HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Suriharn, B. Articles by Hoogenboom, G. Search for Related Content PubMed Articles by Suriharn, B. Articles by Hoogenboom, G. Agricola Articles by Suriharn, B. Articles by Hoogenboom, G. Related Collections Other Legumes Crop Models

CROP BREEDING & GENETICS

Determination of Cultivar Coefficients of Peanut Lines for Breeding Applications of the CSM-CROPGRO-Peanut Model

^a Dep. of Agronomy, Faculty of Agriculture, Khon Kaen Univ., Khon Kaen 40002, Thailand
^b Dep. of Biological and Agricultural Engineering, the Univ. of Georgia, Griffin, GA 30223-1797, USA

^* Corresponding author (aran{at}kku.ac.th).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Breeding applications of a crop simulation model require cultivar coefficients for a large number of breeding lines, making the recommendation to determine these coefficients from field experiments conducted over several environments impractical. The objective of this study was to confirm our earlier findings that experimental data from only two seasons are sufficient to determine the cultivar coefficients of peanut lines for breeding applications of the CSM-CROPGRO Peanut model. Seventeen peanut lines varying in yield level, seed type and maturity duration were selected for this study and were grown during the 2002 rainy and 2003 dry seasons. Data were collected on plant development and growth characteristics and were then used for model calibration to determine the cultivar coefficients of the individual peanut lines. A separate experiment was conducted in the dry season of 2004 for model evaluation. The calibration resulted in cultivar coefficients that provided simulated values of the various traits that were close to their corresponding observed values. Similar results were obtained for model evaluation with an independent data set. Differences among peanut lines were also expressed in the derived cultivar coefficients. These results confirmed our earlier finding that the cultivar coefficients of peanut lines derived from detailed experimental data for two seasons are sufficiently accurate for breeding applications of the CSM-CROPGRO-Peanut model.

Abbreviations: CV, coefficient of variation • CSDL, critical daylength for photoperiod • EMFL, emergence to flowering • FLLF, first flowering to end of leaf expansion • FLSD, first flowering to first seed • FLSH, first flowering to first pod • LAI, leaf area index • LFMAX, maximum leaf photosynthetic rate • PD, photothermal days • PODUR, time required for a cultivar to reach final pod load • PPSEN, sensitivity to photoperiod • RMSE, root mean square error • SFDUR, seed filling duration for an individual pod cohort • SDPM, first seed to physiological maturity • SIZELF, maximum size of a full leaf • SLA, specific leaf area • SLAVR, specific leaf area • SLPF, soil fertility factor • SDPDV, averaged seed number per pod • WP, wettable powder • WTPSD, individual seed size • XFRT, maximum fraction of daily growth that is allocated to seed and shell.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
CROP MODELS provide a dynamic simulation of crop growth and development through numeric integration of the underlying physiological processes with the aid of computers (Sinclair and Seligman, 1996). These models can be useful tools for the synthesis of research understanding on the interactions of genetics, physiology, and environmental factors that influence crop growth and yield (de Wit, 1982; Whisler et al., 1986; Uehara, 1998). In addition, they can be used to support the decision making process for cropping system management and agricultural policy (Boote et al., 1996). Crop simulation models also have the potential to contribute to the crop improvement process (White, 1998; Matthews and Stephens, 2002). These include assisting with multi-location evaluation (Aggarwal et al., 1995; Banterng et al., 2003), understanding of the genotype by environment (G x E) interaction (Hammer et al., 1996; Chapman et al., 2002), identification and evaluation of desirable traits or combination of traits leading to the design of a crop ideotype for a specific environment (Boote and Tollenaar, 1994; Aggarwal et al., 1995; Bastiaans et al., 1997; Boote et al., 2001) and evaluation of breeding strategies for drought tolerance (Spitters and Schapendonk, 1990).

The CSM-CROPGRO-Peanut model is one of the crop simulation models that encompass the Decision Support System for Agrotechnology Transfer (DSSAT; Tsuji et al., 1994; Hoogenboom et al., 1999; 2004; Jones et al., 2003). This model is process-oriented and designed to simulate growth and development on a daily basis, using carbon, nitrogen, and water balances. The model requires inputs that include environmental conditions, management practices, and characteristics of crop genotype or cultivar-specific genetic coefficients (Boote et al., 1998). Before the application of a crop simulation model, such as the CSM-CROPGRO-Peanut model, it is necessary to first determine the cultivar coefficients and evaluate model performance if the cultivars are new breeding lines or local cultivars that have not been used previously with the model. The experiments to determine cultivar coefficients should be conducted under optimum conditions and be free from drought and other environmental and biotic stresses. Data to be collected include the duration of phenological development stages and dry matter accumulation of different plant parts, leaf area index (LAI), and specific leaf area (SLA) at different growth stages (IBSNAT, 1988). In addition, it is recommended that these cultivar coefficient determination experiments are conducted across several planting dates at the same location or across multiple locations for the same planting date (Hoogenboom et al., 1999). This makes the determination of cultivar coefficients a laborious and time-consuming process. The recommended procedure, therefore, is not very practical for plant breeding programs that have large numbers of breeding lines. Banterng et al. (2004) reported a study in which data from detailed experiments grown in two seasons were found to be sufficient to determine cultivar coefficients of peanut lines for breeding applications of the CSM-CROPGRO-Peanut model. However, in their study of genetic coefficients of 26 peanut lines determined by model calibration, they evaluated cultivar coefficients of only four lines with an independent data set. Because Banterng et al. (2004) only evaluated four lines with independent data, it is important to repeat this type of study with a larger set of cultivars calibrated with in-season data and then evaluated with data from an independent season. Therefore, more empirical evidence is needed before a conclusion can be obtained with confidence. The overall goal of this study was to confirm the findings of Banterng et al. (2004) with a different set of peanut lines. The objective was to evaluate whether the cultivar coefficients of these peanut lines determined from plant growth and development data for only two growing seasons are sufficiently accurate for breeding applications of the CSM-CROPGRO-Peanut model.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Determination of Cultivar Coefficients
Field Experiments
Thirteen peanut lines and four check cultivars were used in this study. These lines were selected from preliminary yield trials conducted by the peanut breeding program of Khon Kaen University, Thailand to provide a range of yield levels, seed types, and maturity durations (Table 1). To obtain the required data for determination of the cultivar coefficients, two field experiments were conducted during the rainy season of 2002 and the dry season of 2003 at the Field Crop Research Station of Khon Kaen University located in Khon Kaen, Thailand (16°28' N lat; 102°48' E long, 200 meters above sea level). The rainy season experiment was sown on 8 June 2002, while the dry season experiment was sown on 14 Dec. 2002. A randomized complete block design with four replications was used. Each plot consisted of 12 rows, 7 m long, with a spacing of 50 cm between rows and 20 cm between plants.

The experimental data collection followed the procedures described in IBSNAT (1988) and Hoogenboom et al. (1999). To determine soil physical and chemical characteristics, soil samples were taken at two different locations in each field before planting at depths of 0 to 15, 15 to 30, 30 to 45, 45 to 60, 60 to 75, 75 to 90, and 90 to 105 cm. The soil samples were analyzed for physical properties, which included soil texture, bulk density, and soil moisture, as well as chemical properties of pH, organic matter, exchangeable K and P, and NO₃^– and NH₄⁺ concentrations. The weather data (daily maximum and minimum air temperatures, solar radiation and precipitation) for each growing season were obtained from the KKU Field Crops Research Station where the experiment was conducted.

For crop development, vegetative and reproductive growth stages were determined using the system developed for peanut by Boote (1982). The developmental stages observed included emergence (VE), four nodes on the main stem (V4), first flower (R1), first peg (R2), first pod (R3), fully expanded pod (R4), first seed (R5), full seed (R6), physiological maturity (R7), and harvest maturity (R8; 70–80% matured pods). Each stage was defined to have occurred if at least 50% of the plants in a plot had reached that stage. The dates of VE to R2 stages were observed daily by inspecting all plants in each plot for VE, V4, and R1 stages and all plants in 4 rows of each plot for the R2 stage. After R2, four plants in each plot were pulled for inspection every 3 d to determine if two or more plants had reached the next developmental stage. For growth analysis, five consecutive and bordered plants were harvested from each plot in a sampling date. Samplings for growth analysis were conducted at five developmental stages, including V4, R4, R6, R7 and R8. The individual samples were separated into leaf, stem, pod, and seed. A subsample of 60 leaflets was measured with a leaf area meter (Hayashi DenKoh AAC-400, Tokyo, Japan) to determine the leaf area. The samples were oven dried at 80°C for 48 h and weighed to determine dry matter. Specific leaf area (SLA) was computed by dividing the area of the leaf subsample with its corresponding dry weight. The leaf area index (LAI) was determined as the ratio of the specific leaf area (of the subsample) multiplied by the leaf mass of the harvested ground area (m²m^–2).

Model Calibration
Model calibration is the adjustment of input parameters of a model to provide an acceptable fit between the simulated and observed plant characters (Boote, 1999). The CSM-CROPGRO-Peanut model requires input data that consist of management practices, e.g., cultivars, spacing, plant population, fertilizer, and irrigation, and local environmental conditions, which include daily maximum and minimum air temperature, solar radiation, precipitation and soil surface and profile characteristics. Model calibration was first conducted for the soil parameters and then for the cultivar coefficients. For the calibration of the soil parameters, soil sample data were used to calculate soil parameters for the entire soil profile and for each soil layer with the soil data retrieval program of DSSAT (Tsuji et al., 1994). Parameters that were obtained for each soil layer were saturated soil water content, drained upper limit, and the lower limit of plant-extractable water. Soil surface parameters determined included reflectance or albedo, evaporation limit, drainage rate, runoff curve number, and mineralization factor. The value for the soil fertility factor (SLPF) was adjusted to obtain a good fit for the mean value of biomass and pod yield over all cultivars in both seasons.

The CSM-CROPGRO-Peanut model has 15 cultivar coefficients that define the growth and development characteristics or traits of a peanut cultivar (Table 2). These include the life cycle parameters, the vegetative growth parameters and the reproductive growth parameters. The life cycle parameters consist of the critical daylength for photoperiod (CSDL), the sensitivity to photoperiod (PPSEN), the number of photothermal days from emergence to flowering (EMFL), the number of photothermal days from first flowering to first pod (FLSH), the number of photothermal days from first flowering to first seed (FLSD), the number of photothermal days from first flowering to end of leaf expansion (FLLF) and the number of photothermal days from first seed to physiological maturity (SDPM) (life cycle phase). The vegetative growth parameters include the maximum leaf photosynthetic rate (LFMAX), specific leaf area (SLAVR), the maximum size of a full leaf (SIZELF) and the maximum fraction of daily growth that is allocated to seed and shell (XFRT) (vegetative traits). The reproductive growth parameters include individual seed size (WTPSD), seed filling duration for an individual pod cohort (SFDUR), averaged seed number per pod (SDPDV) and the photothermal time required for a cultivar to reach final pod load (PODUR) (reproductive traits) (Boote et al., 2003).

Because all peanut lines in this study are insensitive to photoperiod, the critical short daylength was set at 11.84 h, and the photoperiod sensitivity was set at 0.00, i.e., allowing for no photoperiod reaction. Calibration of phenology and growth traits was conducted by minimizing the error between the observed and simulated data. The accuracy of the procedure used to estimate the cultivar coefficients was determined by comparing the simulated values of development and growth characteristics with their corresponding observed values, and by the values of the coefficient of determination (r²), the root mean square error (RMSE) (Wallach and Goffinet, 1987) and the index of agreement (d) (Willmott, 1982). The values of RMSE and d indicate the degree of agreement between the predicted values with their corresponding observed values, and a low RMSE value and a d value approaching unity are desirable. The RMSE was computed using the following equation:

[1]

where n is number of observations, P_i is the predicted value for the ith measurement and O_i is the observed value for the ith measurement. The index of agreement was computed using the following equation:

[2]

where n = number of observations, P_i = predicted value for the ith measurement, O_i = observed value for the ith measurement, O– = the overall mean of observed values, P‘_i = P_i–O–, and O’_i = O_i–O–. The derived genetic coefficients of individual lines were compared to determine whether the model was sensitive enough to capture the differences among the tested peanut lines.

Model Evaluation
To obtain an independent data set for model evaluation, all the peanut lines under study were grown in another experiment during the dry season of 2004 at the KKU Field Crops Research Station. The experiment was planted on 14 Dec. 2003; the experimental design was a randomized complete block with four replications. The plot size was six rows, 6 m long, with a spacing of 50 cm between rows and 20 cm between plants. The experiment was also well managed to avoid stresses from water, nutrients, pests, and diseases as much as possible. Applications of fertilizers, fungicides, pesticides, herbicide, irrigation, and other management measures were the same as the experiments that were conducted to obtain data for model calibration. The experimental data were collected in the same manner as the data that were collected for model calibration. In this experiment, the data that were collected for crop development stages included time to emergence (VE), time to first flower (R1) and time to harvest maturity (R8). Growth analysis samples were collected at 25, 67, 81, 95, 106 d after planting and at harvest maturity. In each plot, five plants were harvested for each sample date and were separated into total above ground biomass, pod (after R4) and seed. Separation of stover into leaves and stem was not done in this experiment. These samples were then oven-dried at 80°C for 48 h and weighed to obtain dry weights of the individual plant parts.

To evaluate model performance, the cultivar coefficients of the individual peanut lines obtained from model calibration data (2002–2003) were used to simulate growth and development of the same lines for the evaluation experiment, e.g., the dry season of 2004 at the KKU Field Crops Research Station. The soil fertility factor (SLPF) value used in model simulation was the value determined from model calibration. Model evaluation was conducted by comparing the simulated values of development and growth characteristics of the individual peanut lines with their corresponding observed values, and by obtaining the values for r², RMSE and d.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Calibration of Cultivar Coefficients
The cultivar coefficients were calibrated using the data from the trials grown during the 2002 rainy and the 2003 dry seasons. The ability of the model to simulate developmental stages was assessed by comparing the simulated durations of five developmental stages with the corresponding observed values (Fig. 1 ). The model predicted the duration from planting to first flowering (R1) reasonably well for both seasons, with an RMSE value of 1.1 d and r² value of 0.98 (Fig. 1a). The number of simulated days from planting to first pod (R3) (Fig. 1b) and from planting to first seed (R5) (Fig. 1c) were also in good agreement with the corresponding observed values for both two seasons, with a RMSE of 3.5 and 3.8 d and a r² of 0.91 and 0.90 for R3 and R5, respectively. However, the prediction of these three developmental stages tended to be slightly underestimated for the dry season and slightly overestimated for the rainy season. This could be due to cool temperatures during the pre-flowering stage in the dry season that caused a delay in flower initiation and extended the number of days for the later developmental stages (data not shown). Therefore, the best fit for model calibration caused the predicted values to be in between the observed values of the two seasons. The disparities between observed and simulated values for days to first pod (R3) and days to first seed (R5) were also somewhat greater than for days to first flower (R1). This could be partially accounted for by the inaccuracy of the determination of observed values, as plants needed to be uprooted to be able to detect these stages of pod and seed development. With three day intervals of inspection, each with four plants inspected for a peanut line, greater variations of observed values would be anticipated for these two developmental stages. There might also be a possibility that the temperature parameterization of reproductive timing is wrong. In general, modeled development should have been faster during the rainy and warmer season and slower during the dry and cool season, implying that one or more of the cardinal temperatures of the model are wrong. The prediction of duration from planting to harvest maturity (R8), however, was very good when compared with the simulated values (Fig. 1d). This was shown by a low value for RMSE, e.g., 1.4 d, and a very high value of r², e.g., 0.98.

View larger version (31K): [in this window] [in a new window]   Figure 1. Simulated versus observed values for the number of days from planting to first flowering (a), to first pod (b), to first seed (c), and to harvest maturity (d) for 17 peanut lines/cultivars grown during the 2002 rainy and 2003 dry seasons.

View larger version (65K): [in this window] [in a new window]   Figure 2a. Simulated (lines) and observed (symbols) values for total biomass, and stem, leaf, pod, and seed mass for all entries grown during the 2002 rainy season (a) (see Table 1 for entry description).

View larger version (60K): [in this window] [in a new window]   Figure 3a. Simulated (lines) versus observed (symbols) values for specific leaf area (SLA) and leaf area index (LAI) for all entries grown during the 2002 rainy season (a) (see Table 1 for entry description).

Model Evaluation
The independent data set of the 2004 dry season was used to evaluate the performance of the model based on the cultivar coefficients determined from data from the previous two seasons. Good agreements were obtained between observed and simulated values for days to first flower (R1) and days to maturity (R8) as indicated by the high values of r², e.g., 0.72 for R1 and 0.97 for R8 (Fig. 4a and 4b) and low values for RMSE, e.g., 1.5 and 2.2 d for R1 and R8, respectively. However, prediction of these two development stages tended to be underestimated for this season. Prediction of pod yield also showed a reasonably good agreement between simulated and observed values, as indicated by the high value of r², e.g., 0.64 (Fig. 4d). However, prediction of final biomass was rather poor and tended to be underestimated as shown by the low value of r², e.g., 0.27 (Fig. 4c).

View larger version (24K): [in this window] [in a new window]   Figure 4. Simulated versus observed values for the number of days from planting to first flowering (a) and to harvest maturity (b), and for biomass (c), and pod yield (d) of 17 peanut lines/cultivars grown during the 2004 dry season.

View larger version (27K): [in this window] [in a new window]   Figure 5. Simulated (lines) versus observed (symbols) values for total biomass, pod, and seed mass for entry nos. 1 to 17 grown during the 2004 dry season (see Table 1 for entry description).

Variation in Cultivar Coefficients
To determine whether model calibration was sensitive enough to capture the differences among the test peanut lines, variations among lines for individual cultivar coefficients were evaluated based on the values of the coefficient of variation (CV) which was used as the indicator of the range of genetic variation. For the phenological coefficients, the greatest variation among lines was observed for the number of photothermal days from first seed to physiological maturity (SDPM), with a CV of 9.5% and a range from 48.5 to 64.0 photothermal days (PD). The other four phenological coefficients, i.e., the numbers of PDs for EMFL, for FLSH, for FLSD, and for FLLF, had smaller variations among lines. Their CV values varied from 5.0 to 7.1% (Table 5).

The cultivar coefficients associated with growth include LFMAX, SIZLF, SLA, WTPSD, SDPDV, SFDUR, and PODUR (Table 2). Variations among lines for these parameters ranged from 3.5 to 26.9% (Table 5). The highest variation was found for the number of PD to reach final pod load, with a CV of 26.9% and a range from 13.0 to 30.0 PD. SIZLF, WTPSD, and SFDUR also showed high variations, with CV values of 18.6, 19.0 and 11.0%, respectively. The remaining coefficients, i.e., LFMAX, SLAVR, XFRT, and SDPDV, had low variations among lines, with the CV ranging from 3.5 to 6.4%.

On average, the medium duration lines had the highest LFMAX while the short and long duration groups were similar for this cultivar coefficient. However, the highest value for SLAVR was found for the long duration group, followed by the short duration group and then the medium duration group. SIZLF was largest for the short duration group and smaller for the medium and long duration groups, respectively, while the opposite was found for the values for XFRT. The three groups did not differ in WTPSD and SDPDV (Table 5).

Entries no. 10 {[(Nc.17090 x B₁)–25 x China 97–2] F6–2–2}, no. 1 [(Luhua 11 x KK60–3) F6–15] and no. 11 [(China 97–2 x Singburi) F6–13–1] were the three highest yielding cultivars. These lines had moderately high values for LFMAX, XFRT, and SFDUR but a moderately low value for PODUR. A high LFMAX could be attributed to traits that relate to a high leaf photosynthesic rate, including a high percentage of leaf N, slower N mobilization, and better disease and nematode resistances (Mavromatis et al., 2001). Also a longer seed-filling duration is an essential trait for increasing the yield potential (Kropff et al., 1995; Boote et al., 2001; Aggarwal et al., 1997). These observations are in line with the finding of both Mavromatis et al. (2001) and Banterng et al. (2004) who reported that high-yielding peanut cultivars should have high or moderately high values for LFMAX, FLSH, and SFDUR. These results indicated that model calibration was sensitive enough to capture the differences among the test peanut lines.

The use of crop simulation models in assisting the multi-environment evaluation of crop breeding lines is of great interest to plant breeders as it will help overcome some of the difficulties in the operation of multi-location trials and improve breeding efficiency. This process consumes a lot of time and resources, and the scope of the test environments normally does not cover the entire range of environmental conditions for the different production areas in the country. The potential of such an application as a breeding tool has been demonstrated by our recent studies using the CSM-CROPGRO-Peanut model (Banterng et al., 2003, 2006). We found that the model could predict the relative mean performances of 12 advanced large-seeded peanut lines tested in 12 environments and of 14 advanced small-seeded peanut lines tested in three environments. The model correctly predicted all the six top-yielding lines in the large-seeded group and five out of seven top yielding lines in the small-seeded group identified by actual testing. The model could also predict the regression coefficients of mean yields of individual large-seeded lines against site mean yields quite well, and thus could be used for assessing yield stability of the test breeding lines. However, more empirical evidence is still needed to confirm these results. If there is sufficient scientific support for this application, the model could be used for simulating the yields of the test breeding lines for a wider range of environments than would be possible in actual testing. Yield stability of the test lines could be evaluated over a wide range of environments and the nature of genotype x environment interactions could also be investigated. This would certainly improve the efficiency and effectiveness of breeding line evaluation and selection.

The use of the simulation model in assisting the multi-environment evaluation of crop breeding lines would be more useful during the early stages of line evaluation as actual yield testing is normally done in only a few environments. Simulated values could be used in extending the range of the test environments, making line selection more accurate and effective. At this stage of line evaluation, however, the number of lines will be large and the amount of seeds available for each line would be limited. Although the results of this study as well as of Banterng et al. (2004) indicated that data from two seasons of the cultivar-coefficient determination experiment would be sufficient to determine the estimates of the required cultivar coefficients, the experiment is still laborious and requires a considerable amount of seeds. Thus, determination of cultivar coefficients would still be a major limitation to the use of crop simulation model in assisting breeding line evaluation at the early stages. More research is needed to reduce the data required for the determination of cultivar coefficients, particularly reducing the number of destructive samplings, to make such an application practically possible.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The cultivar coefficients of 17 peanut lines varying in yield level, seed type, and maturity duration determined from data of detailed field experiments conducted in two growing seasons provided simulated values of various development and growth parameters that were in good agreement with their corresponding observed values for almost all parameters. Model evaluation with an independent data set gave similar results for all the peanut lines. The determined cultivar coefficients also showed differences among lines indicating the sensitiveness of model calibration. The results, thus, confirmed the finding of Banterng et al. (2004) that the cultivar coefficients of peanut lines determined from experimental data from two growing seasons are sufficiently accurate for breeding applications of the CSM-CROPGRO-Peanut model.

	ACKNOWLEDGMENTS

 
Financial support for this study was provided by the Thailand Research Fund through the Royal Golden Jubilee Ph.D. Program (Grant No. PHD/0108/2544) and the Senior Research Scholar Project of Professor Dr. Aran Patanothai. Assistance was also received from the Peanut Project, Department of Agronomy, Faculty of Agriculture, Khon Kaen University, Khon Kaen, Thailand and by State and Federal funds allocated to Georgia Agricultural Experiment Station Hatch project GEO00895.

Received for publication January 28, 2006.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Aggarwal, P.K., R.B. Matthews, and M.J. Kropff. 1995. Opportunities for the application of systems approaches in plant breeding. p. 135–144. In P. K. Aggarwal et al. (ed.) SARP Research Proceedings. International Rice Research Institute, Los Banos, Philippines.
• Aggarwal, P.K., M.J. Kropff, K.G. Cassman, and H.F.M. Ten Berge. 1997. Simulating genotypic strategies for increasing rice yield potential in irrigated, tropical environments. Field Crops Res. 51:5–17.[CrossRef]
• Banterng, P., A. Patanothai, K. Pannangpetch, S. Jogloy, and G. Hoogenboom. 2003. Applicability of the CROPGRO-Peanut model in assisting multi-location evaluation of peanut breeding lines. Thai J. Agric. Sci. 37:407–418.
• Banterng, P., A. Patanothai, K. Pannangpetch, S. Jogloy, and G. Hoogenboom. 2004. Determination of genetic coefficients for peanut lines for breeding applications. Eur. J. Agron. 21:297–310.[CrossRef]
• Banterng, P., A. Patanothai, K. Pannangpetch, S. Jogloy, and G. Hoogenboom. 2006. Yield stability evaluation of peanut breeding lines: A comparison of an experimental versus a simulation approach. Field Crops Res. 96:168–175.[CrossRef]
• Bastiaans, L., M.J. Kropff, N. Kempuchetty, A. Rajan, and T.R. Migo. 1997. Can simulation models help design rice cultivars that are more competitive against weeds. Field Crops Res. 51:101–111.[CrossRef]
• Boote, K.J. 1982. Growth stages of peanut (Arachis hypogaea L.). Peanut Sci. 9:35–40.
• Boote, K.J., J.W. Jones, and N.B. Pickering. 1996. Potential uses and limitations of crop models. Agron. J. 88:704–716.[Abstract/Free Full Text]
• Boote, K.J. 1999. Concepts for calibrating crop growth models. p. 179–200. In G. Hoogenboom et al. (ed.) DSSAT v3. Vol. 4. Univ. of Hawaii, Honolulu, HI.
• Boote, K.J., M.J. Kropff, and P.S. Bindraban. 2001. Physiology and modeling of traits in crop plants: Implications for genetic improvement. Agric. Syst. 70:395–420.[CrossRef][ISI]
• Boote, K.J., and M. Tollenaar. 1994. Modeling genetic yield potential. p. 533–565. In K. J. Boote et al. (ed.) Physiology and determination of crop yield. ASA–CSSA–SSSA, Madison, WI.
• Boote, K.J., J.W. Jones, and G. Hoogenboom. 1998. Simulation of crop growth: CROPGRO Model. p. 651–692. In R.M. Peart and R.B. Curry (ed.) Agricultural systems modeling and simulation. Marcel Dekker, New York.
• Boote, K.J., J.W. Jones, W.D. Batchelor, E.D. Nafziger, and O. Myers. 2003. Genetic coefficients in the CROPGRO–Soybean model: Links to field performance and genomics. Agron. J. 95:32–51.[Abstract/Free Full Text]
• Chapman, S.C., G.L. Hammer, D.W. Podlich, and M. Cooper. 2002. Linking bio-physical and genetic models to integrate physiology, molecular biology and plant breeding. p. 167–188. In M. S. Kang (ed.) Quantitative genetics, genomics, and plant breeding. CABI Publishing, New York.
• de Wit, C.T. 1982. Simulation of living systems. p. 3–8. In F.W.T. Penning de Vries, and H.H. van Laar (ed.) Simulation of plant growth and crop production. Simulation Monographs, Pudoc, Wageningen, The Netherlands.
• Hammer, G.L., D. Butler, R.C. Muchow, and H. Meinke. 1996. Integrating physiological understanding and plant breeding via crop modeling and optimization. p. 419–441. In M. Cooper, and G.L. Hammer (ed.) Plant adaptation and crop improvement. CAB Int., ICRISAT, and IRRI, Wallingford, UK.
• Hoogenboom, G., P.W. Wilkens, and G.Y. Tsuji. 1999. DSSAT v3, Vol. 4. University of Hawaii, Honolulu, HI.
• Hoogenboom, G., J.W. Jones, P.W. Wilkens, C.H. Porter, W.D. Batchelor, L.A. Hunt, K.J. Boote, U. Singh, O. Uryasev, W.T. Bowen, A.J. Gijsman, A. du Toit, J.W. White, and G.Y. Tsuji. 2004. Decision Support System for Agrotechnology Transfer Version 4.0 [CD-ROM]. University of Hawaii, Honolulu, HI.
• IBSNAT (International Benchmark Sites Network for Agrotechnology Transfer Project). 1988. Technical report 1. Experimental design and data collection procedure for IBSNAT. The minimum data sets for systems analysis and crop simulation, 3rd ed. University of Hawaii, Honolulu, HI.
• 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 system model. Eur. J. Agron. 18:235–265.[CrossRef]
• Kropff, M.J., A.J. Haverkort, P.K. Aggarwal, and P.L. Kooman. 1995. Using systems approaches to design and evaluate ideotypes for specific environments. p. 417–435. In J. Bouma et al. (ed.) Eco-regional approaches for sustainable land use and food production. Kluwer Academic Publishers, Dordrecht, The Netherlands.
• Mavromatis, T., K.J. Boote, J.W. Jones, A. Irmak, D. Shinde, and G. Hoogenboom. 2001. Developing genetic coefficients for crop simulation models with data from crop performance trials. Crop Sci. 41:40–51.[Abstract/Free Full Text]
• Matthews, R., and W. Stephens. 2002. Crop-soil simulation models: Applications in developing countries. CABI Publishing, New York.
• Sinclair, T., and N.G. Seligman. 1996. Crop modeling: From infancy to maturity. Agron. J. 88:698–704.[Abstract/Free Full Text]
• Spitters, C.J.T., and A.H.C.M. Schapendonk. 1990. Evaluation of breeding strategies for drought tolerance in potato by means of crop growth simulation. Plant Soil 123:193–203.[CrossRef][ISI]
• Tsuji, G.Y., G. Uehara, and S. Balas. 1994. DSSAT Version 3, vols. 1–3. University of Hawaii, Honolulu, HI.
• Uehara, G. 1998. Synthesis. p. 179–188. In G. Y. Tsuji et al. (ed.) Understanding options for agricultural production. Kluwer Academic Publishers, Dordrecht, The Netherlands.
• Wallach, D., and B. Goffinet. 1987. Mean squared error of prediction in models for studying ecological and agronomic systems. Biometrics 43:561–573.[CrossRef][ISI]
• Whisler, F.D., B. Acock, D.N. Backer, R.E. Fye, H.F. Hodges, and J.R. Lambert. 1986. Crop simulation models in agronomic systems. Adv. Agron. 40:141–208.
• White, J.W. 1998. Modeling and crop improvement. p. 179–188. In G. Y. Tsuji et al. (ed.) Understanding options for agricultural production. Kluwer Academic Publishers, Dordrecht, The Netherlands.
• Willmott, C.J. 1982. Some comments on the evaluation of model performance. Bull. Am. Meteorol. Soc. 63:1309–1313.[CrossRef]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Suriharn, B. Articles by Hoogenboom, G. Search for Related Content PubMed Articles by Suriharn, B. Articles by Hoogenboom, G. Agricola Articles by Suriharn, B. Articles by Hoogenboom, G. Related Collections Other Legumes Crop Models