Rising temperatures can negate CO2 fertilization effects on global staple crop yields: A meta-regression analysis

Abstract

Many meta-analyses have reported increased seed yields in response to rising carbon dioxide concentrations ([CO2]) or decreased yields in response to warming temperatures, but synthesis of their concurrent effects is lacking. We conducted an extensive meta-regression analysis of wheat, rice, soybean, and maize seed yield responses to simultaneous increases in [CO2] and temperature and examined the sensitivity of potential yields to levels projected for mid- and late-century periods using CMIP6 climate model outputs. Positive linear yield responses to increasing [CO2] were observed in maize, rice, and soy, with slopes ranging from +4 % - (100 µmol mol−1) -1 in rice to +10 % - (100 µmol mol−1) -1 in soy. In contrast, wheat yield response to increasing [CO2] was not linear, instead flattening by +200 µmol mol−1 of CO2 elevation. Negative linear yield responses to warming were modeled for the three cereal crops, with slopes of −7.6% -°C −1 in wheat and −9.5% -°C −1 in rice and maize. Soy yield did not decrease with increasing warming treatments in this analysis, but experimental warming treatments did not exceed optimal soy growth temperature ranges. Across the three crops with available observations, yield responses were significantly lower in Free-Air CO2 Enrichment than in studies using other fumigation methods, likely due to the associated rapid fluctuations in [CO2]. Considering potential future eC and eT levels simultaneously, we find that eC fertilization effects on seed yield will likely be greatly reduced or entirely negated in the major crops maize, rice, and wheat. Net yield responses to concurrent [CO2] and temperature increases should not be based on limited observations of responses to either factor in isolation, because the levels of each factor as well as experimental conditions influence responses.