Despite real-world adaptation, global breadbaskets remain highly exposed to intensifying heat, and the world’s food supply could shrink faster than expected.
Image credit: Fotokostic/Shutterstock.com
In a recent article published in the journal Nature, researchers assessed how real-world farmers respond to climate stressors across diverse geographical and socioeconomic settings, and projected the future impacts of climate change on staple crop production.
Challenges In Real-World Farming
The foundation of this research rests on extensive prior work indicating agricultural productivity is highly sensitive to climatic variables, notably temperature and precipitation. Earlier models, including process-based simulations calibrated on experimental and managed farm data, have projected varying outcomes; some suggest possible productivity gains through technological innovations and optimal adaptation, while others forecast significant losses owing to climatic stressors.
Traditionally, models assume idealized adaptation, such as the adoption of new crop varieties or advanced irrigation systems, often overlooking real-world constraints, including financial limitations, market failures, or the human capacity to implement such measures. The importance of understanding how resource access, economic development, and technological adoption influence adaptive capacity has become especially critical in an era of accelerated climate change.
Recent advances have facilitated better mapping and analysis of regional responses; however, capturing the full spectrum of adaptive responses across diverse contexts remains a challenge, particularly given that many process-based models rely on data from highly managed experimental fields rather than real-world farms.
Building a Global Yield Model
This research employs an innovative, data-driven empirical approach that centers on analyzing longitudinal, high-resolution crop yield data across a diverse array of regions and socioeconomic backgrounds. The dataset encompasses approximately 12,658 subnational administrative units spanning 54 countries, covering a spectrum of climates and income levels.
The analysis focuses on six key staple crops, selected systematically based on their contribution to global caloric intake and food security. To understand the influence of weather on crop yields, the researchers employ cross-validated, high-dimensional statistical models that allow the data to identify relevant climate variables without imposing rigid assumptions.
They analyze a broad set of meteorological inputs, such as average temperature and rainfall, as well as more complex measures like degree days, extreme heat exposure, and precipitation extremes, for their potential to inform crop biophysical responses, and use cross-validation procedures to select the most predictive climate measures specific to each crop and region.
A crucial aspect of their methodology is capturing real-world adaptations by farmers. Unlike traditional models that impose predefined adaptation scenarios, this study uses empirical data to estimate the net effect of all adaptive adjustments that producers actually undertake, including but not individually identifying actions such as changing crop varieties, altering planting dates, and applying different fertilizers or irrigation practices.
By correlating climatic variables with yield responses while controlling costs and resource accessibility, the researchers develop a reduced-form model that integrates biophysical responses with socioeconomic factors. They also incorporate projections of future resource access, such as income levels and irrigation availability, based on socio-economic scenarios, to simulate potential adaptive behaviors under different climate pathways.
However, some adaptation mechanisms, such as crop switching or shifting planting dates outside current growing-season windows, are not captured directly, and their effects remain outside the model’s scope.
Where Losses Hit the Hardest
The findings reveal a stark picture of climate change’s ongoing and future impacts on global agriculture. The data indicate that crop production declines approximately linearly with rising average global temperatures, with each 1°C increase in global mean surface temperature potentially reducing caloric output by about 120 kilocalories per person per day. This translates to a significant loss in food availability, posing a threat to nutritional security worldwide.
However, the analysis does not treat all regions equally. Instead, it shows that the severity of impacts varies according to both geographical location and the degree of existing adaptive capacity. Regions with hotter current climates and limited access to technological or resource-based adaptation strategies are more vulnerable; some of the most affected areas are the world's breadbaskets, regions that have historically enjoyed stable climates and high productivity but are now demonstrating limited adaptation to rising temperatures.
The study demonstrates that adaptation efforts and income growth could mitigate roughly one-quarter to one-third of the projected losses by mid-century and at century's end. For instance, in projected scenarios, adaptation is estimated to reduce approximately 23% of the total losses by 2050 and about 34% by 2100 under moderate emissions pathways.
Nonetheless, substantial residual losses remain, especially for staple crops such as maize, wheat, soybeans, and rice, with rice faring somewhat better due to its resilience and existing adaptive practices. However, there is substantial uncertainty around these projections, with some scenarios allowing for modest gains for certain crops, particularly rice
Notably, the analysis highlights that current adaptation levels are highly uneven globally. Cold and moderate-climate, high-income breadbasket regions exhibit limited existing adaptation, whereas some low-income and resource-constrained regions appear more acclimated to heat exposure. Although they remain highly vulnerable due to economic limitations and reliance on climate-sensitive crops, such as cassava.
The model also finds that long-term yield trends are primarily driven by temperature, while precipitation plays a larger role in year-to-year variability rather than long-term yield changes.
Securing Food Systems
As global temperatures continue to climb, it becomes increasingly vital to integrate scientific insights with practical, ground-based solutions that empower local communities and foster sustainable agricultural practices. These actionable measures, rooted in empirical evidence, can help ensure that the global food system remains robust and capable of feeding an expanding population under changing climatic conditions.
Ultimately, this work provides a roadmap for understanding where vulnerabilities lie and how adaptive capacity can be strengthened to sustain human well-being in the face of an uncertain future climate.
Download your PDF copy now!
Journal Reference
Hultgren A., Carleton T., et al. (2025). Impacts of climate change on global agriculture accounting for adaptation. Nature 642, 644–652. DOI: 10.1038/s41586-025-09085-w, https://www.nature.com/articles/s41586-025-09085-w