Key Takeaways
- Non-Linear Effects: Daily temperature extremes have a more significant impact on crop yields than average temperatures.
- Predictive Models: Global meteorological datasets like GMFD and ERA5-Land can accurately predict yield impacts despite lower resolution compared to fine-scaled data.
- Yield Sensitivity: Temperature sensitivity varies across crops, with extreme heat reducing yields significantly.
- Global Applications: These datasets are effective in data-sparse regions like Sub-Saharan Africa, outperforming traditional monthly weather data.
Introduction
Assessing climate change impacts on agriculture requires understanding how temperature extremes affect crop yields. Hogan and Hogan (2024) investigate this by comparing various global meteorological datasets to fine-scaled country-specific data in the US and Sub-Saharan Africa.
Research Overview
The study evaluates the relationship between daily temperature extremes and yields of corn and soybeans in the US. It compares fine-scaled PRISM data with global datasets GMFD and ERA5-Land, and extends the analysis to Sub-Saharan Africa using the Enhanced Vegetation Index (EVI) as a yield proxy.
Methodology
Data Comparison: The study uses yield data from US counties and evaluates predictive performance of weather-yield models based on PRISM, GMFD, and ERA5-Land datasets.
Functional Forms: Various models were tested, including piecewise linear regressions, 8th order polynomials, and semi-parametric specifications.
Regional Extension: The analysis was extended to Sub-Saharan Africa, comparing GMFD and ERA5-Land with CRU, a monthly weather dataset.
Results
US Analysis:
- Non-Linear Impact: All datasets revealed that temperature extremes have a non-linear effect on yields, with moderate temperatures increasing yields and extreme heat reducing them.
- Model Performance: PRISM data showed the highest predictive accuracy, but global datasets also performed well, capturing essential temperature-yield relationships.
Sub-Saharan Africa:
- Predictive Power: GMFD and ERA5-Land outperformed CRU, showing better predictive power for crop yields in this data-sparse region.
Conclusion
The study by Hogan and Hogan (2024) underscores the importance of accounting for daily temperature extremes in climate impact models. Global meteorological datasets like GMFD and ERA5-Land, despite lower spatial resolution, provide valuable insights into yield sensitivity and can be crucial for regions lacking detailed weather data. These findings emphasize the need for accurate temperature measurements to inform agricultural adaptation strategies.
For more detailed information, you can access the full study here.
Photo by Andriy Babchiy on Unsplash
