Various maize yield losses and their dynamics triggered by drought thresholds based on Copula-Bayesian conditional probabilities

Abstract

In the context of global warming and human interventions, the impact of climate on crop yield may change over time. Therefore, assessing the dynamics of drought thresholds that trigger various maize yield losses is critical for food security under climate change. To this end, this study aims to assess the vulnerability of maize to drought stress in three provinces of Northeast China (Heilongjiang, Jilin and Liaoning provinces) and to quantify the drought thresholds that cause different levels of maize yield loss. A Copula-Bayesian conditional probability bivariate model is constructed to combine drought conditions and maize yield. Results indicate that: (1) for the whole study period 1980–2018, drought thresholds that induced different levels of maize yield reduction were significantly different in the three northeastern provinces of China; on average, the drought thresholds that induced 30%, 40% and 50% maize yield losses were −1.06, −1.53 and −2.23 in the three provinces of Northeast China; (2) during the transition from mild to moderate and severe drought, maize vulnerability in Liaoning province gradually exceeded that of Jilin and Heilongjiang province; (3) from 1980 to 1999–2000–2018, the drought thresholds that triggered the same percentage of maize yield reduction increased in all three provinces, suggesting a dramatically increasing trend in the vulnerability of maize yields to drought; (4) the changes in precipitation and evapotranspiration leading to increased drought severity were the main factors inducing drought threshold dynamics in both sub-study periods. The probabilistic assessment of the impact of drought on maize yield is expected to provide useful insights into the mitigation of drought and its effects under climate change.