Companies' sustainable sourcing practices play an increasing role in addressing the social and environmental challenges in agricultural supply chains. Yet the approaches companies take to regulate their supply chains continue to evolve. I use the chocolate industry as a critical case to explore how and why companies have changed their approaches to sustainable cocoa sourcing over the last 20 years. Drawing on an analysis of 205 company documents, 95 newspaper articles and over 50 in‐depth interviews, I trace the evolution of chocolate manufacturers' sustainable sourcing practices from a focus on industry initiatives to a commitment to sustainability certification and now to companies increasingly moving toward own‐supply chain programs. These shifts can in part be explained by the evolving salience of different stakeholder groups over time. This study highlights the dynamic nature of sustainable sourcing practice adoption and suggests companies are building upon previous strategies to incorporate more stakeholder voices over time.

Solar radiation management is increasingly considered to be an option for managing global temperatures1,2, yet the economic effects of ameliorating climatic changes by scattering sunlight back to space remain largely unknown3. Although solar radiation management may increase crop yields by reducing heat stress4, the effects of concomitant changes in available sunlight have never been empirically estimated. Here we use the volcanic eruptions that inspired modern solar radiation management proposals as natural experiments to provide the first estimates, to our knowledge, of how the stratospheric sulfate aerosols created by the eruptions of El Chichón and Mount Pinatubo altered the quantity and quality of global sunlight, and how these changes in sunlight affected global crop yields. We find that the sunlight-mediated effect of stratospheric sulfate aerosols on yields is negative for both C4 (maize) and C3 (soy, rice and wheat) crops. Applying our yield model to a solar radiation management scenario based on stratospheric sulfate aerosols, we find that projected mid-twenty-first century damages due to scattering sunlight caused by solar radiation management are roughly equal in magnitude to benefits from cooling. This suggests that solar radiation management—if deployed using stratospheric sulfate aerosols similar to those emitted by the volcanic eruptions it seeks to mimic—would, on net, attenuate little of the global agricultural damage from climate change. Our approach could be extended to study the effects of solar radiation management on other global systems, such as human health or ecosystem function.

Rising atmospheric carbon dioxide concentrations are anticipated to decrease the zinc and iron concentrations of crops. The associated disease burden and optimal mitigation strategies remain unknown. We sought to understand where and to what extent increasing carbon dioxide concentrations may increase the global burden of nutritional deficiencies through changes in crop nutrient concentrations, and the effects of potential mitigation strategies.

Climate-induced shocks in grain production are a major contributor to global market volatility, which creates uncertainty for cereal farmers and agribusiness and reduces food access for poor consumers when production falls and prices spike. Our study, by combining empirical models of maize production with future warming scenarios, shows that in a warmer climate, maize yields will decrease and become more variable. Because just a few countries dominate global maize production and trade, simultaneous production shocks in these countries can have tremendous impacts on global markets. We show that such synchronous shocks are rare now but will become much more likely if the climate continues to warm. Our results underscore the need for continued investments in breeding for heat tolerance.

Integrated assessment models generate climate change mitigation scenarios consistent with global temperature targets. To limit warming to 2 °C, cost-effective mitigation pathways rely on extensive deployments of CO2 removal (CDR) technologies, including multi-gigatonne yearly CDR from the atmosphere through bioenergy with carbon capture and storage (BECCS) and afforestation/reforestation. While these assumed CDR deployments keep ambitious temperature targets in reach, the associated rates of land-use transformation have not been evaluated. Here, we view implied integrated-assessment-model land-use conversion rates within a historical context. In scenarios with a likely chance of limiting warming to 2 °C in 2100, the rate of energy cropland expansion supporting BECCS proceeds at a median rate of 8.8 Mha yr−1 and 8.4% yr−1. This rate exceeds—by more than threefold—the observed expansion of soybean, the most rapidly expanding commodity crop. In some cases, mitigation scenarios include abrupt reversal of deforestation, paired with massive afforestation/reforestation. Historical land-use transformation rates do not represent an upper bound for future transformation rates. However, their stark contrast with modelled BECCS deployment rates implies challenges to explore in harnessing—or presuming the ready availability of—large-scale biomass-based CDR in the decades ahead. Reducing BECCS deployment to remain within these historical expansion rates would mean either the 2 °C target is missed or additional mitigation would need to occur elsewhere.

The availability of climate model experiments under three alternative scenarios stabilizing at warming targets inspired by the COP21 agreements (a 1.5 ºC not exceed, a 1.5 ºC with overshoot and a 2.0ºC) makes it possible to assess future expected changes in global yields for two staple crops, wheat and maize. In this study an empirical model of the relation between crop yield anomalies and temperature and precipitation changes, with or without the inclusion of CO2 fertilization effects, is used to produce ensembles of time series of yield outcomes on a yearly basis over the course of the 21st century, for each scenario. The 21st century is divided into 10 year windows starting from 2020, within which the statistical significance and the magnitude of the differences in yield changes between pairs of scenarios are assessed, thus evaluating if, and when, benefits of mitigations appear, and how substantial they are. Additionally, a metric of extreme heat tailored to the individual crops (number of days during the growing season above a crop-specific threshold) is used to measure exposure to harmful temperatures under the different scenarios. If CO2 effects are not included, statistically significant differences in yields of both crops appear as early as the 2030s but the magnitude of the differences remains below 3% of the historical baseline in all cases until the second part of the century. In the later decades of the 21st century, differences remain small and eventually stop being statistically significant between the two scenarios stabilizing at 1.5 ºC, while differences between these two lower scenarios and the 2.0ºC scenario grow to about 4%. The inclusion of CO2 effects erases all significant benefits of mitigation for wheat, while the significance of differences is maintained for maize yields between the higher and the two lower scenarios, albeit with smaller benefits in magnitude. Changes in extremes are significant within each of the scenarios but the differences between any pair of them, even by the end of the century are only on the order of a few days per growing season, and these small changes appear limited to a few localized areas of the growing regions. These results seem to suggest that for globally averaged yields of these two grains the lower targets put forward by the Paris agreement does not change substantially the expected impacts on yields that are caused by warming temperatures under the pre-existing 2.0ºC target.