Effect of environmental variables on performance of Jerusalem artichoke (Helianthus tuberosus L.) cultivars in a long term trial: a statistical approach

Abstract

Tuber yield, together with tuber number and size are the basic agronomic and breeding traits in Jerusalem artichoke and can be significantly affected by environmental factors. We report the results of a long term trial on the performance of 20 Jerusalem artichoke cultivars. The random model for means with restricted maximum likelihood (REML) procedure was used to estimate the overall effects of the genotype, environment and genotype by environment interaction on traits. The partial least square regression (PLSR) model was used for modeling genotype by environment interaction variance components with a set of available correlated environmental variables. The REML variance component estimates model revealed that tuber number and yield are more dependent on GE interaction which allowed identification of best genotypes for specific environments. The PLSR model revealed that the most important climatic variables for optimal emergence, canopy development, high tuber number and yield are adequate soil and air temperatures in April. For larger tuber mass, precipitation variables and even distribution of rainfall were the most important factor, together with soil and air temperature in June when tuber growth is initiated. The knowledge obtained in this study is valuable for the identification and understanding of key environmental factors that contribute to the performance of Jerusalem artichoke.