Detection and Attribution of Climate-Driven Impacts on Crop Yields

Celine Bonfils (17-ERD-115)

Executive Summary

We intend to apply detection-and-attribution techniques to determine the impact of climate change on the recent changes in California’s specialty-crop productivity. This research supports the DOE science and energy goals by improving our understanding of the impact of climate change on California’s economy and food security, and by informing decisions for adaptation strategies.

Project Description

For the past 20 years, scientists have developed and applied detection-and-attribution techniques to better understand the contribution of human activity on the climate system. Scientists at Lawrence Livermore National Laboratory pioneered and continue to contribute to the detection of the discernible human influence on observed changes in both surface temperature and the thermal structure of the atmosphere. In addition, Laboratory scientists have contributed to identifying human impacts on various aspects of the hydroclimate, including global-scale rainfall and water vapor, as well as in snowpack depth and river runoff of the western U.S. With an expected increase in food demand from rising population, agricultural systems will also become more vulnerable to human-induced climate change. We intend to apply detection and attribution techniques (developed at the Laboratory) to determine the impact of climate change on the recent changes in California’s specialty-crop productivity. This will involve investigating the relative impact on crop yields of the externally forced climate-change signal versus the internal climate-variability noise, assuming that adaptation practices and technological advances such as fertilizers, pesticides, and drought-tolerant crops remain unchanged. While these empirical functions are based on a limited number of records, they constitute the most reliable estimate of how individual crop yields respond to weather and will utilize variables that are readily available in most climate simulations. The feasibility of this work is reinforced by a recent study focusing on global yields of wheat and maize. In contrast, perennial crops may be less subject to technology and management innovations.

We intend to use empirical crop functions to determine the range of expected crop-yield outcomes over the historical period using data from state-of-art simulations of historical climate change driven by changes in both natural and human factors. We will compare this historical distribution of yields with that inferred from internal climate noise alone, using data from control climate integrations without forcing changes. Finally, we will assess whether the recent changes in crop yields (either directly observed or calculated using observed climate data) are accounted for by climate noise alone, by recent shifts of climate signal, or by non-climatic factors (lying outside the climate signal and noise distributions) that may impact yields. We expect to identify regional trends, or a fingerprint (i.e, a geographical pattern in yield changes one would expect a priori from climate trends). Alternatively, we will determine the detection time at which an influence of climate change is expected in the yield changes. We will leverage previous studies that employed empirical crop models to assess yield responses of the most valuable California perennial crops (such as grapes, cherries, almonds, and strawberries) to future climate change, taking into account both climate and crop uncertainties. Particular attention will be given to the sensitivity of conclusions to various sources of uncertainty, including the structural uncertainties associated with the crop models, estimation of fingerprint and climate noise, model-quality information, and the incorporation of new yield observations.

Mission Relevance

This research supports the DOE science and energy goals as well as the Laboratory’s strategic focus area of Earth and atmospheric science by improving the understanding of the impact of climate change on California’s economy and food security, and by informing decisions for adaptation strategies.

FY17 Accomplishments and Results

Due to programmatic commitments, this project was postponed to FY18. However, in FY17, the principal investigator attended the California Climate Change Symposium in Sacramento in January to meet with climate scientists, specialists in climate adaptation in agriculture, and economists interested in understanding the broad impact of climate change on the agricultural sector.