Response of global agricultural productivity anomalies to drought stress in irrigated and rainfed agriculture

The response of agricultural productivity anomalies to drought stress plays a crucial role in the carbon cycle within terrestrial ecosystems and in ensuring food security. However, detailed analysis of how global agricultural productivity anomalies response to drought stress, particularly within irrigated and rainfed agricultural systems, remains insufficient. In this study, the impact of drought stress on agricultural productivity anomalies during the growing season（zcNDVIS）, across both irrigated and rainfed agriculture, were analyzed using a suite of hydro-climatic variables. Specifically, the investigation utilized the multi-scalar Standardized Precipitation Evapotranspiration Index（SPEI）, the Multivariate ENSO Index（MEI）, and the Madden-Julian Oscillation（MJO）. Meanwhile, the relationships between hydroclimatic variables and zcNDVISwere analyzed at one, two, three and four months before the ending of growing season（EOS）. Results showed that（1） the percentages of significant（p<0.1） drying trends varied across the globe from 8.30% to 13.42%, 6.50% to 14.63%, 6.52% to 14.23%, and 6.47%to 14.95% at one-, two-, three-, and four-month lead times before EOS, respectively, during 2001–2020, which represented by the multiscalar SPEI. This observation highlights that most regions across the globe tend to be arid, which could significantly impact agricultural productivity;（2） the global mean correlation coefficients（rmax） for SPEI-1, SPEI-3, SPEI-6, SPEI-12（indicating SPEI at 1-, 3-, 6-, and 12-month lags）, MEI, and MJO with zcNDVISranged between 0.24–0.25, 0.27–0.28, 0.25–0.26, 0.21–0.22, –0.02–0.01 and 0.06–0.11, respectively, across both irrigated and rainfed agriculture system from 2001 to 2020.Agricultural productivity anomalies demonstrated a significant correlation with drought stress. The strongest correlations were noted for SPEI-3 and SPEI-6, suggesting a delayed response of crops to drought conditions. This indicates that agriculture ecosystem experiences prolonged disturbances due to abiotic drought stress; and（3） the percentages of regions that showed significant correlations（p<0.1） between zcNDVISand drought indices（SPEI-1, SPEI-3, SPEI-6, and SPEI-12）, as well as climate indices（MEI and MJO） ranged as follows: 14.77%–20.27%, 21.51%–32.55%, 22.60%–35.68%, 21.89%–35.16%,7.93%–11.20% and 9.44%–17.94%. Quantitatively identifying how zcNDVIS spatially responds to hydro-climatic variables can help us better understand the impact of drought on agricultural productivity anomalies worldwide.