, early leaf senescence and yield losses. Detailed physiological analyses can contribute to an improved understanding of peanut disease interactions and cultivar improvement. A study was conducted evaluating two peanut cult ivars with more (York) and less (Carver) quantitative resistance to C personatum grown under fungicidesprayed and nonsprayed conditions in the field at Citra, Florida over two years. Data were collected on disease severity using the Florida 1 to 10 visual rating scale and by direct measurement of percent canopy lesion area. Leaf lifespan, total canopy photosynthesis (TCP), plant growth, and pod yield were also measured. Disease severity based on canopy lesion area was reduced by 30% in York compared to C arver. No additive effects of combining the resistant cultivar with fungicide were seen, as fungicide use increased yield by 364 kg ha1 for both cultivars. Yield was more strongly related to disease severity based on canopy lesion area than to the Florida scale. Yield improvement with York was not as closely related to disease severity with only a 6% gain in pod yield in York compared to Carver. In addition, reduction in TCP was greater in York compared to Carver given their respective disease severity. These results indicated that combining resistance with the maintenance of physiological function during LLS infection could result in improved peanut yields under diseased conditions.

PAGE 36

36 Background Peanut ( Arachis hypogaea L.) is one of the major sources of protein and oil in the world. It is c ultivated on 24 million ha in more than 100 countries, generating an annual production of nearly 3 7 Tg (FAO, 2011 ). Nevertheless, worldwide peanut production is severely hampered by the incidence of numerous diseases. Earl y leaf spot (caused by Cercospora arachidicola S. Hori), and late leaf spot [caused by Cercosporidium personatum (Berk. & Curt.) Deighton] are among the most widespread and damaging foliar diseases of peanut in the southeastern United States (Nutter Jr. and Shokes, 1995). Pod yield losses can be greater than 50% when fungicides are not applied (Shokes and Culbreath, 1997). In Florida, late leaf spot (LLS) is the pr edominant disease (Jackson, 1981 ), causing yield losses of up to 50% (Pixley et al., 1990a). C onsequently, regular and costly fungicide applications are currently used to minimize yield losses from peanut diseases (Woodward et al., 2008; Monfort et al., 2004). Improved cultivars with moderate resistance to late leaf spot, along with other integrate d disease management practices, have also been successfully used to reduce inputs and production costs (Woodward et al., 2010; Woodward et al., 2008; Monfort et al., 2004). However, the effects of LLS on the physiological responses in cultivars of differing leaf spot resistance is not well understood and could contribute to improved cultivar development for disease resistance. C ercosporidium personatum is a hemibiotrophic fungal pathogen that infects peanut leaves and stems (Mims et al., 1988). The initial source of inoculum is primarily conidia from crop residues in the soil. Conidia are rainsplashed or windblown onto leaf surfaces where they initiate infection. Symptoms are first recognizable as small necrotic flecks that enlarge to dark brown lesions f rom 1 to 10 mm in size (Smith et al., 1992).

PAGE 37

37 Lesions generally develop within 1014 days of initial infection. Symptoms are influenced by host genotype and environmental conditions, such as high temperature, rainfall, and relative humidity (Shew et al., 19 88). Peanut economic yield is a function of cumulative biomass and harvest index, which is determined by partitioning of assimilates to pod and effective duration of pod fill (Phakamas et al., 2008; Williams and Boote, 1995; Duncan et al., 1978). Premature loss of green leaf area (by necrotic tissue and defoliation) and reduction of leaf photosynthetic capacity due to disease contribute to a loss of canopy carbon assimilation, and thus a loss of yield. Many older peanut cultivars such as Florunner and Georg ia Green have poor resistance to LLS. Loss of leaf area due to accelerated senescence was reported to be the predo minant mechanism of yield loss in these cultivars (Bourgeois and Boote, 1992; Boote et al., 1980). However, for cultivars with improved resist ance to LLS that experience less defoliation, yield reduction may also be related to leaf physiological response to disease instead of to loss of leaf area alone. The breeding and selection of cultivars with partial resistance to LLS has been an important part of integrated disease management programs for reducing yield losses in peanut. Several new releases have shown good resistance associated with delayed disease progress and decreased defoliation. Components of resistance identified include extended lat ent period of the fungus, reduced sporulation and smaller lesion diameters (Chiteka et al., 1988; Dwivedi et al., 2002; Cantonwine et al., 2008). Selection of these resistant cultivars is typically based on visual disease ratings (e.g., Florida 1 to 10 sca le) that combine both visual lesion disease severity and defoliation (Gorbet and Tillman, 2008). Direct measures of canopy lesion severity using image analysis may

PAGE 38

38 improve estimates of disease severity, especially in resistant cultivars that exhibit decreased defoliation. While these measures of disease severity work well for monitoring disease dynamics, they do not always correlate well with yield reductions (Bergamin Filho et al., 1997; Jesus Jr. et al., 2001), due to a disconnect between the ratings and actual functional impairment (Bastiaans, 1991). In addition, host functional response to pathogens can be variable depending on environment, genotype, and physiological status (Zhang et al., 2009; Erickson et al., 2003). Better understanding of the physiol ogical responses to LLS related to yield in cultivars differing in resistance is needed to contribute to improved cultivar selection and modeling growth and yield responses of peanut to leaf spot. The objective of this study was to characterize LLS severit y and its effects on growth and partitioning, leaf lifespan, canopy photosynthesis, and pod yield of York, a relatively resistant cultivar, compared to Carver, a cultivar with relatively poor resistance to LLS in a field environment. Materials and Methods Experimental Site and Design Field experiments were conducted during the 2008 and 2009 growing seasons at the Plant Science Research and Educat ion Unit in Citra, Florida (29o2360 N, 82o12 W) on a Gainesville loamy sand (Hyperthermic, coated Typic Quartzipsamments) soil. The experiment was a multi factorial design with the main factors being cultivar, fungicide application and year. Cultivar and fungicide application were arranged in a randomized complete block (RCB) with four replications of each tre atment. Two cultivars were selected for di fferences in resistance to LLS: Carver (Gorbet, 2006) has poor resistance to LLS; while York (Gorbet and Tillm an, 2011) has moderate resistance to LLS (Tillman et al., 2008). Fungicide application included: (i) no fungicide application;

PAGE 39

39 and (ii) an industry standard fungicide schedule (Table 31) applied on a 14day interval commencing from approximately 40 DAP. Fungicides were applied using a CO2 backpack sprayer calibrated to deliver 328 and 374 L ha1 during 2008 and 2009, respectively. A handheld boom with five flat fan nozzles, spaced 45.7 cm apart was used to spray two rows at a time (spray coverage of 182 cm wide). Plots were previously sown with bahiagrass ( Paspalum Notatum Fluegg) and rye ( Secale cereale L .) in a four year rotation with rye (nurse crop to establish bahiagrass) followed by two years of bahiagrass and then peanut. Sowing occurred during the latter part of the recommended planting window for North Central Florida on May 20 in 2008 and May 27 in 2009 to maximize LLS pressure (Wright et al., 2006). Each plot consisted of 6 rows spaced 0.91 m apart and 4.6 m long. Each block was separated by 3.7 m fallow alleys and the entire study was surrounded by two border rows. Seeds were sown at a rate of 1 7 20 seeds per meter row using a conventional planter. Infurrow application of azoxystrobin was conducted at a rate of 0.16 kg a.i. ha1 whi le planting to control seedling diseases. Irrigation was applied as needed with a linear move system. Standard management practices for irrigated peanuts were employed during both years (Wright et al. 2006), including a 39 18 blended granular fertilizer that was broadcast before planting at a rate of 560 kg fertilizer ha1 during both growing seasons. To satisfy the c alcium requirement for pod and kernel formation, gypsum was broadcast at a rate of 2240 kg ha1 split equally in two applications around 3540 DAP followed by another application 1014 days later. Disodium octaborate tetrahydrate was applied with the first two fungicide sprays at a rate of 5.6 kg ha1 per application to supply boron.

PAGE 40

40 Measures of Disease Severity and Growth Late leaf spot intensity was assessed with the widely used Florida 1 to 10 scale ( Table B 1 Woodward et al., 2010; Gorbet and Tillman, 2008; Cantonwine et al., 2008; Chiteka et al., 1988). Values of 1 to 4 indicate increasing leaf spot incidence on leaflets within the lower or upper canopy, but no defoliation. Ratings from 4 to 10 are associated with increasing levels of defoliation (Chit eka et al., 1988). Ratings began when visual symptoms first appeared (87 and 77 DAP in 2008 and 2009, respectively) and continued every 710 days until harvest. Area under disease progress curve (AUDPC) values were calculated for each plot from these disease ratings (Shanner and Finney, 1977) and were standardized by dividing AUDPC values by the number of days from the first observed symptoms to harvest to account for differences in the duration of LLS epidemics (Woodward et al., 2010; Woodward et al., 2008). Microscopic examination of lesions on leaflets indicated that C personatum was the dominant pathogen in both years (Figure 3 1) Tomato spotted wilt (caused by Tom ato spotted wilt virus) and whit e mold (caused by Sclerotium rolfsii Sacc.) was not obser ved in the field plots during both growing seasons. Canopy defoliation and disease severity, the components that make up the Florida scale ratings, were also measured objectively throughout the growing season to compare to the more subjective Florida 1 to 10 scale assessment. Approximately biweekly, a randomly selected 61cm segment of the outer two rows of each plot was harvested, minimizing disturbance or border effects on the inner two final harvest yield rows. A representative subsample excluding the lar gest and the smallest plants was selected from each harvested sample (Bourgeois et al., 1991; Pixley et al., 1990a). Forty leaflets were randomly selected throughout the canopy from this subsample plant.

PAGE 41

41 All leaflets were scanned at 300 dpi using a flatbed scanner (Microtek ScanMaker 5800, Microtek Int. Inc., Industrial Park Hsinchu, Taiwan) and stored as .tiff files. Leaf images were processed using ASSESS ver 2.0 image analysis software (American Phytopathological Society, St. Paul, MN) to give the percent canopy lesion area ( Figure B 2, Erickson et al., 2003). AUDPC values for serial measurements of canopy lesion area were calculated and standardized for each plot similarly to the AUDPC values from the Florida 1 to 10 scale disease progression assessment. The remaining harvested sample was immedi ately oven dried for 72 h at 60oC and subsequently weighed. Leaflets and pods were separated from all subsamples. Pods were counted and then leaves, stems and pods were oven dried to a constant mass. Stem, leaf and pod dry weights (DW) were determined for the entire sample by multiplying their respective fractions of the subsample times the total weight of the harvested sample. In the central two rows of each 6row plot, five plants were chosen at random and the fi rst fully expanded leaf on each main stem was tagged using colored plastic tags at 49 and 92 DAP in 2008, and at 50, 65, and 79 DAP in 2009. These leaves were examined at weekly intervals until defoliation to calculate the total leaf lifespan in days for a ll the leaflets. Measures of Canopy Photosynthesis and Yield Starting approximately 35 DAP, a 61cm section of row was selected randomly from the outer two rows to measure canopy photosynthesis. Measurements were taken at 10 15 d intervals, using a 91 cm x 61 cm aluminum frame mylar chamber and a portable photosynthesis system (LICOR LI 6200, Li Cor Inc., Lincoln, NE) as explained by Bourgeois and Boote (1992). Carbon exchange rate was measured on two plots from each treatment under full sunlight and total darkness (achieved by covering the large

PAGE 42

42 chamber with a black plastic sheet) conditions between 10:00 and 14:00 h. Measured carbon exchange rates under dark conditions were considered to represent canopy, root, and soil respiration. Total canopy photosynthesis (TCP, Boote et al., 1983 b ) was calculated by adding the absolute dark respiration to the observed carbon exchange rate. The central two rows of each 6row plot in each genotype were dug at maturity (determined by hull scrape method; Williams and Drexl er, 1981) using a conventional two row digger shaker inverter. Plants were allowed to sundry in the field for 34 days. Afterwards, stationary threshers were used to harvest pods. Peanut yields were determined after drying to uniform moisture content of 9% (wt/wt). Sprayed plots of Carver were inverted 135 and 127 DAP in 2008 and 2009, respectively. Nonsprayed plots were harvested approx. 7 days earlier in each year due to leaf spot pressure. Both sprayed and nonsprayed York plots were inverted 149 and 1 45 DAP, respectively. In 2009, a subsample of 200 g of pods per plot was subjected to a standard analysis for peanut grade. Pod samples were graded using standard farmer stock grading equipment in accordance with the federal state inspection service method. Pod grades were defined as percent total sound matur e kernels (TSMK) which is the sum of sound mature kernels and sound split kernels. Statistical Analysis Statistical analyses were performed using analysis of variance procedures in the GLIMMIX procedure of SAS (SAS Institute, 2009). Cultivar, fungicide regime, year, and their interactions were considered fixed effects and block by year as a random effect. Degrees of freedom were determined using the KenwardRoger method. Where significant ( P < 0.05) fixed effects were seen, pairwise comparisons were made using

PAGE 43

43 the LSMEANS statement with TUKEY method. Relations between yield and disease severity were analyzed using linear regression procedures. Statistical analyses of total biomass and its partitioning, an d TCP were performed using nonlinear regression procedures of the nlme library of R (R Development Core Team, 2008). A 3parameter logistic function (Eq. [ 3 ] in Yin et al., 2003) was employed to fit stem, pod, and total biomass data, which provided a y asy mptote value, shape parameter related to growth rate, and DAP value at inflection point, which represent s the DAP at half of the maximum value on the y axis. Leaf weight and TCP were fit with a 3parameter gaussian function (Gauch Jr. and Chase, 1974), whi ch provided the maximum value on the y axis, DAP at which the maximum value was achieved, and a peak width parameter at of the maximum value. Analysis of variance was run on these parameters using GLIMMIX of SAS, as explained earlier (except for TCP as data was collected for only two replicates ). Results of this analysis are reported only when significant. Results Growth Environment Environmental conditions during the 2008 and 2009 growing seasons were quite favorable for LLS development (Figure 3 2 ). Rai nfall from mid May through harvest in mid October was 481 and 745 mm in 2008 and 2009, respectively. This precipitation was received in 58 events in 2008 and 74 events in 2009. Irrigation was not applied in either year after onset of disease as rainfall was adequate for crop growth. Average daily temperature during the same period was 25.8 and 25.5 C in 2008 and 2009 respectively. Relative Humidity ranged from 62 to 96% in 2008 and 65 to 96% in 2009.

PAGE 44

44 Disease Assessment Late leaf spot epidemics occurred in both years of the study, but appeared earlier in 2009 compared to 2008 (Figure 33 ) consistent with more frequent and abundant rainfall in 2009 compared to 2008. Late leaf spot symptoms were first observed visually in the field around 95 and 80 DAP during 2 008 and 2009 on both cultivars, respectively. Carver, the less resistant cultivar, showed more rapid disease progress than York during both years, especially in nonsprayed plots. Fungicide delayed the initial progress of disease symptoms in both cultivars (Figure 3 3 ). Standardized values for the area under disease progress curves for both Florida 1 to 10 scale ratings (stAUDPCFL) and percent canopy necrotic lesion area (stAUDPCLes) were generally in good agreement and showed significantly reduced disease intensity associated with fungicide inputs and with the moderately resistant cultivar York compared to the poorly resistant cultivar Carver (Table 32). For example, stAUDPCLes and stAUDPCFL were 30% and 19% lower in York compared to Carver, respectively. Similarly, fungicide sprayed plots showed a 43% reduction in stAUDPCLes and a 26% reduction in stAUDPCFL compared to nonsprayed plots. A significant year x cultivar x fungicide effect on stAUDPCLes resulted from higher values in York in 2008 compared to 2 009, whereas higher values in Carver were seen in 2009 compared to 2008 (Table 3 2). This pattern was not seen in stAUDPCFL, as 2009 values were significantly higher in both cultivars, resulting in a significant year effect. Plant Growth and Development A lthough the cultivars did not differ ( P > 0.05) in their maximum stem or leaf DW (or leaf area index, data not shown), Carver achieved maximum leaf DW 10 days earlier ( P = 0.03, Figure 34 ) and the DAP value at inflection was 10 days earlier ( P < 0.001)

PAGE 45

45 fo r stem DW. Maximum leaf DW was attained at 79 and 89 DAP in Carver and York, respectively, across both growing seasons. In both years, following attainment of maximum leaf DW, defoliation was observed in all treatments, but defoliation in nonsprayed plots generally exceeded that of fungicidesprayed plots, as indicated by narrower peak widths for leaf DW ( P < 0.01). This effect was greater in Carver compared to York as leaf lifespan data of tagged leaf cohorts showed greater differences in leaf lifespan in sprayed plots compared to nons prayed plots for Carver (Table 33). In addition, defoliation occurred more quickly and to a greater extent in Carver compared to York, as indicated by narrower peak widths ( P = 0.03) in Carver (Figure 3 4 ). Notably, partiti oning to leaf and stem weight largely occurred before appreciable disease was found, whereas much of the partitioning to pod weight occurred after disease (Figure 3 4 ). For example, DAP at inflection for stem weight in Carver was at 53, while DAP for pod w eight was 82. In addition, DAP value at inflection for pod weight occurred sooner ( P < 0.001) in Carver (82 DAP) compared to York (101 DAP). Canopy photosynthesis was in agreement with seasonal patterns of leaf and stem accumulation, as maximum TCP occurred at 70 DAP in Carver and 80 DAP in York, but maximum TCP was sim ilar between cultivars (Figure 35 ). In addition, similar peak width values indicated similar declines in TCP between cultivars, despite disease progress that was comparatively slower in York than Carver (Figure 3 3 ). However, fungicide application resulted in a slower decline in TCP, as indicated by a peak width of 34 days in fungicidesprayed plots compared to 28 days in their no n sprayed counterparts (Figure 35 ).

PAGE 46

46 Pod Yield and Quality Mean pod yields across all treatments ranged from 2500 to 3500 kg ha1 (Table 32). Fungicide application resulted in a significant increase in pod yield (12.5%) over nonsprayed plots. However, there was a significant year x fungicide interaction whereby diff erences were significant in 2009, but not in 2008. Averaged across all treatments, pod yields were not different among growing seasons ( P = 0.71). This was due to a significant cultivar x year interaction, whereby the poorly resistant cultivar (Carver) out yielded the moderately resistant cultivar (York) in 2008, whereas the opposite was true in 2009. Notably, there was no cultivar x fungicide interaction seen in either year of the study, indicating no diminished response of fungicide on absolute yield gain of York Averaged across all treatments, number of pods per unit area was greater ( P = 0.03) while average pod size was smaller ( P < 0.01) in 2009 compared to 2008. Pod yield was negatively related to stAUDPCFL and stAUDPCLes and the slopes of these relati onships were not affected by cultivar or fungicide schedule (Figure 3 6 ). Overall, the relationship between pod yield and stAUDPCLes was better than that between pod yield and stAUDPCFL, which was especially evident at relatively low disease severities. Fi nally, neither cultivar nor fungicide affected peanut TSMK during 2009 (Table 32). Discussion The overall objective of this study was to gain an improved understanding of peanut response to disease by looking at effects of LLS on peanut physiology, growth and yield of two cultivars differing in resistance, which will be important for continued cultivar improvement and lower fungicide input in peanut production. T he more resistant cultivar contributed to delayed disease progress, which resulted in slower development of canopy lesion area and less defoliation. Improved yield in the more resistant cultivar

PAGE 47

47 was seen in one year of the study when the LLS disease severity was high. N o additive effects of combining improved cultivar resistance and application of f ungicide on pod yield were found, as the absolute gains in yield associated with fungicide treatment were the same between both cultivars across both years of the study. Pod yield was better related to stAUDPCLes compared to stAUDPCFL. Finally, TCP was fou nd to decline similarly in both cultivars despite the slower progress of disease noted in the more resistant cultivar. Delayed disease progress in more resistant cultivars like that seen in the present study has been demonstrated in other studies using the Florida 110 scale ratings (Woodward et al., 2010; Monfort et al., 2004) and canopy disease severity (Pixley et al., 1990b). Visual disease presence in the improved cultivar appeared to start at the same time as in the less resistant cultivar during both years; however, the progress of the disease was slower in the improved cultivar. This differing pattern of disease progress could be explained by a number of factors including a reduced number of initial infection points (foci) and/or differences in the latent period of the fungus. Prior studies have found little difference in the incubation period among a wide range of peanut genotypes, while the latent period tended to be longer in more resistant genotypes, resulting in slower temporal progression of the disease (Cantonwine et al., 2008; Dwivedi et al., 2002; Chiteka et al., 1988). Although the 14day calendar based fungicide program did not achieve 100% disease control, fungicide application delayed the progress of disease symptoms (Pixley et al., 1990b; Bourgeois et al., 1991). Substantial necrosis and defoliation due to LLS was observed in the control plots (Figures 33 3 4 and 3 5 ; Table 33) during both

PAGE 48

48 growing seasons which is typical for nonsprayed peanut. This was also observed by other studies conducted under different growing seasons and locations (Woodward et al., 2010; Mon fort et al., 2004). This study also showed that yield benefits associated with applying fungicide did not differ significantly between cultivars varying in their resistanc e to LLS. Therefore, based on the results, growers might be reluctant to reduce fungicide applications even on more resistant cultivars. However, other studies have shown nonsignificant yield losses in more resistant cultivars with reduced fungicide applicati on compared to a 14day calendar based schedule (Woodward et al., 2010; Monfort et al., 2004). This discrepancy might be due to differences in peanut cultivars, LLS severity, environment and/or fungicide schedule. Resistance to LLS in southeastern U.S. run ner type peanut cultivars has generally been associated with later maturing varieties that possess a later onset of pod fill and a reduced pod growth rate, but possess longer effective pod fill duration (Pixley et al., 1990a). In the present study, York showed later initiation of pod fill, slower pod growth rate, and longer duration of pod fill compared to Carver (Figure 3 4 ). Implications of these growth patterns for LLS effects on yield depended on onse t of the disease epidemic in this study. In 2008, when LLS was relatively late in arrival, partitioning to pod yield was nearly complete in Carver, and thus relatively high yields were attained with Carver with little effect of fungicide on yield. In contrast, in 2009 when LLS arrived about 2 weeks earlier c ompared to 2008, LLS effects on pod yield were greater and effect of fungicide was greater. Thus, where later planting dates are desired (e.g., to minimize incidence of tomato spotted wilt virus), cultivars with improved LLS resistance are beneficial. Finally, since LLS had no effect on TSMK or average pod size in this

PAGE 49

49 study (Table 3 2), the determinant of yield impacted by LLS was pod number, which is consistent with the Phakamas et al. (2008) study that showed that peanut yield was primarily determined by pod number and not pod size across genotypes. Relations between yield and disease severity measurements are often weak (Jesus Jr. et al., 2001) ; however significant re gression relationships were found in this study (Figure 3 6 ). This finding may be due to the wide ranges of disease s everity and yield observed in this study. In addition, yield was more strongly related to stAUDPCLes compared to stAUDPCFL. This suggests that pod yield response to disease epidemics is better explained by measured canopy lesion area rather than the visually determined Florida 1 to 10 scale, which is likely due to the fact that stAUDPCLes was determined using an objective digital image analysis instead of subjective visual ratings. While pod yield reductions were generally relat ed to disease ratings (Figure 36 ), yield reductions in York due to LLS were greater than the ratings indicated. Reduction in disease severity under nonsprayed conditions in York compared to Carver were 22 and 34% based on stAUDPCFL and stAUDPCLes, respec tively. Moreover, the leaf lifespan in nonsprayed York was longer than Carver (Table 3 3). However, this relatively lower disease severity resulted in only 8% yield improvement in York compared to Carver. Thus, the yield improvement in York was not propor tional to the reduction in disease severity in this study. One potential explanation for this disconnect between disease reduction and yield improvement is the existence of at least two separate mechanisms: (i) the ability to sustain leaf photosynthesis du ring disease progression; and (ii) resistance to the progression of disease symptoms. I n this study, the more resistant cultivar, York, may lack the ability to sustain photosynthesis at a given disease severity.

PAGE 50

50 This idea is supported by similar reductions in TCP in both York and Carver despite reduced disease severity in York (Figure 3 3 ). Thus, a combination of LLS resistance (i.e., delayed disease progress) combined with host physiological tolerance (i.e., maintenance of physiological function in the presence of disease) may offer the most promising approach for peanut cultivar improvement and reduced fungicide input production systems. In conclusion, this study demonstrates that cultivar resistance is an important component for integrated disease managem ent of LLS in peanut, particularly during years with high disease pressure. Nevertheless no diminished effect of fungicide with improved cultivar on absolute yield gain was observed. So, foliar application of fungicide still seems to play an important rol e in minimizing damage caused by LLS epidemics. Despite substantial reduction in disease severity and defoliation in the resistant cultivar York, yield improvement over the less resistant cultivar, Carver, was marginal and most beneficial under heavy LLS pressure. T hese findings were attributed in part to a lack of improved physiological tolerance to LLS in York. These results indicate that combining resistance to disease progression with enhanced ability to sustain canopy photosynthetic capacity in the cul tivar selection procedure could provide significant improvement in our efforts to improve peanut yields under diseased conditions.

PAGE 51

51 Table 3 1. Fungicide spray schedule for the field experiments at Citra, FL. Spray Fungicide 1 Chorothalonil (1.26) 2 Cho rothalonil (1.26) 3 Pyraclostrobin (0.18) 4 Azoxystrobin (0.33) 5 Chorothalonil (0.63) + Tebuconazole (0.23) 6 Chorothalonil (0.63) + Tebuconazole (0.23) 7 Chorothalonil (1.26) 8 Chorothalonil (1.26) Numbers in the parentheses denote the rate of f ungicide application (kg a.i. ha-1)

PAGE 52

52 Table 3 2. Treatment means ( n = 4) and analysis of variance results for standardized area under the disease progress curve for Florida 1 to 10 scale (stAUDPCFL) and percent canopy lesion area (stAUDPCLes), pod yield, pod number, average pod weight, and total sound mature kernels (TSMK). Fungicide treatments were no fungicide application (NF) and standard 14day calendar based application (F). Cultivars were Carver (C) and York (Y). Year Cult ivar Fung icide stAUDPC FL stAU DPC Les Pod Yield Pod No. Pod Weight TSMK 2008 C NF 4.13 3.36 Kg ha 1 3098 m 2 406 g 0.95 % F 3.11 2.23 3290 395 0.95 Y NF 3.45 2.97 2925 284 1.03 F 2.71 2.06 3122 354 1.08 2009 C NF 5.17 4.93 2498 509 0.90 72.7 F 3.51 2.25 3144 533 0.92 75.2 Y NF 3.78 2.53 3136 511 0.87 74.9 F 2.91 1.39 3556 550 0.90 74.7 SIGNIFICANCE Cultivar Fungicide Cult x Fung Year Cult x Year Fung x Year Cult x Fung x Year *** *** ns ns ns *** *** *** ns ns ns *** * ns ns ns ns ns ns * ** *** *** * ** ns ns ns ns ns ns ***, **, and ns = P < 0.001, P < 0.01, P < 0.05 and P > 0.05 r espectively; Data not recorded

PAGE 53

53 Table 33. Treatment means ( n = 4) for leaf lifespan of leaf cohorts tagg ed at different times (DAP) throughout the growing season. Fungicide treatments were no fungicide application (NF) and standard 14day calendar based application (F). Cultivars were Carver (C) and York (Y). Tagging date (DAP) 2008 2009 Cult ivar Fung icide 49 92 50 65 79 C NF d 66ab d 29c d 53b d d 41c 31c F 69a 38b 62a 50a 42a Y NF 63bc 44a 52b 43bc 35b F 60c 47a 47c 44b 42a Numbers followed by the same letter within a column do not differ ( P > 0.05)

PAGE 54

54 Figure 31. Peanut leaf with leaf spot disease. (A) abaxial side covered with fungal hyphae, (B) adaxial side with necrotic lesion but no conidiophores, (C) abaxial side with conidiophores of Cercospora arachidicola, and (D) abaxial side with conidiophor es of Cercosporidium personatum A C D B

PAGE 55

55 Figure 3 2 Average daily temperature (A), relative humidity (B), and cumulative rainfall (C) for the field experiment during the study period (Source: Florida Automated Weather Network, Citra, FL).

PAGE 56

56 Figure 33 Progress of late leaf spot as estimated with the Florida 1 to10 scale and percent canopy lesion area during 2008 and 2009 growing seasons for the two peanut cultivars (C Carver; Y York) grown under fungicide sprayed (F) and nonsprayed (NF) conditions Vertical bars greater than symbols represent standard error of the mean ( n =4).

PAGE 57

57 Figure 34 Leaf, stem, pod, and total dry matter accumulation vs. days after planting (DAP) for two peanut cultivars Carver (C) and York (Y) grown under fungicide spr ayed (F) and nonsprayed (NF) conditions at C itra, FL during 2008 and 2009. Symbols represent treatment means ( n =4) while regression lines represents gaussian (for leaf biomass) and logistic (for stem, pod and total biomass) model fits.

PAGE 58

58 Figure 3 5 M id day total canopy photosynthesis (TCP) for two peanut cultivars (C Carver; Y York) grown under fungicide sprayed (F) and nonsprayed (NF) conditions at Citra, FL during 2008 and 2009. Symbols represent treatment means ( n =2) while regression lines represents gaussian model fits.

PAGE 59

59 Figure 36 The relationship between pod yield and the standardized area under the disease progress curve based on the Florida 1 to 10 scale (stAUDPCFL) and percent canopy lesion area (stAUDPCLes) for the two peanut culti vars, Carver (C) and York (Y).

PAGE 60

60 CHAPTER 4 PHOTOSYNTHETIC CONSE QUENCES OF LATE LEAF SPOT DIFFER BETWEEN TWO PEANUT CULTIVARS WITH VARIABLE LEVELS OF RESISTANCE Abstract Late leaf spot (LLS) caused by Cercosporidium personatum (Berk. & Curt.) Deighton red uces leaf CO2 assimilation rate ( Asat) and accelerates leaf defoliation, which together lead to major reductions in peanut ( Arachis hypogaea L.) yield worldwide. This study was conducted to determine whether differences in photosynthetic response to LLS severity exist among peanut cultivars of differing resistance. Field experiments were conducted in 2008 and 2009 to study the effects of LLS on Asat of tagged leaf cohorts, and photosynthetic response of similar age leaves to LLS in peanut cultivars with mor e (York) and less (Carver) quantitative resistance. A nonlinear model, y = (1 x) was used to analyze Asat data, where y is relative Asat, x is measured visual lesion area, and represents the relationship between virtual and visual lesion area. Progression of LLS severity on leaf cohorts was slower in York compared to Carver. However, the reduction in Asat with leaf cohort age was similar across the cultivars. This paradox could be explained by a higher value in York (4.6) compared to Carver (3.6), indicating a greater relative reduction in Asat beyond the necrotic lesion area in Y ork. This greater reduction in Asat in York compared to Carver was most closely related to a reduction in maximum carboxylation velocity. Results indicated that future efforts to improve LLS resistance should include sustaining Asat in response to LLS infection along with slower disease progress. Background Late leaf spot (LLS), caused by Cercosporidium personatum (Berk. & Curt.) Deighton, (telemorph = Mycosphaerella berkeleyi Jenk.) is among the most widespread

PAGE 61

61 and damaging foliar diseases (Nutter Jr. and Shokes, 1995) of peanut ( Arachis hypogaea L.) in the southeastern United States. Pod yield losses of up to 50% have been reported when fungicides are not applied (Shokes and Culbreath, 1997). Consequently, regular and costly fungicide applications are curr ently used to minimize yield losses from LLS (Monfort et al., 2004; Woodward et al., 2008). In addition, breeding and selection for improved cultivars with moderate resistance to leaf spot have recently been used with integrated disease management practices to reduce inputs and production costs (Monfort et al., 2004; Woodward et al., 2008). However, LLS effects on peanut yield are often variable and do not always correlate with cultivar LLS resist ance ratings (Chapter 3). An improved understanding of the ef fects of LLS on leaf level physiological responses could help to explain these variable yield responses and result in better identification of improved cultivars in breeding programs and better modeling estimate s of peanut yield loss to LLS. Cercosporidium personatum is a hemibiotrophic fungal pathogen that infects leaves, the photosynthetic unit of the peanut plant (Mims et al., 1988). Conidia, produced by conidiophores, on peanut residues in soil and off season plants typically serve as a source of initial inoculum. Conidia are dispersed by wind, splashing water, and insects. Pathogenesis of C personatum starts with the development of spore germination tubes which enter plant cells via stomata or directly through the epidermis, allowing intercellular mycelium growth. Lesions generally develop within 1014 days of initial infection (Shokes and Culbreath, 1997) and cause reductions in leaf carbon assimilation, premature leaf senescence, and pod detachment that result in yield losses (Bourgeois et al., 1991; Bourgeois and Boote, 1992).

PAGE 62

62 Several new releases from breeding programs (Branch, 2002; Holbrook and Culbreath, 2008; Tillman et al., 2008) have shown moderate resistance levels associated with slower disease progress and less diseaseinduced defoliation. C omponents of identified resistance include a lengthened latent period of the fungus, reduced sporulation, and smaller lesion diameters (Chiteka et al., 1988). Selection of resistant cultivars is typically based on visual disease ratings (e.g., Florida 1 to 10 scale) that combine both visual lesion disease severity and defoliation (Chiteka et al., 1988). While these ratings work well for monitoring disease dynamics, they do not always correlate well with yield reductions, especially at low disease severities (e.g., Bergamin Filho et al., 1997; Jesus Jr. et al., 2001). For example, similar yield losses between cultivars differing significantly in disease progress have b een reported (Chapter 3). It has often been found that the lack of a correlation between dis ease severity and yield loss is due to a disconnect between disease ratings and the actual functional impairment, which has been related to a loss of photosynthetic activity beyond the visual lesion area (Bastiaans, 1991). Reductions in the photosynthetic capacity of infected leaves have been shown in several pathosystems (Boote et al., 1983a ; Bourgeois and Boote, 1992; Shtienberg, 1992; Kumudini et al., 2008). In order to relate reductions in leaf photosynthesis to visual lesion area, Bastiaans (1991) proposed a relatively simple model, y = (1 x), where y is the relative net assimilation rate of a diseased leaf compared to that of an asymptomatic leaf, x is the measured visual lesion area, and describes the relationship between virtual and visual lesion area. The virtual area represents loss of photosynthetic capacity beyond the visual lesion area. Thus, indicates whether the

PAGE 63

63 effect of disease on photosynthesis is higher ( > 1), lower ( < 1), or equal ( = 1) to that accounted for by the measured visual lesion area. Using this model, several studies have shown that the relationship between photosynthesis and visual disease severity is related to host pathogen interactions (Bassanezi et al., 2002; Erickson et al., 2003; Zhang et al., 2009). In general, studies have indicated that values are relatively low (< 2.5) for biotrophic pathogens (Bassanezi et al., 2001; Lopes and Berger, 2001; Robert et al., 2005; Kumudini et al., 2010), intermediate (3 to 6) for hemibiotropic pathogens (Bassanezi et al., 2001; Erickson et al., 2003; Roloff et al., 2004) and highest for necrotrophic pathogens (Bassanezi et al., 2001; Lopes and Berger, 2001). Reasons for reduced photosynthesis beyond the measured visual lesion area, and thus differences in are still not clear, but have been related to reductions in carboxylation efficiency (Nogues et al., 2002) and chlorophyll (Lopes and Berger, 2001; Moriondo et al., 2005). However, understanding why values differ, and perhaps even more importantly, whether they differ by cultivar within species is critical for improved cultivar selection and for modeling effects of disease on carbon assimilation, growth and yield (Bastiaans, 1993; Adomou et al., 2005; Bancal et al., 2007). Although little difference in values are generally observed among cultivars infected by biotrophic pathogens (Bassanezi et al., 2001; Kumudini et al., 2010), genotypic differences in have been observed among cultivars in response to hemibiotrophic pathogens (Erickson et al., 2003). Despite the importance of LLS in peanut production, there are no published values for comparisons among peanut cultivars. The main objective of this study was therefore to evaluate and compare the effects of LLS severity on photosynthetic

PAGE 64

64 metabolism in two peanut cultivars with variable levels of resistance to LLS. It was hypothesized that peanut cultivars would differ in their photosynthetic response to LLS, which could help explain variable yield losses due to disease (e.g., Chapter 3) and improve screening of cultivars and modeling growth and yi eld responses of peanut to LLS. Material and Methods Experimental Site and Design Peanut leaves were sampled from field experiments conducted during the 2008 and 2009 growing seasons at the Plant Science Researc h and Educat ion Unit in Citra, Florida (29o230 N, 82o120 W) to study leaf photosynthetic responses to LLS. These experiments were part of a larger study conducted to quantify the growth and yield losses due to LLS in peanut cultivars with variable l evels o f resistance ( C hapter 3). The experiment was a 2x2 factorial arranged in a randomized complete block design with four replications. Cultivar and fungicide application were treated as fixed effects. Two cultivars were selected for differences in resistance to LLS: Carver (Gorbet, 2006) has poor resi stance to LLS; while York (Gorbet and Tillm an, 2011) has moderate resistance to LLS (Tillman et al., 2008). Fungicide treatments included no fungicide application and an industry standard fungicide schedule (Table 3 1 of C hapter 3) applied on a 14d interval commencing from approximately 40 days after planting (DAP). Plots were planted on 20 May in 2008 and 27 May in 2009 to maximize LLS pressure (Wright et al., 2006). Each plot consisted of six rows with a row spacing of 0.91 m and a row length of 4.6 m. Seeds were sown at a rate of 17 to 20 seeds per m of row using a conventional planter. Infurrow application of azoxystrobin was applied at a rate of 0.16 kg a.i. ha1 to help control soilborne diseases. I rrigation was applied as

PAGE 65

65 needed with an overhead linear move irrigation system. Plots were fertilized preplant with a 39 18 blended granular fertilizer at a rate of 560 kg fertilizer ha1. Gypsum was broadcast at a rate of 2240 kg ha1 in two equal appli cations around 35 to 40 DAP followed by the second application 10 to 14 d later. Disodium octaborate tetrahydrate was applied with the first two sprays at a rate of 5.6 kg ha1 to supply boron. A preemergence broadcast application of pendimethalin (0.92 kg a.i. ha 1) + diclosulam (0.42 kg a.i. ha1) and a postemergence application of imazapic (0.07 kg a.i. ha1) were used for weed control. Measures of Asat on Tagged Leaf Cohorts over Time In the central two rows of each plot, five plants were chosen at rand om and most recent fully expanded leaves on each main stem were tagged using colored plastic tags at 49 and 92 days after planting (DAP) in 2008, and at 50, 65, and 79 DAP in 2009, respectively. At approximately weekly intervals, around five leaflets per cultivar were randomly selected and analyzed for Asat on 6 cm2 leaf area using a LI 6400XT portable, openflow photosynthesis system (LI COR Inc., Lincoln, NB). Measurements were made between 1000 to 1400 h under partly cloudy to cloudfree days at 1800 m ol m 2 s 1 photosynthetic photon flux density (PPFD) using the 640002 LED light source (LI COR). The sample chamber CO2 concentration was maintained at 400 mol CO2 mol 1 air, and the flow rate of air through the sample chamber was set at 500 mol s1. Te mpe rature was maintained at 30 1C and relative humidity was maintained between 6070%, similar to environmental conditions. Severity of LLS was assessed on all of the tagged leaflets at approximately weekly intervals until defoliation using the ICRISAT diagrammatic scale ( Figure B 1, Subrahmanyam et al., 1995) to estimate the percent disease severity. The ICRISAT

PAGE 66

66 scale is a visual diagrammatic scale which depicts the proportion of the leaf area (%) that has necrotic lesions. Disease severity was defined as the proportion of total necrotic lesion area to total leaf area. Microscopic examination of conidiophores and conidia on diseased leaves indicated that C personatum was the dominant pathogen in both years (Figure 3 1 in C hapter 3) Relations between Ph otosynthesis and Disease Severity To examine the effects of disease severity on photosynthesis while controlling for leaf age, measures of leaf Asat were collected on a separate group of leaves of similar age. Measurements were conducted on fully expanded leaves (one to two leaves below the youngest fully expanded leaf) collected from the central two rows in Carver and York at around 25 days after appearance of visual disease symptoms. Symptomatic leaves were selected from nonsprayed plots to determine the reduction of photosynthesis at a given disease severity, whereas asymptomatic leaves were selected from fungicidesprayed plots on the same day to determine photosynthetic rates under diseasefree conditions. Leaflets were selected across a range of LLS s everities to quantify the relationship between host genotype photosynthesis and disease severity. Relative leaf Asat was calculated as the Asat of the LLS infected leaflet relative to the average of asymptomatic leaflets, collected on the same day. Leaf l ight response curves, CO2 response curves, and measures of dark adapted chlorophyll fluorescence were made using the LI 6400XT photosynthesis system, to evaluate determinants of photosynthetic metabolism. Across the two years, another separate group of 2030 leaflets of similar age, but differing disease severities (including asymptomatic leaflets) were selected for each cultivar as described above and analyzed for these variables. For light response curves, leaf CO2 assimilation rate ( A ) was

PAGE 67

67 measured at ei ght light levels ranging from 0 to 1800 mol m2 s1, which were used to estimate quantum efficiency of CO2 assimilation (CO2) as described by Boote et al. (1985). Plots of A vs. substomatal CO2 concentration ( ci) were used to estimate maximum carboxylati on velocity of Rubisco ( Vc,max) using the equations of Farquhar et al. (1 980). Data were corrected to 25oC using published temperature responses (Long and Bernacchi, 2003). Dark adapted chlorophyll fluorescence was measured using LI 6400 fitted with a leaf chamber fluorometer (Model LI640040), after dark adapting the leaves for around 30 min. Dark adapted efficiency of Photosystem II (PSII) photochemistry ( Fv/ Fm) was determined on asymptomatic and symptomatic leaves. Chlorophyll extraction was performed as described by Inskeep and Bloom (1985). Leaf discs (area = 2 cm2; diam. = 1.6 cm) were excised from asymptomatic ( n = 4) and symptomatic diseased leaflets ( n = 25). Leaf discs were placed in glass tubes wrapped in aluminum foil with 5 mL of N, Ndimethylformamide. The tubes were capped to minimize evaporation and then placed on a horizontal shaker for 72 h to extract leaf chlorophyll. Following extraction, 1 ml of solution was placed in a quartz cuvette and absorbance was measured at 647 and 664.5 nm with a spectrophotometer (BW VIS Model, StellarNet Inc., Tampa, FL). Chlorophyll content (g cm2) was calculated for each leaf sample using the equations described by Inskeep and Bloom (1985). Following the photosynthetic measurements, leaflets were detached and taken to the laboratory. All harvested leaves (and leaf discs) were scanned at 300 dpi using a flatbed scanner (Microtek ScanMaker 5800, Microtek Int. Inc., Industrial Park Hsinchu, Taiwan) and stored as .tiff files. Leaf images were processed using ASSE SS ver 2.0 image analysis software (American Phytopathological Society, St. Paul, MN) to give the

PAGE 68

68 percent disease severity ( Figure B 2, Erickson et al., 2003). Disease severity was defined as the fraction of total necrotic lesion area to total leaf area. D ata Analysis On a given sampling date, significant cultivar differences between disease severity and Asat on tagged leaf cohorts through time were analyzed using analysis of variance in the GLIMMIX procedure of SAS (SAS Institute, 2009). Relations between relative Asat and disease severity were analyzed using a nonlinear model, y = (1 x), as described by Bastiaans (1991). The nonlinear mixed effects (NLME) library of R (R Development Core Team, 2008) was used to estimate and assess significant diffe rences ( P < 0.05) in values between cultivars. Since, values did not differ by year within cultivar ( P > 0.05), the years were pooled for estimates. Comparisons of the impact of LLS between two peanut cultivars on other photosynthetic variables ( Vc,m ax, CO2, Fv/ Fm, and chlorophyll) were made under three disease categories: no disease (0% disease severity), low disease (015% disease severity), and high disease (15 30% disease severity). Categories were chosen to get approximately equal numbers ( n = 1 0 20) in each of three pooled categories (Kumudini et al., 2010). Results Disease Severity and Asat on Tagged Leaf Cohorts over Time Late leaf spot epidemics occurred in both years of the study, but appeared earlier in 2009 compared to 2008 (Figure 4 1). Late leaf spot symptoms were first observed visually in the field around 91 and 82 DAP during 2008 and 2009 on both cultivars, respectively. Across all tagged leaf cohorts, average maximum disease severity was 27 and 19% during 2008, and 25 and 12% during 2009 in Carver and York, respectively. Moreover, the time required from leaf tagging to attain these severities was 51 and 63

PAGE 69

69 days during 2008, and 42 and 49 days during 2009 in Carver and York, respectively. Hence, Carver showed more rapid disease progress than York during both years of the study (Figure 41). Initial net carbon assimilation rate of tagged leaves did not differ between cultivars across all the tagged cohorts, averaging 38.3 and 37.9 mol m2 s1 during 2008 and 2009, respectively. However Asat declined in all tagged cohorts with increasing leaf cohort ag e and disease severity (Figure 42). Despite differences in disease severity on individual leaf cohorts (Figure 4 1), Asat did not differ between the two cultivars on any given sampling date (except at 107 DAP in leaves tagged at 65 DAP during 2009) throughout the growing season. Relations between Photosynthesis and Disease Severity The mean photosynthetic rate of asymptomatic leaves averaged 36.0 mol m2 s1 and did not differ ( P > 0.05) between the two cultivars. In both cultivars, relative Asat was strongly and negatively related to increasing disease severity (Figure 4 3). The value obtained for both cultivars was greater than 1.0 ( P < 0.001), indicating that the photosynthetic impair ment extended beyond the necrotic lesion area. In addition, the value for York ( = 4.6 0.15) was significantly greater ( P < 0.05) than the value for Carver ( = 3.6 0.12), which indicated a greater reduction in Asat at a given disease severity for York compared to Carver. Notably, values were the same for a given cultivar across both years of the study. The cultivars did not differ in Vc,max for asymptomatic leaves with an average value of 148.7 mol m2 s1, indicating similar carboxylation capacity bet ween the two cultivars (Figure 4 4A). Maximum carboxylation velocity of Rubisco declined rapidly with increasing disease severity. Moreover, this decline was greater in the more resistant

PAGE 70

70 cultivar York compared to the cultivar with poor resistance to LLS (Carver). In fact, Vc,max of leaves with low disease severity (< 15%) was reduced by 62 and 84% compared to asymptomatic leaves in Carver and York, respectively. For leaves with high disease severity, the decline in Vc,max was 78 and 83% compared to asymptomatic leaves in Carver and York, respectively. Therefore, these results indicated that the reduction in Vc,max was much greater proportionately than disease severity and also that the cultivars differed in Vc,max more at relatively low disease severities compared to higher disease severities. Quantum efficiency of CO2 assimilation (CO2) of asymptomatic leaves also did not differ between the cultivars (0.064 mol CO2 mol1 quanta) and declined rapidly with disease severity (Figure 4 4B). However, the relative decline was not as severe as the decline in Vc,max. Quantum efficiency of leaves with low disease severity declined 23 and 43% compared to asymptomatic leaves in Carver and York, respectively. For leaves with high disease severity, this decline was 39 and 53% in Carver and York, respectively. Thus, the overall decline in CO2was g reater in York compared to Carver. As was observed in Vc,max, differences among cultivars in their response to disease severity was most evident in low disease category. While Vc,max and CO2 were strongly affected at low disease severities, dark adapted m aximum efficiency of PSII ( Fv/ Fm) was relatively unaffected in both cultivars at low disease (Figure 4 4C). In fact, even at high disease severities, Carver and York showed a similar reduction in Fv/ Fm of about 13% compared to their asymptomatic counterpar ts.

PAGE 71

71 Finally, leaf chlorophyll content also declined with increasing disease severity (Figure 4 4D). Mean chlorophyll content of asymptomatic leaves averaged 53.3 g cm2 and did not dif fer between the two cultivars. However, the reduction in chlorophyll at low disease severities was greater in York (43%) than Carver (26%). For leaves with high disease severity, this decline was 32 and 50% in Carver and York, respectively. Thus, the overall decline in chlorophyll content was greater in York compared to Carve r. Discussion The present study demonstrated differing disease severity and photosynthetic responses to LLS between two peanut cultivars. Despite differences in disease severity, similar reductions in Asat with leaf age were observed for all leaf cohorts across the two cultivars. These similar reductions in carbon assimilation between the cultivars could be explained in part by differing photosynthetic reductions in the leaf tissue beyond the necrotic leaf area (i.e., different values). A greater decline in Asat at a given disease severity was observed for York compared to Carver. Consistent with a greater value for York, greater reductions in chlorophyll, CO2, and especially Vc,max were observed in York compared to Carver. Delayed LLS disease progress, including reduced necrotic lesion area and/or defoliation has been demonstrated in many recently released cultivars (Tillman et al., 2008; Woodward et al., 2008). For example, for the same two cultivars used in the present study, disease severity, based o n area under canopy lesion area curve, was reduced by 30% in the York cultivar, which has greater resis tance to LLS (Chapter 3 ). After the onset of LLS, disease severity on tagged leaves differed on all sampling dates between the two cultivars (Figure 4 1) which was attributed to the slower progress of

PAGE 72

72 necrotic lesion area in York compared to Carver. Less necrotic area can lead to less sporulation for this polycyclic disease. Moreover, slow disease progress in resistant genotypes has been related to a long er latent period for the fungus (Chiteka et al., 1988). Although the cultivars differed in disease severity, leaf photosynthetic behavior was similar with leaf age during the growing season. Light saturated net assimilation rates did not differ between cul tivars at the first sampling date after tagging, which indicated similar photosynthetic potential between the cultivars. Throughout the season, Asat of tagged leaves declined with leaf age (Figure 4 2), attributable to both natural senescence (Guinn and Br ummett, 1993) and the incidence of LLS (Bourgeois and Boote, 1992; Nogues et al., 2002). Bassanezi et al. (2002) suggested that reduction in photosynthesis associated with disease was related to the trophic relationship of the pathogen with the host. Values of reported for some biotrophic pathosystems include 1.3 and 2.2 for Uromyces appendiculatus (Pers.:Pers) Unger on common bean ( Phaseolus vulgaris L.) (Lopes and Berger, 2001 and Bassanezi et al., 2001, respectively) and 2.3 for Phakospora pachyrhizi S yd. & P. Syd. on soybean ( Glycine max L. Merr.) (Kumudini et al., 2010). Thus, values of biotrophs have generally been equal to or very close to one, indicating minimal effects of biotrophic pathogens on photosynthesis beyond the visual lesion areas of t he leaf. However, values reported for hemibiotrophic pathosystems have generally been greater than 1.0 and often greater than 3.0. For example, Erickson et al. (2003) reported 6.1 for Marssonina brunnea f. sp. brunnea on poplar ( Populus spp.), Bassanezi et al. (2001) reported 3.8 for Phaeoisariopsis griseola (Sacc.) Ferr. on common bean, and Roloff et al. (2004) found

PAGE 73

73 values of 2.8 and 3.1 for Septoria albopunctata Cooke on Vaccinium spp. Consequently, greater reductions in photosynthesis beyond the visual lesion area seem to be more common with hemibiotrophs compared to biotrophs. Equally important to the magnitude of is whether cultivars differ in as this could be key for cultivar improvement and for modeling assimilation responses to disease. In this study, intraspecific differences in the photosynthetic response to LLS disease severity were found. Disease induc ed reduction in Asat was more severe ( = 4.6) in the cultivar with a higher level of resistance to LLS compared to the cultivar with a low level of resistance ( = 3.6). Bourgeois and Boote (1992) reported a decline in relative photosynthesis with disease severity of 4.0 (comparable to ) for peanut (cv. Florunner). Thus, genotypic variability in is likely to extend beyond the peanut cultivars examined here, and intraspecific variation in values has been reported in other pathosystems as well (Erickson et al. 2003; Zhang et al., 2009), but is often not seen in biotrophs (Bassanezi et al., 200 1; Kumudini et al., 2010). In this study, low disease severity was associated with a high parameter, but Zhang et al. (2009) reported low values associated with low disease severities in Populus cathayana Rehd., indicating that is not necessarily inversely correlated with disease severity. The greater reduction in photosynthetic capacity beyond the necrotic lesion area in York compared to Carver at a given disease severity could help to explain the similar reductions in canopy carbon assimilation through time between the two cultivars and their respective yield responses in the field (Chapter 3 ), despite differences in disease s everities as explained earlier.

PAGE 74

74 Re ductions in Asat with increasing disease severity were also associated with reductions in Vc,max, CO2, Fv/ Fm, and chlorophyll. R esults indicated that reductions in all the determinants of photosynthetic metabolism were much greater proportionally than the necrotic lesion area except for Fv/ Fm. Among the various determinants of photosynthesis studied, Vc,max showed the greatest relative decline, especially in the low disease category. Nogues et al. (2002) also concluded that decreased Vc,max was likely the primary determinant underlying the decline in Asat of tomato ( Lycopersicon esculentum Mill.) infec ted by Fusarium oxysporum f.sp. lycopersici. Moreover, the cultivars exhibited a differential response in Vc,max to disease severity, which could potentially explain the differing photosynthetic response between the cultivars. Taken together, these results indicated that the resistant cultivar, York, was unable to sustain photosynthesis in response to LLS despite its ability to slow the progression of disease (i.e. resistance). In conclusion, this study demonstrates the variability in photosynthetic respons e to LLS among cultivars with differing resistance levels. York, the cultivar with the higher level of resistance to LLS showed more photosynthetic impairment beyond the necrotic lesion area at a given disease severity compared to Carver, the cultivar with a poorer level of resistance. Possible mechanisms responsible for this greater photosynthetic impairment in York included a reduction in carboxylation velocity of Rubisco T hese findings were attributed in part to a lack of photosynthetic tolerance to LLS in the more resistant cultivar York. These results have potential implications in our efforts for selecting improved cultivars and predicting growth and yield responses of new peanut cultivars to leaf spot. For example, combining visual disease ratings wi th physiological

PAGE 75

75 measures of the parameter could result in the identification and selection of cultivars with slow disease progress (e.g., like York) and relatively low parameters (e.g., like Carver), which could contribute to reduced yield loss due to LLS, especially under low fungicide input production. Since collection of values is relatively intensive, spectral methods that detect chlorophyll as a surrogate for represents a future research need along with evaluation of a greater diversity of cul tivars.

PAGE 76

76 Figure 41. Progress of late leaf spot (LLS) severity (percent necrotic lesion area) on the individual leaf cohorts during 2008 and 2009 growing seasons for the two peanut cultivars Carver and York, grown under nofungicide application conditi ons. Leaf cohorts were tagged at 49 and 92 days after planting (DAP) during 2008 and at 50, 65, and 79 DAP during 2009 growing seasons, respectively. Significant differences ( P < 0.05) between cultivars for each time point are indicated by an asterisk.

PAGE 77

77 Figure 42. Light saturated CO2 assimilation rate (Leaf Asat) of individual leaf cohorts during 2008 and 2009 growing seasons for the two peanut cultivars Carver and York, grown under nofungicide application conditions. Leaf cohorts were tagged at 49 and 92 days after planting (DAP) during 2008 and at 50, 65, and 79 DAP during 2009 growing seasons, respectively. Significant differences ( P < 0.05) between cultivars for each time point are indicated by an asterisk.

PAGE 78

78 Figure 43. Relative light saturated leaf CO2 assimilation rate (Relative leaf Asat) in relation to disease severity (fraction necrotic lesion area) for two peanut cultivars Carver and York. parameter for York (4.6, n = 183) was greater ( P < 0.05) compared to that of Carver (3.6, n = 160).

PAGE 79

79 Figure 44. Changes in ( A ) maximum carboxylation velocity of Rubisco ( Vc,max), (B) quantum efficiency of CO2 assimilation (CO2), (C) dark adapted maximum efficiency of PSII photochemistry ( Fv/ Fm), and (D) chlorophyll content of leaves at no (0% disease severity), low (0 15% disease severity), and high (15 30% disease severity) disease categories for two peanut cultivars, Carver and York.

PAGE 80

80 CHAPTER 5 USING THE CROPGRO PEANUT MODEL TO SIMU LATE GROWTH AND YIEL D IN PEANUT CULTIVARS WIT H VARIABLE RESI STANCE LEVELS TO LAT E LEAF SPOT Abstract Late leaf spot (LLS) caused by Cercosporidium personatum (Berk. & Curt.) Deighton leads to significant reductions in peanut ( Arachis hypogaea L.) yield worldwide. This study was conducted to determine whether LLS e ffects on defoliation and photosynthesis can be incorporated into the CROPGRO Peanut model to simulate growth and yield reductions in peanut cultivars. Field experiments were conducted in 2008 and 2009 to collect data on the effects of LLS on biomass accum ulation and partitioning, leaf necrosis and defoliation, and total canopy photosynthesis (TCP) in peanut cultivars with more (York) and less (Carver) quantitative resistance to LLS After incorporating LLS damage as percent defoliation and necrotic area ( w ith cultivar specific value) the model accurately simulated crop growth and development for both cultivars despite different disease dynamics Simulated leaf, total crop, and pod yield values were in good agreement with measured data. Agreement between measured and simulated TCP values indicated correct crop C balance. A modification in the model to directly reduce leaf photosynthesis and quantum efficiency resulted in improved simulations of LLS effects on growth and yield of both cultivars. Correlations among measured defoliation and n ecrotic area with disease ratings indicated that visual disease ratings could be successf ully used to estimate necrosis and defoli ation and to correctly simulate LLS induced reductions in growth and yield. Results indicate d that the CROPGRO Peanut model has adequate capabilities to simulate LLS effects on growth and yield in peanut cultivars with differing levels of resistance to LLS when inputs on

PAGE 81

81 canopy necrotic area and defoliation are provided, which could be used to im prove model predictions to help reduce fungicide use and improve cultivar development Background In the southeastern USA e arly leaf spot (caused by Cercospora arachidicola S. Hori), and late leaf spot [caused by Cercosporidium personatum (Berk. & Curt.) Deighton] are among the most widespread and damaging foliar diseases of peanut ( Arachis hypogaea L.) (Nutter Jr. and Shokes, 1995). Yield reductions associated with these foliar diseases are related to premature loss of green leaf area (by necrotic tissue and defoliation) and reduction of leaf photosynthetic capacity (Pixley et al., 1990a; Bourgeois and Boote, 1992; C hapter 4). Late leaf spot (LLS) can cause yield losses of up to 50% in some cases (Pixley et al., 1990a ; Bourgeois et al., 1991). Current c rop prot ection strategies rely largely on fungicide applications cultural practices, and resistant cultivars. But economic and environmental concerns of fungicides have increased the demand for improved management strategies based on minimizing crop losses r ather than minimizing disease outbreaks. Decision support systems that can predict yield losses rather than controlling the outbreak of disease and hence provide improved disease management options are required to meet this demand. Crop models are important tools to evaluate growth and yield losses due to various biotic and abiotic stresses ( Boote et al., 1983 a ; Naab et al., 2004; Timsina et al., 2007). Models that have been used to predict the impact of foliar diseases on yield have generally incorporated the disease effects on defoliation and photosy nthesis (Batchelor et al., 1993; Boote et al 1993; Teng et al., 1998 ) Disease induced defoliation has been incorporated into the models by simply reducing the leaf area (Batchelor et al., 1993; Williams and Bo ote, 1995) ; however the reduction in leaf photosynthesis due to disease

PAGE 82

82 is more complex to model. The impact of LLS on leaf photosynthesis has been shown to be greater than can be accounted for by the visual lesion area (Bourgeois and Boote, 1992). In order to relate reductions in leaf photosynthesis to visual lesion area, Bastiaans (1991) proposed a relatively simple model, y = (1 x), where y is the relative photosynthetic rate of a diseased leaf compared to that of an asymptomatic leaf, x is the measured visual lesion area, and describes the relationship between virtual and visual lesion area. The virtual area represents loss of photosynthetic capacity beyond the visual lesion area. Thus, indicates whether the effect of disease on photosynthesis is higher ( > 1), lower ( < 1), or equal ( = 1) to that accounted for by the measured visual lesion area. Several studies have incorp orated this parameter into crop growth model s to estimate growth and yield losses due to biotrophic pathogens at the canopy level (Bastiaans 1993; Bassanezi et al., 2001; Robert et al ., 2004; Bancal et al., 2007) but all these researchers used a single v alue for all cultivars. However, the photosynthetic response ( parameter) can differ among various genotypes in their response to a pathogen (Erickson et al., 2003; Zhang et al., 2009; C hapter 4). In these cases, using a cultivar specific parameter in t he crop model should achiev e better predictions of diseaseinduced reductions in carbon assimilation, growth, and yield. Quantifying the effects of LLS on peanut cultivars with variable levels o f resistance and inclusion of the effect on phot osynthesis in yield loss model s are of great importance for a more complete understanding of growth and yield responses to diseases and should increase the accuracy of yield loss estimates The CROPGRO Peanut model (Boote et al., 1998a, 1998b) is a process oriented mech anistic crop growth model which considers crop carbon balance, crop and soil N

PAGE 83

83 balance, and soil water balance at the process level. This model has coupling points and procedures for entering pest damage to simulate growth and yield reductions associa ted w ith foliar pathogens like LLS (Batchelor et al., 1993; Boote et al., 1993). The primary impacts of disease are simulated as defoliation; however the impacts of virtual lesion area are currently simulated by simply defoliating more leaf area (hence zero photosynthesis on that area) rather than creating direct impact at the leaf level photosynthesis. This subroutine has been tested by some previous studies (Naab et al., 2004; Adom ou et al., 2005) to simulate LLS effects on peanut growth and yield. However, th ese studies did not include measured data on leaf necrosis and/or defoliation required by the disease subroutine in the model Either visual ICRISAT ratings were linearly regressed against necrosis values of 0 to 9% (based on data from Bourgeois et al., 1991) to obtain hypothesized necrosis values (Adomou et al., 2005) or variable defoliation and necrosis values were used to mimic leaf weight loss (Naab et al., 2004). Moreover, these studies were conducted only on cultivars lacking any k nown level of resis t ance to LLS and not on cultivars differing in their resistance levels to LLS. This signifies the need of having reliable disease assessment methods to provide accurate assessment of disease effects in cultivars with differing disease resistance levels wh ich could then be used as input to crop growth models and as evaluation methods in breeding programs. The overall objective of this study was therefore to evaluate the CROPGRO Peanut model for its ability to simulate the impacts of LLS on growth and yield reductions in peanut cultivars with variable resistance levels when inputs on necrosis and defoliation were provided. The different physiological response ( values) among

PAGE 84

84 peanut cultivars to LLS was incorporated into the model to test its ability to accurately simulate reductions in growth and yield In addition the model was modified to directly impact the light saturated leaf photosynthetic rate ( Asat) an d quantum efficiency of CO2 assimilation (QE ) depending on the percent necrotic area as observed in the physiological data on disease effec ts from the field experiments (C hapter 4) Multiple disease assessment methods were also analyzed to determine the b est method to estimate necrosis and defoliation, or to derive them from visual ratings (e.g. Florida 110 scale). Thus, t h ese objectives represented a step towards improved model simulation of LLS induced growth and yield losses that when coupled with a di sease simulation model or subroutine will contribute to reduced fungicide use and improv ed peanut cultivar development. Materials and Methods Experimental Site and Design Data were obtained from field experiments conducted during the 2008 and 2009 growing seasons at the Plant Science Research and Education Unit in Citra, Florida (29o2360 N, 82o120 W) to simulate growth and yield reductions associated with LLS epidemics on peanut cultivars of differing resistance using CROPGRO Peanut model. These exper iments were part of a larger study conducted to quantify the growth and yield losses and underlying physiological determinants due to LLS in peanut cultivars with variable levels of resistance. The soil at the experimental site was Gainesville loamy sand ( hyperthermic, coated Typic Quartzipsam ments). The experiment was a two by two factorial arranged in a randomized complete block design with four replications. Cultivar and fungicide application were treated as fixed effects. Two cultivars were selected for differences in resistance to LLS: Carver (Gorbet, 2006) has poor resistance

PAGE 85

85 to LLS; while York (Gorbet and Tillman, 2011 ) has moderate resistance to LLS (Tillman et al., 2008). Fungicide application included: (i) no fungicide application and (ii) an indus try standard fungicide schedule (Table 3 1) applied on a 14d interval commencing from approximately 40 days after planting (DAP). Sowing occurred during the latter part of the recommended planting window for North Central Florida on May 20 in 2008 and May 27 in 2009 to maximize LLS pressure (Wright et al., 2006). Each plot consisted of 6 rows spaced 0.91 m apart and 4.6 m long. Seeds were sown at a rate of 1720 seeds per meter row using a conventional planter. Standard management practices for irrigated peanut were employed during both years (Wright et al. 2006) to manage the crop as described in C hapter 3. Measures of Growth and Yield Starting 35 DAP, a 61cm section of row was harvested randomly from the outer two rows in each plot at approximately biweekly intervals to measure growth and partitioning A repre sentative subsample plant was selected from each harvested sample (Bourgeois et al., 1991; Pixley el al., 1990b). The remaining harvested sample was immediately oven dried for 72 h at 60oC and subsequently weighed. Leaflets and pods were separated from all subsamples and then l eaves, stems, and pods were oven dried to a constant weight Stem, leaf, and pod dry weights (DW) were determined for the entire sample by multiplying their respective fractions of the subsample times the total weight of the harvested sample. A 61 cm section of row was selected randomly from the outer two rows starting around 35 DAP to measure total canopy photosynthesis (TCP), using a 91 by 61 cm aluminum frame mylar chamber and a portable photosynthesis system (LICOR LI 620 0, Li Cor Inc., Lincoln, NE). Total canopy photosynthesis was calculated by adding the

PAGE 86

86 absolute dark respiration (measured under dark conditions) to the measured carbon exchange rate under full sunlight condi t ions, as described in C hapter 3. Measures of Disease Injury Occurrence and severity of late leaf spot were assessed visually based on the widely used Florida 110 scale ( Table B 1, Chiteka et al., 1988; Woodward et al., 2008; Woodward et al., 2010). Values of 1 to 4 indicate increasing leaf spot incidence on leaflets within the lower or upper canopy, but with no defoliation. Ratings from 4 to 10 are associated with increasing levels of defoliation (Chiteka et al., 1988). Ratings began when visual symptoms first appeared and continued every 710 days until harvest Microscopic examination of lesions on leaflets indicated that C. personatum was the dominant pathogen in both years. S potted wilt (caused by Tomato spotted wilt virus) and whit e mold (caused by Sc lerotium rolfsii Sacc. ) w ere not observed in the field plots during either growing season. Canopy defoliation and necrosis, the components that make up the Florida scale ratings, were also measured objectively by other methods throughout the growing season to compare to the more subjective Florida 1 to 10 scale assessment. To determine canopy lesion area, forty leaflets were randomly selected from the subsample plant. All leaflets were scanned at 300 dpi using a flatbed scanner (Microtek ScanMaker 5800, Mic rotek Int. Inc., Industrial Park Hsinchu, Taiwan) and stored as .tiff files. Leaf images were processed using ASSESS ver 2.0 image analysis software (American Phytopathological Society, St. Paul, MN) to give the percent necrotic area ( Figure B 2, Erickson et al., 2003). Leaf hue was used by the program to distinguish necrotic lesion area.

PAGE 87

87 To calculate percent canopy defoliation, different methods were employed as follows: (i) Loss of leaf DW from peak leaf weight was used to calculate percent defoliation for each treatment. After the occurrence of peak leaf DW, percent defoliation was calculated for that treatment on a given day as the ratio of the difference between peak leaf DW and leaf DW on the given sampling day to the peak DW observed (Pixley et al., 1 990b) (ii) Number of total nodes and nodes with missing leaflets were counted on the mainstem of each subs ample plant (Pixley et al., 1990b; Adomou et al., 2005). Percent defoliation was calculated as the ratio of missing to total nodes for each plant. T he first six and eight nodes were subsequently not considered in counting for Carver and York, respectively to account for the differences in nondisease induced leaf senescence which varied with life cycle and branch formation from lower nodes between the two cultivars (Ado mou et al., 2005). The i ntercept of the linear relation between missing nodes on the main stem and defoliation (based on leaf weight loss from peak weight) was used as a starting point to reach the number of nodes to exclude (Figure A 1 ) and was further optimized by the best CROPGRO Peanut model fit against the actual leaf DW data (iii) Number of total nodes and nodes with missing leaflets were also counted on the four dominant lateral branches and percent defol iation was determined as describ ed above. The percent defoliation values obtained through these methods ( Table s A 1 and A 2) were used as input for model simulations in file T under the header PCLA as described later To obtain estimates of necrosis and defoliation as a function of visual disease ratings, Florida 110 scale ratings were cor related to necrosis and to defoliation by comparing the slopes and intercepts of the relations using generalized least squares

PAGE 88

88 procedure of the nlme library of R (R development core team, 2008). Data were combined if slope and intercept of the given relation was not significant ( P > 0.05). Description of the CROPGRO Peanut Model The CROPGRO Peanut model is a mechanistic, process oriented model designed to simulate growth and development on a dail y basis using crop C, crop and soil N, and soil water balances (Boote et al, 1998a, 1998b) The code is modular and generic, such that crop specific parameters were removed from the code and placed into read in species, ecotype, and cultivars files. Crop development includes the rates of vegetative and reproductive development ( expressed as physiological days as a function of temperature, photoperiod, water, and N deficit) that governs dry matter partitioning to plant organs over time. Crop N balance includ es daily soil N uptake, N2 fixation, N mobilization from vegetative tissues, and N loss from abscised parts. Soil water balance includes infiltration of irrigation and rainfall, runoff, drainage, root uptake, soil evaporation, and plant transpiration. Crop C balance includes daily photosynthesis, growth and maintenance respiration, conversion and condensation of C to crop tissues, and C losses to abscised parts. The CROPGRO Peanut model computes canopy photosynthesis at hourly time steps using leaf level ph otosynthesis and hedgerow light interception (Boote and Pickering, 1994). This approach is more mechanistic and responsive to row spacing and plant density. Absorption of direct and diffused irradiance by sunlit versus shaded leaves is computed based upon canopy height and width, row direction, leaf angle, latitude of the site, day of year and time of day along with the predicted LAI (Boote and Pickering, 1994) Photosynthesis of sunlit and shaded leaves is computed using the asymptotic exponential light r esponse equation:

PAGE 89

89 )] / exp( 0 1 [max maxAPPFD QE A A (5 1) Where A is leaf CO2 assimilation rate ( mol m2 s1), Amax is the light saturated A (defined at 30oC, 350 mol CO2 mol1 air), QE is the quantum efficiency of the leaf (referenced at the same conditions as Amax) and PPFD is the photosynthetic photon flux density Both Amax and QE can be adjusted based on the following factors: ) ( ) ( ) ( ) ( ) ( ) (max 2 max, maxf SLW f leafN f CO f Chill f Temp f A Ag (5 2) ) ( ) ( ) (2 QE gf leafN f CO f QE QE (5 3) Where Amax,g is a cultivar specific coefficient, QEg is a C3 speciesspecific coefficient ; and f represents 01 adjustment function based on temperature of given day (temp), minimum temperature of previous night (Chill), atmospheric CO2 concentration (CO2), leaf N concentration (leafN), leaf thickness (specific lea f weight, SLW), and disease impact ( m a x or QE). Both m a x and QE are not present in the default model and have been introduced into the model as a part of t his study (as explained later) and are calculated as: ) 1 (maxPDLA orQE (5 4) Where PDLA is the fraction necrotic lesion area and repre sents the relationship between virtual and visual lesion area. Hourly canopy photosynthesis on a land area basis is computed by multiplying the photosynthetic rates for the sunlit and shaded leaves by t heir respective LAIs. The hourly rates are integrated over the 24 hr period to yield the total daily gross photosynthesis. Since CROPGRO is a sourcedriven crop model, correctly predicting canopy assimilation is important. Finally, growth of new tissues depends on daily

PAGE 90

90 carbohydrate availability, partitioning to various tissues, and respiration costs of tissue synthesis. CROPGRO Peanut Model Inputs D aily weather data (i.e. maximum and minimum temperature, rainfall, and solar radiation) were obtained on site from the Florida Automated Weather Network (FAWN 2011) for both growing seasons. D ata on site (latitude, longitude, and elevation), crop management practices (e.g. sowing date, spacing, plant population), and soil type and properties were provided as inpu t to the model as well. All data were entered in the standard file formats (*.PNX, *.PNA, *.PNT, *.WTH, and SOIL.SOL) needed for execution of the CROPGRO Peanut model in DSSAT ver 4.5 (Jones et al., 2003; Hoogenboom et al., 2009). Data on diseaseinduced necrosis and defoliation were also entered in the model input file to simulate disease damage as explained later. Procedure for Calibration of Genetic Coefficients The CROPGRO Peanut model requires genetic coefficients that describe crop phenology vegetat ive growth traits, and reproductive growth traits unique to a given cultivar (Boote et al., 1998b). Carver, the cultivar with poor resistance to LLS, is phenotypically similar to Florunner; thus the genetic coefficients of Florunner available in DSSAT vers ion 4.5 were chosen as the starting point for calibration. For York, genetic coefficients of Southern Runner were chosen initially, as this is a longer cycle cultivar that also has moderate levels of resistance to LLS. These coefficients were modified slig htly by comparing the simulated phenology, time series growth and yield with observed data from the fungicidesprayed plots only up to midseason, following the procedures outlined by Boote (1999). The extent of changes in coefficients was small

PAGE 91

91 (Table 5 1 ) and this exercise was done for more accurate prediction of timing of vegetative and reproductive growth. Model calibration was not the point of the study. Procedure for Simulating Disease Effects The effects of LLS on growth and yield of peanut were sim ulated by entering the measured levels of percent necrosis due to disease and associated percent leaf defoliation for the corresponding day of year into the crop performance file (File T) of DS SAT, under the headers of PDLA (necrosis percentage) and P CLA ( defoliation percentage, Tables A 1 and A 2 ) The default model code is designed to read this file and interpolates between dates to create leaf ar ea loss (also leaf mass and N loss) (Batchelor et al., 1993; Boote et al., 1993). The effect of necrosis (PDLA ) in the model is amplified by the presence of virtual lesion area (Bastiaans 1991; Erickson et al., 2003; C hapter 4). The mode l uses a virtual lesion effect of 4.0 irrespective of the cultivar used (estimated for Florunner by Bourgeois and Boote, 1992), m eaning there is an effective four unit decrease in photos ynthesizing leaf area for every unit of necrotic disease area. The model accounts for this effect by defoliating more leaf area based on this virtual effect. The model then runs the remainder of the season with reduced leaf area (and mass) resulting in reduced light interception, canopy photosynthesis and yield. In C hapter 4, the virtual lesion effect ( value) differed for the two cultivar s used in this study To account for this, values of 3.6 and 4.6 were used for Carver and York respectively in all the simulations conduc t ed in this study. Moreover, a modification was made in the model, whereby the effect of necrosis wa s placed directly on Asat (eq. 5 2) and QE (eq. 5 3) of single leaf to correspond directly to the physiological data collected in the field study So, instead of defoliating more leaf area as a result of necrosis and virtual lesion ( as in the default model by assuming linear relationship between necrotic

PAGE 92

92 area and effective leaf area) this modification to the model caused a cultivar specific reduc tion in Asat and QE assuming that the necrotic spot was directly in the photosynthetic area (Figure 5 1 eq. 5 1 to 5 4 ). Simulation runs conducted by this routine are referred to as modified model simulations. Statistical Evaluation of Model Performance Evaluation of model performance was conducted using root mean square error (RMSE) and the Willmott (1 981, 1982) index of agreement (D index). These statistical indicators were computed from observed and simulated variables (e.g. leaf mass, pod mass, and total crop biomass). The RMSE reflects the magnitude of the root mean sum of square differe nces between the predicted ( P ) and observed ( O ) values over time and is calculated as: n O P RMSEn i i i 1 2) ( The D index is a descriptive index that measures dispersion of the simulated and observed data, calculated as: n i i i n ii iO O O P O P index D1 2 1 2| | | | 1 Where n is the total number of observations, Pi is the predicted value for the i th measurement, Oi is the observed value for the i th measurement, and O is the overall mean of the observed values. A model performs well when the RMSE approaches zero and the D index is close to 1.0.

PAGE 93

93 Results and Discussion Leaf Weight and Leaf Area Simulations The temporal changes in measured leaf dry weight along with the simulated values are presented in Figures 5 2 and 5 3 The model (with no defoliation function) predicted leaf dry weight with good accuracy until around 95 and 85 DAP during 2008 and 2009 respectively for Carver and York, after which the model consistently over predicted the leaf dry matter accumulation. This occurred as leaf dry weight began to declin e due to LLS disease around that time and the model was unable to simulate those impact s (the model assumes no diseaseinduced necrosis and defoliation). This decline was observed even in fungicidesprayed plots (Figures 5 2 and 5 3 ), indicating that the 14d calendar based fungicide program did not achieve 100% disease control. To account for the LLS induced damage, the defoliation function in the model was used as explained earlier, whereby defoliation was input over time as calculated by different me thods. Inclusion of the defoliation function resulted in improved simulations of leaf weight (Figures 5 2 and 5 3 Table 5 2). Using leaf weight loss as defoliation function resulted in good agreement between simulated and measured values, as shown by high D index values (close to 1) and low RMSE values (Table 5 2). However, destructive growth sampling is required to collect leaf weight loss data, which is very labor intensive and time consuming. Counting main stem nodes and/or branch nodes to calculate def oliation is comparatively easier, but resulted in overestimation of leaf weight loss due to LLS (Figures 5 2 and 5 3 ). This could occur due to natural leaf senescence occurring on lower nodes, and can be accounted by subtracting the first few nodes on the main stem (Adomou et al., 2005). The f irst six and eight nodes on main stem in Carver and York were excluded to calculate defoliation, resulting in good agreement between

PAGE 94

94 simulated and measured values (Figures 5 2 and 5 3 Table 5 2). More nodes were excl uded in York to acc ount for higher leaf senescence of lower nodes which could occur due to longer life cycle and more branching from lower nodes in York Adomou et al. (2005) did not count the first four nodes in their study conducted on short duration cul tivars. These results showed the importance of a proper missing node method to estimate defoliation caused by LLS in peanut. The D index values for leaf weight obtained in this study (0.961.00) a fter accounting for defoliation (using leaf weight loss) w ere higher than those obtained in other studies using the same method ( 0.6 8 0.8 4 Adomou et al., 2005). Overall, simulated LAI followed the same trend as described above for leaf mass (data not shown). In both cultivars, simulated LAI was close to measured values until around 90 and 80 DAP during 2008 and 2009 respectively without using the defoliation function, and further values were im proved by including the diseaseinduced defoliation. Because LAI in the model is a function of both leaf mass and specifi c leaf area (SLA), this indicated that the SLA simulations were also good. Simulations of Total Biomass and Pod Weight The temporal changes in observed total biomass and pod weight along with the predicted values are presented in Figures 5 4 and 5 5 Witho ut the disease function, the model simulated total biomass and pod weight accurately up to around 110 DAP for both cultivars after which disease reductions on leaf area and assimilation become important. Thereafter in the model over predicted both total bi omass and pod yield resulting in high RMSE and only moderate dindex (Tables 5 3 and 5 4). To account for LLS induced damage, both measured necrosis and defoliation (calculated from leaf weight loss as explained in the previous section) were used as a disease function to run

PAGE 95

95 the simulations. This resulted in good agreement between the simulated and measured total biomass and pod weight across the growing season, as shown by D index values approaching 1.0 and low RMSE values (Tables 5 3 and 5 4). This w as ev ident under both fungicidesprayed and nonsprayed conditions during both years of the study. Reductions in total biomass and pod yield even under fungicidesprayed conditions corroborates other studies which showed less than optimal control of LLS with st andard program (Monfort et al., 2004; Woodw ard et al., 2010). Naab et al. ( 2004) also showed good agreement between simulated and measured total biomass (D index = 0.98) and pod yiel d (D index = 0.97) after entering hypothesized necrosis and defoliation values to mimic the leaf weight loss. All model simulations for York were run using a different virtual lesion effect in York ( = 4.6) and Carver ( = 3.6) Table 5 5 shows improvement in RMSE values for total biomass and pod yield simulations in York with the use of higher value for York. Although the improvement in RMSE value is small this shows so me importance of using cultivar specific virtual lesion effect in modeling growth and yield l osses where intraspecific differences in the physiological response to disease are seen Moreover, use of cultivar specific v alue (3.6 in Carver and 4.6 in York) resulted in improved predictions in 2009 when disease incidence occurred earlier in the season and progressed faster compared to 2008 (data not shown) This shows that the importance of the use of different values may depend on disease severity and the factors that contribute to greater disease severity such as time of onset, weather conditions, etc. In other cases where intraspecific variability does not ex ist or is minimal, it is possible to

PAGE 96

96 use same value for c ultivars (Robert et al., 2006; Bancal et al., 2007; Kumudini et al., 2010). Simulations showed that the slope of biomass accumulation rose smoothly after a short lag phase early in the vegetative g rowth period. This lag phase was longer in York compared to Carver. Both measured and simulated values indicated a peak dry weight of around 10,000 kg ha1 in both cultivars. Thereafter, the simulated dry m atter accumulation showed a decline in both cultiv ars which occurred due to LLS incidence. Comparison of the seasonal patterns of simulated pod weight among the cultivars also showed later initiation of pod fill, slower pod growth rate, and longer duration of pod fill in York compared to Carver. Simulatio ns of Canopy Photosynthesis Figure 5 6 shows model simulations of TCP, with and without the disease function. Without the necrosis and defoliation inputs, the model simulations were poor with low D index and high RMSE values. Canopy photosynthesis simulati ons improved with the use of the disease f unction, except for the sharp drops caused by cloudy days (Figure 5 6 Table 56) On cloudy days, TCP was actually measured under full sun ( late morning and between cloud breaks), whereas simulated photosynthesis rates were based on total daily solar radiations and hence were low on cloudy days Agreement between simulated and measured TCP values indicated correct crop C balance in the model. Dec line in TCP in fungicide sprayed plots after the incidence of LLS agai n indicated less than 100% control of disease by calendar based fungicide schedule. Reductions in TCP due to LLS have also been observed by Bourgeois and Boote (1992) while working with Florunner cultivar in peanut.

PAGE 97

97 Simulations with the Modified Model The modified model was used to directly impact Asat and QE (C hapter 4) to account for the virtual lesion area ( value). Figure 5 7 show s model simulations of total biomass and pod weight under nonsprayed conditions for 2009, with the default and modified mo del The modified model resulted in some improvement in the simulations as shown by the increase in D index and decrease in RMSE values compared to the default model simulations for both years (Table 5 7 and 5 8). Simulations for all treatments in 2008 and under fungicidesprayed conditions for 2009 are not shown as the trends were similar to the ones shown in Figure 5 7 Improvement in model predictions of total crop and pod weight showed the importance of more mechanistically including the effects of LLS on Asat and QE in modeling LLS impacts. Estimating Disease Induced Percent Necrosis and Defoliation from Florida 1 10 Scale Percent canopy necrosis was positively related to Florida 110 visual rating scale ( P < 0.001). The slope of the relation between percent necrosis (measured with scan system) and Florida 110 scale was not affected ( P > 0.05) by cultivar, fungicide schedule, or year (Figure 5 8 ). Adomou et al. 2005 also developed a relationship between percent necrosis and the ICRISAT 1 9 scale which i s similar to Florida scale. However, their relation was developed by linearly regressing necrosis from zero at ICRISAT score of 1 to a maximum necrosis of 9% at ICRISAT score of 8, based on necr osis data from Bourgeois et al. ( 1991) The relation establish ed in this study is more reliable and rigorously tested as it was developed from actual scan measured data o n two peanut cultivars under fung icide sprayed and nonsprayed conditions across two years of study. T his relationship between the visual rating scale and necrosis did no t

PAGE 98

98 differ among treatments in this study and can be used to derive percent necrosis from visual rating scale where measured data on necrosis is not available Percent canopy defoliation (calculated from leaf weight loss) was also posit ively related to Florida 110 scale ( P < 0.001). The slope of the relation between defoliation and Florida 110 scale was not affected by fungicide schedule or year, however, it was influenced by the cultivar used (Figure 5 9 ). This showed that the Florida scale ratings were not consistent for defoliation among the two cultivars These results indicated that a cu ltivar specific relation is required to derive defoliation from visual rating scale. Simulations U sing Estimated Necrosis and Defoliation from Flor ida 1 10 Scale Simulations conducted using percent necrosis and defoliation inputs derived from visual disease ratings (Florida 110 scale) were in good agreement with the simulated values using measured disease data (Figure 510; Tables 59 and 5 10). The temporal dynamics of total crop biomass and pod weight were consistent between both simulations and were in good agreement with the measured data as well, as shown by D index values approaching 1.0 and RMSE values approaching zero. These results indicated that input for disease function can be derived from the less labor intensive and less time consuming visual disease scouting methods rather than relying on intensive measurements for diseaseinduced necrosis and defoliation The D index values for pod wei ght (0.950.99) obtained in this study were higher than the ones (0.690.87) obtained by Adomou et al. (2005) while using actual defoliation and estimated necrosis data. This indicated more accuracy of the relations developed in this study in estimating qu antitative disease da mage from visual ratings.

PAGE 99

99 Conclusions T he CROPGRO Peanut model can be used to simulate the influence of foliar diseases (e.g. late leaf spot) on photosynthesis, growth partitioning, and yield reductions in peanut cultivars with differ ing levels of resistance to LLS when inputs on percent canopy necrotic area and defoliation are provided. Estimating defoliation from main stem nodes worked well, but obtaining the appropriate starting point (node 7 in Carver and node 9 in York) for diseas e induced defoliation is important. Significant relations between Florida 110 visual rating scale and measured necrosis and defoliation can be used to simulate LLS induced grow th and yield reductions where detailed sampling on disease damage is not conduc ted. This approach of entering LLS damage based on scouting information can be extended to other important foliar diseases of peanut and ot her legumes to simulate diseaseinduced growth and yield reductions. The adjustments made to the model code were rel atively minor but resulted in improved predictions (10% over prediction of final pod yield by the modified model compared to 14% by the default model across all treatments) of t he effect of LLS on grow th and development of peanut. It is recommend ed that th e ne xt version of the CROPGRO model should include the changes to simulate effects of necrotic area directly on leaf photosynthetic traits Use of cultiva r specific parameter was warranted as it resulted in improved simulations of growth and yield. The s ensitivity of the model to percent necrosis (PDLA) and parameter among different cultivars needs further investigation. Future model development should include an independent disease simulator which can predict LLS induced necrosis and defoliation based on weather conditions. This would allow improved predicti ons of the impacts of foliar diseases like

PAGE 100

100 LLS on growth and yield and hence result in reduced fungicide use and improved cultivar development without relying on the disease damage inputs from scouti ng

PAGE 101

101 Table 5 1. Genetic coeffici ents of the cultivars Carver, Florunner, York and Southern Runner used for model simulations. Cultivar Genetic coefficient Abbreviation Carver Florunner York S. Run Time from emergence to flower appearance, pd EM FL 20.2 21.2 24.0 22.9 Time from beginning flower to beginning pod, pd FL SH 9.2 9.2 10.6 9.2 Time from beginning flower to beginning seed, pd FL SD 18.8 18.8 20.4 18.2 Time from beginning seed to maturity, pd SD PM 74.3 74.3 76.0 82.6 Time fro m beginning flower to end of leaf expansion, pd FL LF 85 88 86 91 Maximum leaf photosynthetic rate, mg CO2 m 2 s 1 LFMAX 1.40 1.40 1.33 1.30 Specific leaf area, cm 2 g 1 SLAVR 232 260 232 265 Maximum size of full leaf, cm 2 SIZELF 18 18 14 17 Maximum f raction of daily growth partitioned to seed and shell XFRT 0.92 0.92 0.83 0.85 Maximum weight per seed, g WTPSD 0.68 0.69 0.68 0.63 Seed filling duration, pd SFDUR 42.9 40.0 43.0 40.0 Seeds per pod, no. pod 1 SDPDV 1.72 1.65 1.71 1.65 Time to reach ful l pod load, pd PODUR 26 24 32 30 pd, Photothermal days S. Run. Southern Runner

PAGE 102

102 Table 5 2. Root mean square error (RMSE) and index of agreement (D index) values for leaf dry weight for two peanut cultivars grown under fungicidesprayed (Fung) and nonspra yed (NF) conditions during 2008 and 2009. Defoliation inputs include no defoliation input (No def ), defoliation based on leaf weight loss (LW), main stem defoliation (MS), branch defoliation (Branch), and main stem defoliation excluding first si x and eight nodes in Carver and York (MS + offset). Defoliation input Year Cultivar Fung icide No def LW MS Branch MS+offset No def LW MS Branch MS+offset -----------------------D index -----------------------------------------RMSE (kg ha 1 ) ---------------2008 Carver NF 0.66 1.00 0.99 0.97 1.00 951 112 133 293 108 Fung 0.67 0.96 0.99 0.95 0.98 822 298 98 277 205 York NF 0.60 0.98 0.91 0.86 0.99 1066 235 399 487 184 Fung 0.68 0.99 0.88 0.78 0.98 892 174 446 621 201 2008 Carv er NF 0.70 0.99 0.93 0.95 0.98 913 196 409 392 238 Fung 0.66 0.97 0.97 0.93 0.96 834 260 237 356 300 York NF 0.61 0.99 0.87 0.84 0.99 1105 184 485 591 170 Fung 0.71 0.96 0.79 0.65 0.96 807 269 552 809 282

PAGE 103

103 Table 5 3. Root mean square error (RMSE) and index of agreement (D index) values for total biomass for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions at Citra, FL during 2008 and 2009. Simulations were conducted without (No dis) and with disease function including necrosis and defoliation (Dis function). Year Cultivar Fung icide No dis Dis function No dis Dis function ------D index -------RMSE (Kg ha 1 ) -2008 Carver NF 0.93 0.99 1698 517 Fung 0.93 0.98 1702 941 York NF 0.81 0.95 2968 1357 Fung 0.90 0.97 2311 1072 2009 Carver NF 0.92 0.97 1869 1071 Fung 0.95 0.99 1463 664 York NF 0.88 0.98 2517 921 Fung 0.94 0.98 1838 1091 Table 5 4. Root mean square error (RMSE) and index of agreement (D in dex) values for pod weight for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions at Citra, FL during 2008 and 2009. Simulations were conducted without (No dis) and with disease function including necrosis and defolia tion (Dis function). Year Cultivar Fung icide No dis Dis function No dis Dis function ------D index -------RMSE (Kg ha 1 ) -2008 Carver NF 0.97 0.99 638 359 Fung 0.96 0.96 751 700 York NF 0.93 0.95 802 622 Fung 0.97 0.97 522 497 2009 Carver NF 0.95 0.98 790 522 Fung 0.98 0.99 596 477 York NF 0.95 0.97 824 537 Fung 0.98 0.99 526 455

PAGE 104

104 Table 5 5. Virtual lesion effect ( value) on r oot mean square error (RMSE) values for total biomass and pod yield for York grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. Year Cultivar Fung icide = 3.6 = 4.6 = 3.6 = 4.6 -Total bio mass (Kg ha 1 ) --Pod weight (Kg ha 1 ) -2008 York NF 1393 1357 647 622 Fung 1108 1072 508 497 2009 York NF 945 921 549 537 Fung 1106 1091 472 455 Table 5 6. Root mean square error (RMSE) and index of agreement (D index) values for total canopy photosynthesis (TCP) for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. Simulations were conducted without (No dis) and with disease function including necrosis and defoliation (Dis f unction). Year Cultivar Fung icide No dis Dis function No dis Dis function ------D index -------RMSE (mg m 2 s 1 ) -2008 Carver NF 0.60 0.89 0.74 0.43 Fung 0.57 0.78 0.67 0.51 York NF 0.71 0.96 0.68 0.27 Fung 0.71 0.93 0. 61 0.31 2009 Carver NF 0.50 0.89 0.82 0.42 Fung 0.50 0.71 0.61 0.47 York NF 0.64 0.93 0.65 0.32 Fung 0.70 0.86 0.52 0.36

PAGE 105

105 Table 5 7. Comparison of default vs. modified model for r oot mean square error (RMSE) and index of agreement (D index ) values for total biomass for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. Year Cultivar Fung icide Default Modified Default Modified -----D index ------RMSE (Kg ha 1 ) -2008 Carver NF 0.99 0.99 517 522 Fung 0.98 0.98 941 845 York NF 0.95 0.96 1357 1194 Fung 0.97 0.98 1072 930 2009 Carver NF 0.97 0.97 1071 1122 Fung 0.99 0.99 664 623 York NF 0.98 0.98 921 798 Fung 0.98 0.98 1091 100 7 Table 5 8. Comparison of default vs. modified model for r oot mean square error (RMSE) and index of agreement (D index) values for pod weight for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. Year Cultivar Fung icide Default Modified Default Modified ------D index -------RMSE (Kg ha 1 ) -2008 Carver NF 0.99 0.99 359 281 Fung 0.96 0.97 700 617 York NF 0.95 0.96 622 540 Fung 0.97 0.98 497 423 200 9 Carver NF 0.98 0.97 522 553 Fung 0.99 0.99 477 435 York NF 0.97 0.98 537 482 Fung 0.99 0.99 455 406

PAGE 106

106 Table 5 9. Statistics of total biomass as simulated by CROPGRO Peanut model using disease function derived from measured data (Dis funms d) and estimated data from Florida 110 scale (Dis funest) for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. Year Cultivar Fung icide Dis fun msd Dis fun est Dis fun msd Dis fun est ------D index -------RMSE (Kg ha 1 ) --2008 Carver NF 0.99 0.99 522 580 Fung 0.98 0.97 845 1004 York NF 0.96 0.94 1194 1427 Fung 0.98 0.96 930 1242 2009 Carver NF 0.97 0.96 1122 1171 Fung 0.99 0.99 623 644 Yo rk NF 0.98 0.98 798 788 Fung 0.98 0.98 1007 842 D index, index of agreement; RMSE, root mean square error. Table 5 10. Statistics of pod yield as simulated by CROPGRO Peanut model using disease function derived from measured data (Dis funmsd) and estimated data from Florida 110 scale (Dis funest) for two peanut cultivars grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. Year Cultivar Fung icide Dis fun msd Dis fun est Dis fun msd Dis fun est -----D index ------RMSE (Kg ha 1 ) --2008 Carver NF 0.99 0.99 281 297 Fung 0.97 0.97 617 664 York NF 0.96 0.95 540 645 Fung 0.98 0.97 423 526 2009 Carver NF 0.97 0.97 553 559 Fung 0.99 0.99 435 398 York NF 0.98 0.98 482 415 Fung 0.99 0.99 406 330 D index, index of agreement; RMSE, root mean square error.

PAGE 107

107 Figure 51. Relationship between necrotic area and effective leaf area in the default model and between necrotic area and r elative Asat (ratio of photosynthetic rate of diseased leaflet to the averag e of asymptomatic leaflets) and QE (quantum efficiency of CO2 assimilation ) in the modified model.

PAGE 108

108 Figure 5 2 Simulated and measured leaf dry weight vs. days after planting (DAP) for peanut culti vars Carver (C) and York (Y) grown under fungicidesprayed (F) and nonsprayed (NF) condit ion s during 2008. Symbols represent treatment means ( n =4). Lines represents simulations based on defoliation inputs including no defoliation input (No def), defoliati on based on leaf weight loss (LW), main stem defoliation (MS), branch defoliation (Branch), and main stem defoliation excluding first six and eight nodes in Carver and York (MS + offset ). Vertical bars represent standard error of the mean.

PAGE 109

109 Figure 5 3 Simulated and measured leaf dry weight vs. days after planting (DAP) for peanut cultivars Carver (C) and York (Y) grown under fungicidesprayed (F) and nonsprayed (NF) condit ions during 2009. Symbols represents measured treatment means ( n =4). Lines represents simulations based on defoliation inputs including no defoliation input (No def), defoliation based on leaf weight loss (LW), main stem defoliation (MS), branch defoliation (Branch), and main stem defoliation excluding first six and eight nodes in Carver and York (MS + offset ). Vertical bars represent standard error of the mean.

PAGE 110

110 Figure 5 4 Simulat ed and measured total biomass for peanut cultivars Carver ( C) and York (Y) grown under fungicidesprayed (F) and nonsprayed (NF) conditions at Cit ra, FL during 2008 and 2009. Symbols represents measured treatment means ( n =4). Lines represent simulations including no disease function (No dis) and disease function with necrosis and defoliation (Dis function). Vertical bars represent standard error o f the mean.

PAGE 111

111 Figure 5 5 Simulated and measured pod weight for peanut cultivars Carver ( C) and York (Y) grown under fungicidesprayed (F) and nonsprayed (NF) conditions at Citra, FL during 2008 and 2009. Symbols represents measured treatment means ( n = 4). Lines represent simulations including no disease function (No dis) and disease function with necrosis and defoliation (Dis function). Vertical bars represent standard error of the mean.

PAGE 112

112 Figure 5 6 Simulated and measured mid day total ca nopy photosynthesis (TCP) for peanut cultivars Carver ( C) and York (Y) grown under fungicidesprayed (F) and nonsprayed (NF) conditions at Citra, FL during 2008 and 2009. Symbols represents measured treatment means ( n =4). Lines represent simulations including no disease function (No dis) and disease function with necrosis and defoliation (Dis function).

PAGE 113

113 Figure 5 7 Simulated and measured total biomass and pod weight for peanut cultivars Carver (C) and York (Y) grown under nofungicide application (NF) conditi ons during 2009. Symbols represents measured treatment means ( n =4). Lines represents simulations with default model routine (Default model) and modified model routine (Modified model). Vertical bars represent standard error of the mean.

PAGE 114

114 Figure 5 8 The relationship between percent necrosis and Florida 1 10 visual rating scale.

PAGE 115

115 Figure 5 9 The relationship between defoliation and Florida 110 visual rating scale for the two peanut cultivars Carver and York.

PAGE 116

116 Figure 5 10. Simulated and measur ed total biomass and pod weight for peanut cultivars Carver (C) and York (Y) grown under nofu ngicide application (NF) conditions during 2009. Symbols represents measured treatment means ( n =4). Lines represents simulations with disease function derived from measured data (Dis functionmsd) and estimated data from Florida 110 scale (Dis function est) Vertical bars represent standard error of the mean.

PAGE 117

117 CHAPTER 6 SUMMARY AND CONCLUSI ONS The overall goal of this study was to characterize LLS severity and progression and its impact on growth, yield and photosynthetic metabolism of peanut cult ivars with differing levels of resistance to LLS. To that end, field experiments were conducted during the growing season of 2008 and 2009 at Citra, FL evaluating tw o peanut cultivars with more (York) and less (Carver) quantitative resistance to LLS grown under fungicidesprayed and nonsprayed conditions. In addition, a simulated experiment was created based on f ield experiment data to predict the growth and yield reductions associated with LLS. This study demonstrated that the more resistant cultivar York contributed to delayed disease progress resulting in slower development of canopy lesion area and reduced defoliation. Despite this, yield improvement over the less resistant cultivar, Carver, was marginal and only occurred during the second year of the study when LLS pressure was high. A d iminishing effect of fungicide on the more resistant cultivar for pod yield was not observed, as yield gains associated with fung icide a pplication were the same for both cultivars across both years of the study. So, foliar application of fungicides still play ed an important role in minimizing crop yield loss caused by LLS epidemics T hese findings were attributed in part to a lack o f improved physiological tolerance to LLS in York as shown by similar reductions in TCP in both cultivars despite reduced disease severity in York These results indicate that c ombining resistance to disease progression with enhanced ability to sustain canopy photosynthetic capacity in the cultivar selection procedure could provide significant improvement in our efforts to improve peanut yields under diseased conditions.

PAGE 118

118 T his study also demonstrated variability in the leaf level photosynthetic response to L LS between cultivars with differing resistance levels to LLS York, the cultivar with the higher level of resistance to LLS showed more photosynthetic impairment beyond the necrotic lesion area at a given disease severity compared to Carver, the cultivar w ith a poorer level of resistance. Possible mechanisms responsible for this greater photosynthetic impairment in York included a reduction in carboxylation velocity of Rubisco and/or reduction in chlorophyll content T hese findings were attributed in part t o a lack of photosynthetic tolerance to LLS in the more resistant cultivar York. Thus, r eductions in photosynthetic rate in the virtual area were most likely due to a decline in protein content This decline could be occurring uniformly throughout the nonnecrotic leaf area in response to LLS infection rather than just the area around the necrotic lesions. These results have potential implications in our efforts for selecting improved cultivars and predicting growth and yield responses of new peanut cultivars to LLS For example, combining visual disease ratings with physiological measures of the parameter could result in the identification and selection of cultivars with slow disease progress (e.g., like York) and relatively low parameters (e.g., like Carver), which could contribute to reduced yield loss due to LLS, especially under low fungic ide input production. In this study, the CROPGRO Peanut model was successfully used to simulate the influence of foliar diseases (e.g. late leaf spot) on photosynthesis, growth partitioning, and yield reductions in peanut cultivars with differing levels of resistance to LLS given inputs on canopy necrotic area and defoliation. Estimating defoliation from main stem nodes worked well, but obtaining the appropriate starting point (node 7 in Carver and

PAGE 119

119 node 9 in York) for diseaseinduced defoliation is import ant. Significant relations between the Florida 1 10 visual rating scale and measured necrosis and defoliation can be used to simulate LLS induced grow th and yield reductions where detailed sampling on disease damage is not conducted. The adjustments made t o the model code were rel atively minor but resulted in improved predictions of t he effect of LLS on growth and developm ent of peanut. So, the next version of the CROPGRO model should include the changes to simulate effects of necrotic area directly on leaf photosynthetic traits Use of a cultivar specific parameter was warranted as it resulted in improved simulations of growth and yield. The sensitivity of the model to percent necrosis (PDLA) and parameter among different cultivars needs further investi gation. Future model development should include an independent disease simulator which can predict LLS induced necrosis and defoliation based on weather conditions. This would allow improved predicti ons of the impacts of foliar diseases like LLS on growth and yield, and hence result in reduced fungicide use and improved cultivar development without relying on the disease damage inputs from scouting

PAGE 120

120 APPENDIX A DEFOLIATION, NECROSI S, DRY BIOMASS, AND CANOPY PHOTOSYNTHESI S VALUES FOR CARVER AN D YORK Table A 1. Calculated percent canopy defoliation (PCLA) and necrosis (PDLA) for peanut cultivars Carver and York grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 across various DAP (days after planting) and DoY (day of year). PCLA values were calculated based on leaf weight loss (LW), main stem defoliation (MS), branch defoliation (Branch), and main stem defoliation excluding first six and eight nodes in Carver and York (MS + offset). Fungicide PCLA (%) Cultivar treatment DAP DoY LW MS Branch MS+offset PDLA (%) Carver NF 35 176 0.0 0.0 0.0 0.0 0.00 49 190 0.0 0.0 0.0 0.0 0.00 64 205 0.0 13.6 0.0 0.0 0.25 77 218 0.0 19.1 16.6 0.0 0.55 91 232 0.0 23.2 39.8 0.0 1.26 105 246 16.0 39.3 52.5 19.9 2.65 119 260 82.2 90.6 89.3 87.9 9.33 128 269 98.7 100.0 99.3 100.0 9.33 Fung 35 176 0.0 0.0 0.0 0.0 0.00 49 190 0.0 0.0 0.0 0.0 0.00 64 205 0.0 11.3 0.0 0.0 0.37 77 218 0.0 20.1 14.8 0.0 0.49 91 232 0.0 24.4 42.9 1.7 0.88 1 05 246 21.2 47.7 59.1 29.9 1.31 119 260 41.4 65.4 73.6 54.2 3.55 133 274 90.3 95.7 94.4 94.2 8.65 York NF 35 176 0.0 0.0 0.0 0.0 0.00 49 190 0.0 0.0 0.0 0.0 0.00 64 205 0.0 16.3 0.0 0.0 0.36 77 218 0.0 26.6 22.9 0.0 0.54 91 2 32 12.1 42.7 52.5 13.6 0.62 105 246 27.5 57.9 60.9 39.6 1.37 119 260 43.7 65.0 76.3 49.2 3.63 133 274 84.1 88.2 90.7 84.3 5.83 147 288 93.4 98.7 98.0 98.2 10.4 Fung 35 176 0.0 0.0 0.0 0.0 0.00 49 190 0.0 0.0 0.0 0.0 0.00 64 205 0.0 17.4 0.0 0.0 0.15 77 218 0.0 22.9 31.0 0.0 0.50 91 232 0.0 39.0 44.9 11.0 0.54 105 246 19.9 54.7 63.4 35.4 0.86 119 260 45.6 58.1 81.6 40.2 2.61 133 274 66.0 75.9 82.9 66.0 3.55 147 288 83.6 94.6 91.3 92.5 7.75

PAGE 121

121 Tabl e A 2. Calculated percent canopy defoliation (PCLA) and necrosis (PDLA) for peanut cultivars Carver and York grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2009 across various DAP (days after planting) and DoY (day of year). P CLA values were calculated based on leaf weight loss (LW), main stem defoliation (MS), branch defoliation (Branch), and main stem defoliation excluding first six and eight nodes in Carver and York (MS + offset). Fungicide PCLA (%) Cultivar treatme nt DAP DoY LW MS Branch MS+offset PDLA (%) Carver NF 35 182 0.0 0.0 0.0 0.0 0.00 49 196 0.0 0.0 0.0 0.0 0.00 63 210 0.0 18.1 0.0 0.0 0.31 77 224 0.0 23.9 13.6 0.0 0.58 90 237 9.6 43.2 41.3 22.1 3.00 103 250 60.8 66.2 85.3 54.1 9.1 9 118 265 96.1 92.7 96.7 90.3 12.20 Fung 35 182 0.0 0.0 0.0 0.0 0.00 49 196 0.0 0.0 0.0 0.0 0.00 63 210 0.0 16.7 0.0 0.0 0.24 77 224 0.0 22.8 14.5 0.0 0.25 90 237 0.0 35.9 45.2 14.1 0.51 103 250 17.1 41.9 53.1 21.6 2.45 118 265 65.9 68.3 78.7 57.0 5.66 125 272 81.6 87.8 93.9 84.1 8.40 York NF 35 182 0.0 0.0 0.0 0.0 0.00 49 196 0.0 0.0 0.0 0.0 0.00 63 210 0.0 23.8 0.0 0.0 0.09 77 224 0.0 34.7 18.5 0.0 0.18 90 237 0.0 42.2 47.0 11.6 0.93 103 250 28.2 59.1 73.4 40.4 1.44 118 265 72.7 81.9 91.9 75.0 3.92 131 278 81.6 86.6 91.9 81.7 5.19 142 289 85.9 88.3 95.0 84.4 8.47 Fung 35 182 0.0 0.0 0.0 0.0 0.00 49 196 0.0 0.0 0.0 0.0 0.00 63 210 0.0 24.4 0.0 0.0 0.12 77 2 24 0.0 24.1 20.6 0.0 0.05 90 237 0.0 42.7 61.0 16.8 0.05 103 250 8.7 57.7 69.2 37.5 0.94 118 265 47.3 65.5 80.4 50.6 1.86 131 278 52.6 81.1 93.1 73.9 2.87 142 289 70.1 85.5 93.6 80.4 6.11

PAGE 122

122 Table A 3. Leaf, stem, pod, and total dr y weight (DW) and leaf area index (LAI) mean values ( n = 4) vs. days after planting (DAP) and day of year (DoY) for peanut cultivars Carver and York grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008. Cultivar Fungicide DAP Do Y LAI Leaf DW Stem DW Pod DW Total DW Carver NF 35 176 1.13 405 393 0 798 49 190 2.72 1074 1318 5 2396 64 205 5.28 2034 3113 506 5654 77 218 4.84 2074 3434 1453 6960 91 232 5.44 2235 4252 2515 9036 105 246 4.00 1877 4229 4006 1018 9 119 260 0.69 398 4088 4340 8897 128 269 0.03 30 3144 4064 7274 Fung 35 176 1.13 405 393 0 798 49 190 2.72 107 4 1318 5 2396 64 205 4.31 1747 2637 359 4744 77 218 4.72 1976 3161 1285 6422 91 232 4.90 1979 3852 3085 8959 105 246 3.34 1559 3771 3615 8995 119 260 2.26 1159 3326 3571 8064 133 274 0.95 191 3489 4442 8224 York NF 35 176 0.58 222 171 0 393 49 190 1.86 758 890 0 1649 64 205 2.93 1150 1551 70 2770 77 218 5.54 2315 3279 705 6299 91 232 5.60 2035 3324 1041 6428 105 246 3.60 1678 3433 1798 6971 119 260 2.77 1300 3560 3517 8462 133 274 2.15 369 2982 3283 6743 147 288 0.28 153 2770 3249 6214 Fung 35 176 0.58 222 171 0 393 49 190 1.86 7 58 890 0 1649 64 205 3.48 1345 1786 122 3252 77 218 5.35 2102 2922 661 5685 91 232 6.48 2455 4052 1760 8302 105 246 4.71 1966 4228 1831 8100 119 260 2.88 1336 3650 3592 8679 133 274 2.70 835 2962 3409 7298 147 288 0.80 402 3462 4029 7934

PAGE 123

123 Table A 4. Leaf, stem, pod, and total dry weight (DW) and leaf area index (LAI) mean values ( n = 4) vs. days after planting (DAP) and day of year (DoY) for peanut cultivars Carver and York grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2009. Cultivar Fungicide DAP DoY LAI Leaf DW Stem DW Pod DW Total DW Carver NF 35 182 1.81 661 768 0 1429 49 196 2.50 1518 1660 12 3191 63 210 4.28 2001 3250 651 5902 77 224 5.44 2602 4504 1979 9096 90 237 4.93 2352 4221 3487 1007 5 103 250 1.72 1020 3868 4836 9725 118 265 0.17 102 3057 3738 6897 Fung 35 182 1.81 661 768 0 1429 49 196 2.50 1518 1660 12 3191 63 210 4.02 1885 2842 714 5442 77 224 4.38 2096 3406 1926 7447 90 237 4.02 2219 4353 3162 9750 103 250 2.79 1839 3665 4800 10304 118 265 0.98 756 3433 4572 8761 125 272 0.58 407 3655 4691 8754 York NF 35 182 1.20 417 383 0 800 49 196 2.03 1016 1079 2 2097 63 210 3.53 1551 2086 281 3918 77 224 4.49 1950 2734 642 5382 90 237 6.08 2672 3898 1545 8165 103 250 3.84 1919 4702 2463 9084 118 265 1.21 730 3900 2761 7391 131 278 0.78 491 3645 3598 7734 142 289 0.56 301 3632 4420 8352 Fung 35 182 1.20 417 383 0 800 49 196 2.03 1016 1079 2 2097 63 210 3.64 1597 2270 162 4030 77 224 3.99 1696 2521 733 5017 90 237 5.88 2493 4384 1784 8701 103 250 4.60 2276 4142 2597 9015 118 265 2.17 1315 3818 3489 8622 131 278 1.83 1181 3804 3848 8832 142 289 1.16 673 4170 4991 9834

PAGE 124

124 Table A 5. Mid day total canopy photosynthesis (TCP) mean values ( n = 2) vs. days after planting (DAP) and day of year (DoY) for peanut cultivars Carver and York grown under fungicidesprayed (Fung) and nonsprayed (NF) conditions during 2008 and 2009. 2008 2009 Cultivar Fungicide DAP DoY TCP DAP DoY TCP Carver NF 35 176 0.93 44 191 1.29 49 190 1.29 56 203 1.76 64 205 1.93 64 211 1.64 78 219 1.66 78 225 1.78 98 239 1.46 89 236 1.35 107 248 1.06 103 250 0.46 116 257 0.73 117 264 0.16 128 269 0.05 Fung 35 176 0.93 44 191 1.29 49 190 1.64 56 203 1.56 64 205 1.72 64 211 1.6 78 219 1.59 78 225 1.7 98 239 1.3 89 236 1.48 107 248 1.11 103 250 1.12 116 257 0.92 117 264 0.46 130 271 0.26 York NF 35 176 0.69 44 191 0.91 49 190 1.06 56 203 1.29 64 205 1.66 64 211 1.53 78 219 1.87 78 225 1.85 98 239 1.8 89 236 1.67 107 248 1.35 103 250 1.29 116 257 1 .07 117 264 0.74 130 271 0.38 132 279 0.5 142 283 0.2 142 289 0.17 147 288 0.2 Fung 35 176 0.69 44 191 0.91 49 190 1.23 56 203 1.19 64 205 1.65 64 211 1.47 78 219 1.77 78 225 1.91 98 239 1.6 89 236 1.62 107 248 1.42 103 250 1.53 116 257 1.32 117 264 1 130 271 0.65 132 279 0.85 142 283 0.34 142 289 0.35 147 288 0.25

PAGE 125

125 Figure A 1. Relationship between missing nodes on the mainstem and defoliation (based on leaf weight loss from peak weight) for the two peanut cultivars Carver and York

PAGE 126

126 APPENDIX B LEAF SPOT RATING SCALES Table B 1. Florida 1 10 rating scale (based on Chiteka et al., 1988). Rating Description 1 No disease 2 Very few lesions (none o n upper canopy) 3 Few lesions (very few on upper canopy) 4 Some lesions with more on upper canopy and slight defoliation ( 5 Lesions noticeable even on upper canopy with noticeable defoliation ( 6 Lesions numerous and very evident on upper canopy with significant defoliation ( 7 Lesions numerous o n upper canopy with much defoliation ( 8 Upper canopy covered with lesions with high defoliation ( 9 Very few leaves remaining and those covered with lesions (some plants completely defoliated) 10 Plants completely defoliated and killed by lea f spot

PAGE 127

127 Figure B 1. ICRISAT diagrammatic scale to estimate percent leaflet necrosis (from Subrahmanyam et al., 1995).

PAGE 128

128 Figure B 2 Determination of percent necrotic leaf area using ASSESS ver 2.0 image analysis software. (A) scanned leaf sh owing necrotic spots, and (B) nonnecrotic leaf area selected using the software. Percent necrotic area was determined by subtracting nonnecrotic area from 100.

PAGE 129

129 LIST OF REFERENCES Abdou, Y. A M., W.C. Gregory, and W.E. Cooper. 1974. Sources and nature o f resistance to Cercospora arachidicola Hori and Cercosporidium personatum (Beack. And Curtis) Deighton in Arachis species. Peanut Sci. 1:611. Adomou, M., P.V.V. Prasad, K.J. Boote, and J. Detongnon. 2005. Disease assessment methods and their use in simul ating growth and yield of peanut crops affected by leafspot disease. Ann. Appl. Biol. 146:469479. Anderson, W.F., C.C. Holbrook, and T.B. Brenneman. 1993. Resistance to Cercosporidium personatum within peanut germplasm. Peanut Sci. 20:5357. Aquino, V.M., F.M. Shokes, D.W. Gorbet, and F.W. Nutter. 1995. Late leaf spot progression on peanut as affected by components of partial resistance. Plant Dis. 79:7478. Bancal, MarieOdile, C. Robert, and B. Ney. 2007. Modelling wheat growth and yield losses from late epidemics of foliar diseases using loss of green leaf area per layer and preanthesis res erves. Ann. Bot. (London). 100: 777 789. Bassanezi, R.B., L. Amorim, A. Bergamin Filho, and R.D. Berger. 2002. Gas exchange and emission of chlorophyll fluorescence during the monocycle of rust, angular leaf spot and anthracnose on bean leaves as a function of their trophic cha racterization. J. Phytopathol. 150:37 47. Bassanezi, R.B., L. Amorim, A. Bergamin Filho, B. Hau, and R.D. Berger. 2001. Accounting for photosynthetic efficiency of bean leaves with rust, angular leaf spot and anthracnose to ass ess crop damage. Plant Pathol. 50:443 452. Bastiaans, L. 1991. Ratio between virtual and visual lesion size as a measure to describe reduction in leaf photosynthesis of rice due to leaf blast. Phytopathology 81:611615. Bastiaans, L. 1993. Effect of leaf blast on photosynthesis of rice. 1. Leaf photosynthesis. Neth. J. Pl. Path. 99:197 203. Batchelor, W.D., J.W. Jones, K.J. Boote, and H.O. Pinnschmidt. 1993. Extending the use of crop models to study pest damage. Trans. ASAE 36:551558. Bergamin Filho, A., S.M.T.P.G. Carneiro, C.V. Godoy, L. Amorim, R.D. Berger, and B. Hau. 1997. Angular leaf spot of Phaseolus beans: Relationship between disease, healthy leaf area, and yield. P hytopathology 87:506515. Boote, K.J. 1999. Concepts of calibrating crop growth models. p. 179200. In G. Hoogenboom et al. (ed.) DSSAT Version 3. A decision support system for agrotechnology transfer. Vol. 4. Univ. of Hawaii, Honolulu.

PAGE 130

130 Boote, K.J., W.D. B atchelor, J.W. Jones, H. Pinnschmidt, and G. Bourgeois. 1993. Pest damage relations at the field level. p. 277296. In F.W.T. Penning de Vries et al. (ed.) Systems approaches for agricultural development. Kluwer Acad. Publ., Dordrecht, the Netherlands. Boo te, K.J., J.M. Bennett, and J.W. Jones. 1983 b Canopy Photosynthesis: Apparent, Net, or Gross? In Advances in Photosynthesis Research. Proc. 6th Int. Congr. on Photosynthesis, Vrije Univ., Brussels, Belgium. 16 Aug. 1983. Dr. W. Junk Publ., The Hauge. Boo te, K.J., J.W. Jones, and J.M. Bennett. 1985. Factors influencing crop canopy CO2 assimilation of soybean. p 780785. In R. Shibles (ed.) Proc. World Soybean Research Conference III, 1217 August 1984, Ames, IA. Westview Press, Boulder, CO. Boote, K.J., J .W. Jones, and G. Hoogenboom. 1998a. Simulation of crop growth: CROPGRO model. p. 651692. In R.M. Peart and R.B. Curry (ed.) Agricultural Systems Modeling and Simulation. Marcel Dekker, New York. Boote, K.J., J.W. Jones, G. Hoogenboom, and N.B. Pickering. 1998b. The CROPGRO model for grain legumes. p. 99128. In G.Y. Tsuji et al. (ed.) Understanding Options for Agricultural Production. Kluwer Academic Publishers, Dordrecht The Netherlands. Boote, K. J., J. W. Jones, J. W. Mishoe, and R. D. Berger. 1983 a C oupling pests of crop growth simulators to predict yield reductions Phytopathology 73: 15811587. Boote, K. J., J. W. Jones, G. H. Smerage, C. S. Barfield, and R. D. Berger. 1980. Photosynthesis of peanut canopies as affected by leafspot and artificial de foliation. Agron. J. 72:247252. Bourgeois, G., and K. J. Boote. 1992. Leaflet and canopy photosynthesis of peanut affected by late leaf spot. Agron. J. 84:359366. Bourgeois, G., K.J. Boote, and R.D. Berger. 1991. Growth, development, yield, and seed qual ity of Florunner peanut affected by late leafspot. Peanut Sci. 18:137143. Branch, W.D. 2002. Registration of Georgia01R peanut. Crop Sci. 42:17501751. Branch, W.D. 2006. Registration of Georgia05E peanut. Crop Sci. 46:2305. Branch, W. D. and T. Bre nneman. 2008. Registration of Georgia07W Peanut J. Plant Registrations 2:8891. Butler, D.R., K.D.R. Wadia, and R.K. Reddy. 1995. Effects of humidity, leaf wetness, temperature and light on cinidial production by Phaeoisariopsis personata on groundnut. Plant Pathol. 44:662674.

PAGE 131

131 Cantonwine, E.G., A.K. Culbreath, C.C. Holbrook, and D.W. Gorbet. 2008. Disease progress of early leaf spot and components of resistance to Cercospora arachidicola and Cercosporidium personatum in runner type peanut cultivars. Pe anut Sci. 35:1 10. Cantonwine, E.G., A.K. Culbreath, K.L. Stevenson, R.C. Kemerait, Jr., T.B. Brenneman, N.B. Smith, and B.G. Mullinix, Jr. 2006. Integrated disease management of leaf spot and spott ed wilt of peanut. Plant Dis. 90:493500. Chiteka, Z.A., D .W. Gorbet, F.M. Shokes, T.A. Kucharek, and D.A. Knauft. 1988. Components of resistance to late leaf spot in peanut. I. Levels of variability implications for selection. Peanut Sci. 15:2530. Cook M. 1981. Susceptibility of peanut leaves to Cercosporidi um personatum Phytopathology 71:787 791. Duncan, W.G., D.E. McCloud, R.L. McGraw, and K.J. Boote. 1978. Physiological aspects of peanut yield improvement. Crop Sci. 18:10151020. Dwivedi, S.L., S. Pande, J.N. Rao, and S.N. Nigam. 2002. Components of resis tance to late leaf spot and rust among interspecific derivatives and their significance in a foliar disease resistance breeding in groundnut ( Arachis hypogaea L.). Euphytica 125:8188. Erickson, J.E., G.R. Stanosz, and E.L. Kruger. 2003. Photosynthetic consequences of Marssonina leaf spot differ between two poplar hybrids. New Phytol. 161:577583. FAO. 2011 FAOSTAT. Available at http://faostat.fao.org/default.aspx (verified 6 June 201 1 ) FAO, Rome, Italy. FAWN. 2011. Available at http://fawn.ifas.ufl.edu/ (verified 6 June, 2011). University of Florida, Gainesville, FL. Farquhar, G.D., S. von caemmerer, and J.A. Berry. 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149:7890. Gauch Jr., H.G., and G.B. Chase. 1974. Fitting the Gaussian curve to ecological data. Ecology 55:13771381. Gorbet, D.W. 2006. Registration of Carver peanut. Crop Sci. 46:27132714. Gorbet D.W. 200 7 Registration of Hull Peanut J. Plant Registrations 1:125126. Gorbet D. W. A. J. Norden, F.M. Shokes and D. A. Knauft. 1987. Registration of South ern Runner peanut. Crop Sci. 27:817. Gorbet D. W., and F.M. Shokes F. 2002a Registratio n of C 99R peanut. Crop Sci. 42:2207.

PAGE 132

132 Gorbet D. W., and F.M. Shokes F. 2002 b. Registration of Florida MDR 98 peanut. Crop Sci. 42:22072208. Gorbet, D.W. and Tillman, B.L. 2008. Registration of DP 1 peanut. J. Pl ant Registration s 2:200204. Gorbet, D.W., and B .L. Tillman. 2011. Registration of York peanut. J. Plant Registrations 5:16. Gregory, M.P., and W.C. Gregory. 1979. Exotic germplasm of Arachis L. interspecific hybrids. J. Hered. 70:185193. Guinn, G. and D.L. Brummett. 1993. Leaf age, decline in photosynthesis, and changes in abscisic acid, indole 3 acetic acid, and cytokinin in cotton leaves. Field Crops Res. 32:269275. Holbrook C. C. and A.K. Culbreath A. 2007. Registration of Tifrunner Peanut J. Plant Registrations 1:124. Holbrook, C.C., and A. K. Culbreath. 2008. Registration of Georganic peanut. J. Plant Registrations 2:17. Holbrook, C.C., P. Timper, A.K. Culbreath, and C.K. Kvien. 2008. Registration o f Tifguard peanut. J. Plant Registrations 2:9294. Hoogenboom, G., J.W. Jones, P.W. Wilkens, C.H. Porter, K.J. Boote, L.A. Hunt, U. Singh, J.L. Lizaso, J.W. White, O. Uryasev, F.S. Royce, R. Ogoshi, A.J. Gijsman, and G.Y. Tsuji. 2009. Decision Support System for Agrotechnology Transfer (DSSAT) Version 4.5 [CD ROM] University of Hawaii, Honolulu Hawaii. Inskeep, W.P., and P.R. Bloom. 1985. Extinction coefficients of chlorophyll a and b in N,Ndimethylformamide and 80% acetone. Plant Physiol. 77:483485. Jackson, L.F. 1981 Distribution and severity of peanut leafspot in Florida. Phytopathology 71:324 328. Jackson, C.R., and D.K. Bell. 1969. Diseases of peanut (g roundnut) caused by fungi. University of Georgia. Coll. Agric. Exp. Stn. Res. Bull. 56:1 5. Jacob i, J.C., and P.A. Backman. 1995. AU Pnuts Advisory II: Modification of the rulebased leaf spot advisory system for a partially resistant peanut cultivar. Plant Dis. 79:672676. Jacobi, J.C., P.A. Backman, D.P. Davis, and P.M. Brannen. 1995. AU Pnuts Advisory I: Deveolopment of a rule based system for scheduling peanut leaf spot fungicide applic ation. Plant Dis. 79:666671. Jenkins W. 1938. Two fungi causing leafspot of peanuts. J. Agric. Res. 56:317 332

PAGE 133

133 Jesus Jr., W.C., F.X.R. do Vale, R.R. Coelho, B. Hau, L. Zambolim, L.C. Costa, and A. Bergamin Filho. 2001. Effects of angular leaf spot and rust on yield loss of Phaseolus vulgaris Phytopathology 91:10451053. Jones, J.W., G. Hoogenboom, C.H. Porter, K.J. Boote, W.D. Batchelor, L.A. Hunt, P.W. Wilkens, U. Singh, A.J. Gijsman, and J.T. Ritchie. 2003. The DSSAT cropping systems model. Europ. J. A gron. 18:235265. Kemerait, B., T. Brenneman, and A. Culbreath. 2010. Peanut disease update. p 5780. In J. P. Beasley, Jr. (ed.) 2010 Peanut Update. University of Georgia Ext., Georgia. Knauft, D.A., and D.W. Gorbet. 1990. Variability in growth character istics and leafspot resistance parameters of peanut lines. Crop Sci. 30:169175. Knauft D.A., A.J. Norden, D.W. Gorbet and F.G. Martin. 1987. Stability of market quality factors in peanut ( Arachis hypogaea L.). Soil and Crop Science Society of Florida Pr oceedings 46:7274. Krapovickas, A., and W.C. Gregory. 1994. Taxonomy of the genus Arachis (Leguminosae). Bonplandia 8:1186. Kucharek, T. 2005. Extension plant pathology r eport no. 12, Disease control program for peanuts. Plant Pathology department. University of Florida/IFAS, Gainesville, FL. Kumudini, S., C.V. Godoy, B. Kennedy, E. Prior, J. Omielau, H.R. Boerma, and D. Hershman. 2010. Role of host plant resistance development stage on leaf photosynthetic competence of soybean rust infected leaves. Crop Sci. 50:25332542. Kumudini, S., E. Prior, J. Omielan, and M. Tollenaar. 2008. Impact of Phakopsora pachyrhizi infection on soybean leaf photosynthesis and radiation absorption. Crop Sci. 48: 23432350. Leal Bertioli, S.C. D.M., M.C.D. Farias, P.I.T. Silva P.M. Guimaraes, A.C.M. Brasileiro, D.J. Bertioli, and A.C.G. D. Araujo. 2010. Ultrastructure of the initial interaction of Puccinia arachidis and Cercosporidium personatum with leaves of Arachis hypogaea and Arachis stenosperma. J Phytopathol. 158:79279 6. Lin, L., A.S. Hedayat, B. Sinha, and M. Yang. 2002. Statistical methods in assessing agreement: models, issues and tools. J. Amer. Stat. Assoc. 97:257270. Long, S.P., and C.J. Bernacchi. 2003. Gas exchange measurements, what can they tell us about the underlying limitations to photosynthesis? Procedures and sources of error. J. Expt. Bot. 54:23932401.

PAGE 134

134 Lopes, D.B., and R.D. Berger. 2001. The effects of rust and anthracnose on the photosynthetic competence of disease bean leaves. Phytopathology 91:21222 0. McDonald, D., P. Subrahmanyam, R.W. Gibbons, and D.H. Smith. 1985. Early and late leaf spots of groundnut. Info. Bull. no. 21. International Crops Research Institute for the Semi Arid Tropics. Patancheru, Andhra Pradesh, India. Mims, C.W., E.S. Luttrell and S.C. Alderman. 1988. Ultrastructure of haustorium of the peanut late leaf spot fungus Cercosporidium personatum Can. J. Bot. 67:1198 1202. Monfort, W.S., A.K. Culbreath, K.L. Stevenson, T.B. Brenneman, D.W. Gorbet, and S.C. Phatak. 2004. Effects of reduced tillage, resistant cultivars, and reduced fungicide inputs on progress of early leaf spot of peanut ( Arachis hypogaea). Plant Dis. 88:858864. Moriondo, M., S. Orlandini, A. Giuntoli, and M. Bindi. 2005. The effect of downy and powdery mildew on g rapevine ( Vitis vinifera L.) leaf gas exchange. J. Phytopathology 153: 350357. Morton, B.R. 2007. Poor field emergence of latematuring peanut cultivars ( Arachis hypogaea L.) derived from PI 203396.Dissertation. University of Florida, Gainesville, FL. Mos sler, M. and M.J. Aerts. 2007. Florida crop/pest management profiles: peanuts. CIR 1260. Florida Coop. Ext. Serv., University of Florida, Gainesville, FL. Naab, J.B., P. Singh, K.J. Boote, J.W. Jones, and K.O. Marfo. 2004. Using the CROPGRO Peanut model to quantify yield gaps of peanut in the Guinean savanna zone of Ghana. Agron. J. 96:12311242. Nogues, S., L. Cotxarrera, L. Alegre, and M.I. Trillas. 2002. Limitations to photosynthesis in tomato leaves induced by Fusarium wilt. New Phytol. 154:461470. Nut ter Jr., F.W., and F.M. Shokes. 1995. Management of foliar diseases caused by fungi. p. 6574. In H.A. Melouk and F.M. Shokes (eds.) Peanut Health Management. American Phytopathological Society, St. Paul, MN. Perfect, S.E., and J.R. Green. 2001. Infection structures of biotrophic and hemibiotrophic plant pathogens. Molecular Pl. Pathology 2:101108 Phakamas, N., A. Patanothai, S. Jogloy, K. Pannangpetch, and H. Hoogenboom. 2008. Physiological determinants of pod yield of peanut lines. Crop Sci. 48:2351236 0. Pinnachmidt, H.O., W.D. Batchedlor, and P.S. Teng. 1995. Simulations of multiple species p est damage on rice. Agric. Syst. 48:193222.

PAGE 135

135 Pinnachmidt, H.O., Y. Luo, and P.S. Teng. 1994. Methodology for quantifying rice yield effects of blast. p 318408. In R.S. Ziegler et al (ed.) Rice blast diseases. CAB International, Wallingford, Oxon, UK. Pixley, K.V., K.J. Boote, F.M. Shokes, and D.W. Gorbet. 1990a. Growth and partitioning characteristics of four peanut genotypes differing in resistance to late leafspot. Crop Sci. 30:796804. Pixley, K.V., K.J. Boote, F.M. Shokes, and D.W. Gorbet. 1990b. Disease progression and leaf area dynamics of four peanut genotypes differing in resistance to late leafspot. Crop Sci. 30:789796. R Development Core Team. 2008. R: A language and environment for statistical computing. R foundation for statistical computing, Vienna, Austria. Robert, C., MarieOdile Bancal, C. Lannou, and B. Ney. 2006. Quantification of the effects of Septoria tritici blotch on wheat leaf gas exchange wi th respect to lesion age, leaf number, and leaf nitrogen status. J Exp. Bot. 57:225234. Robert, C., MarieOdile Bancal, B. Ney, and C. Lannou. 2005. Wheat leaf photosynthesis loss due to leaf rust, with respect to lesion development and leaf ni trogen sta tus. New Phytol. 165: 227 241. Robert, C., MarieOdile Bancal, P. Nicolas, C. Lannou, and B. Ney. 2004. Analysis and modellling of effects of leaf rust and Septoria tritici blotch on wheat growth. J Exp. Bot. 55:10791094. Roloff, I., H. Scherm, and M. W. Iersel. 2004. Photosynthesis of blueberry as affected by septoria leaf spot and abiotic leaf damage. Plant Dis. 88:397401. SAS Institute. 2009. The SAS system for Windows. Release 9.2. SAS Inst., Cary, NC. Shanner, G., and R.E. Finney. 1977. The effect of nitrogen fertilization on the expression of slow mildewing resistance in knox wheat. Phytopathology 67:10511056. Shew, B.B., M.K. Beute, and J.C. Wynne. 1988. Effect of temperature and relative humidity on expression of resistance to Cercosporidium personatum in peanut. Phytopathology 78:493 498. Shokes, F.M., and A.K. Culbreath. 1997. Early and late leaf spots. p. 1720. In N. Kokalis Burelle, D.M. Porter, R. Rodriguez Kabana, D.H. Smith, and P. subrahmanyam (eds.) Compendium of Peanut Diseases, 2nd Ed. APS Press, St. Paul, MN. Shokes F. M. D.W. Gorbet, and L. Jackson. 1983. Control of early and late leaf spot on two peanut cultivars. Pean ut Sci. 10:1721.

PAGE 136

136 Shokes, F.M., D.W. Gorbet, and G.E. Sanden. 1982. Effect of planting date and date of spray initiat ion on control of peanut leaf spots in Florida. Plant Dis. 66: 574 575. Smith, D.H., and R.H. Littrell. 1980. Management of peanut foliar diseases with fungicides. Plant Dis. 64:356 361 Smith, D.H., G.D.C. Paucer, and F.M. Shokes. 1992. Cercosporidium and cercospora leaf spots of peanut (Groundnut). p. 285304. In U.S. Singh, A.N. Mukhopadhyay, J. Kumar, and H.S. Chaube (eds.) Plant Diseases of International Importance. Prentice Hall Inc., Englewood Cliffs, New Jersey. Shtienberg D. 1992. Effect of foliar disease on gas exchange process: a comparative study. Phytopathology 82:760765. Stalker, H.T. 1997. Peanut ( Arachis hypogaea L.). Field Crops Res. 53:205217. Subrahmanyam, P., D. McDonald, F. Waliyar, L.J. Reddy, S.N. Nigam, R.W. Gibbons, V. Ramanatha R ao, A.K. Singh, S. Pande, P.M. Reddy, and P.V. Subba Rao. 1995. Screening methods and sources of resistance to rust and late leaf spot of groundnut. ICRISAT Information Bulletin No. 47. ICRISAT, Patancheru, Andhra Pradesh, India. Teng, P.S., W.D. Batchelor H.O. Pinnschmidt, and G.G. Wilkerson. 1998. Simulation of pest effects on crops using coupled pest crop models: the potential for decision support. p 221266. In G.Y. Tsuji et al (ed.) Understanding options for agricultural production. Kluwer Academic Publishers, Dordrecht The Netherlands. Tillman, B., D. Gorbet, M. Gomillion, J. McKinney, G. Person, and B. Thomas. 2008. Peanut variety performance in Florida 20042007. SS AGR 311. Florida Coop. Ext. Serv. Univ. of Florida, Gainesville, FL. Tillman, B.L., and H.T. Stalker. 2009. Peanut. p 287317. In J. Vollmann and I. Rajcan (ed.) Oil Crops, Handbook of plant breeding 4. Springer Science. Timsina, J., K.J. Boote, and S. Duffield. 2007. Evaluating the CROPGRO soybean model for predicting impacts of insect defoliation and depodding. Agron. J. 99:148157. USDA NASS. 2011. Crop Acreage and Value: Peanut. Available at http://www.nass.usda.gov (verified 6 June 2011). Washington, D.C. Williams, J.H., and K.J. Boote. 1995. Physiology and modeling predicting the unpredictable legume. p 301353. In H.E. Pattee and H.T. Stalker (ed.) Advances in peanut science. Am. Peanut Res. And Edu. Soc., Stillwater, Ok. Williams, E.J., and J.S. Drexler. 1981. A nondestructive method for determining peanut pod maturity. Peanut Sci. 8:134141.

PAGE 137

137 Willmott, C.J. 1981. On the validation of models. Physical Geography 2:184194 Willmott, C.J. 1982. Some comments on the evaluation of model performance. Bull. Am. Meteorol. Soc. 63:13091313. Woodw ard, J.E., T.B. Brenneman, R.C. Kemerait, Jr., A.K. Culbreath, and N.B. Smith. 2010. Management of peanut diseases with reduced input fungicide programs in fields with varying levels of disease risk. Crop Prot. 29:222229. Woodward, J.E., T.B. Brenneman, R .C. Kemerait, Jr., N.B. Smith, A.K. Culbreath, and K.L. Stevenson. 2008. Use of resistant cultivars and reduced fungicide programs to manage peanut disease in irrigated and nonirrigated fields. Plant Dis. 92:896902. Wright, D.L., B. Tillman, E. Jowers, J. Marois, J.A. Ferrell, T. Katsvairo, and E.B. Whitty. 2006. Management and cultural practices for peanuts. SS AGR 74. Florida Coop. Ext. Serv. Univ. of Florida, Gainesville, FL. Wynne, J.C., M.K. Beute, and S.N. Nigam. 1991. Breeding for disease resistanc e in peanut. Annu. Rev. Phytopathol. 29:279303. Yin, X., J. Goudriaan, E.A. Lantinga, J. Vos, and H.J. Spiertz. 2003. A flexible sigmoid function of determinate growth. Ann. Bot. (London) 91:361371. Zhang, S., S. Lu, X. Xu, H. Korpelainen, and C. Li. 2009. Changes in antioxidant enzyme activities and isozyme profiles in leaves of male and female Populus cathayana infected with Melampsora larici populina. Tree Physiol. 30:116128.

PAGE 138

138 BIOGRAPHICAL SKETCH Maninder pal Singh was born in 1983 in Punjab, India. He is the younger son of Joginder Kaur and Kehar Singh. After attending village school at the primary level, he went to a boarding school, Jawahar Navodaya Vidyalaya, for high school. He earned a Bachelor of Science with Honors in Agriculture from Guru Nan ak Dev University, In dia and a Master of Science in agronomy from Punjab Agricultural University, India. His masters thesis was entitled Effect of plating methods and irrigation schedules on growth and yield of hybrid Bt cotton. In 2007, he joined the U niversity of Florida for his doctorate degree.