Agriculturally Relevant Climate Extremes and Their Trends in the World's Major Growing Regions

Abstract

Climate extremes can negatively impact crop production, and climate change is expected to affect the frequency and severity of extremes. Using a combination of in situ station measurements (Global Historical Climatology Network's Daily data set) and multiple other gridded data products, a derived 1° data set of growing season climate indices and extremes is compiled over the major growing regions for maize, wheat, soybean, and rice for 1951–2006. This data set contains growing season climate indices that are agriculturally relevant, such as the number of hot days, duration of dry spells, and rainfall intensity. Before 1980, temperature‐related indices had few trends; after 1980, statistically significant warming trends exist for each crop in the majority of growing regions. In particular, crops have increasingly been exposed to extreme hot temperatures, above which yields have been shown to decline. Rainfall trends are less consistent compared to temperature, with some regions receiving more rainfall and others less. Anomalous temperature and precipitation conditions are shown to often occur concurrently, with dry growing seasons more likely to be hotter, have larger drought indices, and have larger vapor pressure deficits. This leads to the confluence of a variety of climate conditions that negatively impact crop yields. These results show a consistent increase in global agricultural exposure to negative climate conditions since 1980.

Climate adaptation by crop migration

“…As shown in previous studies 45 , rainfed crops are currently experiencing average growing season temperatures that are hotter than they were in the 1970s because of climate change ( Fig. 2e-h).…”

Projected increase in compound dry and hot events over global land areas

“…The impact may be caused by individual non‐extreme events (Seneviratne et al ., 2012), and thus, the compound event with individual variables at a moderate level could also seriously affect ecosystems and society, which deserves appropriate attention. Following previous studies using percentile‐based thresholds to characterize compound events (Beniston, 2009; López‐Moreno et al ., 2011; Fortin and Hétu, 2013; Hao et al ., 2013; Kirono et al ., 2017; Martin and Germain, 2017; Zhu and Troy, 2018; Li et al ., 2019), we defined compound dry and hot events as the concurrence of seasonal precipitation lower than the 30th percentile threshold and seasonal temperature higher than the 70th percentile threshold (denoted as P < 30th and T > 70th). Based on previous studies (Hansen et al ., 2012; Coumou and Robinson, 2013), we chose 1951–1980 as the reference period because it was identified as a period with relatively stable global mean temperature.…”

A Granger causality analysis of groundwater patterns over a half-century

“…Climatic datasets included precipitation totals (annual and summer), minimum and maximum air temperatures, Growing Degree Days (GDD), and Palmer Drought Severity Index (PDSI). The Growing Degree Days (GDD) is one of the commonly used climate indices that may exert a strong influence on crop production 39 , and it is estimated as mean air temperature above a certain threshold accumulated over time (Table 3). The Palmer Drought Severity Index (PDSI) is a climate index representative of dryness of landscape derived from a combination of variables such as precipitation, air temperature and soil moisture 40 (Table 3).…”