Probability of maize yield failure increases with drought occurrence but partially depends on local conditions in China

Abstract

Extreme climate events such as drought have significantly adversely affected crop yields and challenged the stability of global food production. Assessing the risk posed by drought hazards to crop yield is of great importance for yield forecasting and agricultural risk management; however, the relationship of crop yield with drought occurrence from a probabilistic perspective is still insufficient. Taking major maize cropping regions in China as an example, we integrate maize yield anomalies and standardized precipitation evapotranspiration index (SPEI) values over 1981–2014, develop a bivariate copula-based probabilistic diagram, and compare the probabilities of yield failure under various drought conditions within 1094 counties. The results show that maize yield anomalies are negatively correlated with drought intensity in most counties despite varying timescales, and maize yield decreases by 4.91 %, 10.44 %, and 7.34 % under moderate, severe, and extreme drought conditions, respectively. The higher probability of yield failure confirmed that maize yield is more sensitive to drought conditions than near-normal conditions. The probability of yield failure increases significantly with increasing drought severity, measuring 61.43 %, 66.01 %, and 69.46 % under moderate, severe, and extreme drought conditions, respectively, revealing a nonlinear response of maize yield to drought severity. Additionally, the susceptibility of maize cropped in semiarid regions to yield loss across conditions is attributed to regional climatic conditions. Because of the high risk of yield failure during drought events, suitable adaptations to drought considering local endowments are essential for risk management and decision-making under a continuously changing climate.