Classification of Maize Environments Using Crop Simulation and Geographic Information Systems

productivity over long periods of time, these efforts did not attempt to identify the environmental variables that The effectiveness of a cultivar evaluation system largely depends were most important in influencing GEI and thus the on the genetic correlation between genotype performance in multigenetic correlations for genotype performance among environment trials (MET) and in the target population of environments (TPE). Previous classifications of maize (Zea mays L.) environtesting sites, a key factor in determining the efficiency ments on the basis of climate and soil did not quantify their impact and efficacy of a cultivar evaluation system. Consequently, on the genetic correlations among environments. Consequently, plant plant breeders have more extensively used classificabreeders have favored classifications based on the similarity of cultivar tions of environments based on similarity of cultivar discrimination in trials. However, these efforts frequently fail to prodiscrimination using crop performance data collected vide adequate assessments of the TPE, since they require long-term from their cultivar evaluation or ad hoc trials, rather performance data, which are not normally collected due to high cost. than basing the classification on environmental data. To describe the TPE, we performed crop simulations for each U.S. Cooper et al. (1993) compared the relative merit of Corn Belt Township for the period 1952 through 2002, using standard four strategies for classifying wheat (Triticum aestivum CERES-Maize model inputs. To classify METs, input data were colL.) environments and favored classifications based on lected at or near the trial sites. Grain yield and biotic stress data for model confirmation were collected from 18 hybrids grown in replithe standardized and rank transformations. The value cated trials in 266 environments in 2000–2002. On the basis of prevailof these classifications for predicting cultivar perforing conditions during key growth stages, and observed patterns of mance is enhanced by knowledge of (i) the underlying genotype environment interactions (GEI), six major environment causes of the observed GEI and (ii) whether the classificlasses (EC) were identified. The relative frequency of each EC varied cation adequately depicts long-term patterns. greatly from year to year and significant hybrid EC interaction While requirement (i) can be met by collecting approvariance was observed. Our environmental classification system propriate environmental information from the testing sites, vided a useful description of some of the features of both the TPE these efforts normally fail to provide an adequate longand MET. Knowledge of the spatial (locations) and temporal (years) term description of the TPE, mainly because of the cost distributions of ECs that influence the incidence of GEI can be used and impracticality of collecting empirical performance to improve cultivar performance predictability in the U.S. Corn Belt TPE. data for long-term studies. More recent efforts to characterize environments for crop production have utilized crop models to integrate weather, soil, and management information and to proT effectiveness of a corn cultivar evaluation sysduce categorical outputs that describe environments in tem largely depends on the degree to which the terms of levels of stress that impact crop productivity. MET represents the TPE (Comstock, 1977). Using comUsing the Agricultural Production Systems sIMulator puter simulation, Cooper and Podlich (1999) and Pod(APSIM) crop growth simulation model, Chapman et lich et al. (1999) demonstrated the value of using a al. (2000) integrated soils and between 80 and 105 yr of weighted selection strategy when the environments samweather data to classify sorghum [Sorghum bicolor (L.) pled in the MET did not match the expectations in the Moench] environments in Queensland, Australia. For TPE. The advantage of the weighted strategy increased a subset of six testing locations, they found that three as the amount of crossover GEI observed in the MET environment types, described in terms of the timing and increased. Clearly, an adequate classification of the enintensity of water stress, had a consistent relationship vironments that compose the TPE constitutes a prereqwith simulated yield. uisite for implementing a successful weighted selecSince the advent of crop simulation with the pioneertion strategy. ing work from de Wit (1965), the CERES-Maize model Previous efforts to classify maize environments relied was developed by the USDA-ARS primarily for asmainly on climate and soils data (e.g., Runge, 1968; sisting with crop management decisions, strategic planPollak and Corbett, 1993; Hartkamp et al., 2000). While ning, yield forecasting, and definition of research needs useful to describe environmental variables affecting crop Abbreviations: APSIM, agricultural production systems simulator; AWC, available water capacity; CERES, crop environment resource Pioneer Hi-Bred International, Inc., 7250 NW 62nd Ave., P.O. Box synthesis; CRM, corn relative maturity; EC, environment class; ECB, 552, Johnston, IA 50131. Received 17 June 2004. *Corresponding European corn-borer (Ostrinia nubilalis H.); GEI, genotype by enviauthor ([email protected]). ronment interactions; GGE, genotype main effects plus genotype by environment interaction effects; GIS, geographic information system; Published in Crop Sci. 45:1708–1716 (2005). Crop Breeding, Genetics & Cytology MET, multienvironment trials; NOAA, National Oceanic and Atmospheric Administration; RCB, randomized complete block design; doi:10.2135/cropsci2004.0370 © Crop Science Society of America STATSGO, state soil geographic database; TPE, target population of environments. 677 S. Segoe Rd., Madison, WI 53711 USA 1708 Published online August 1, 2005