Photo credit: phys.org
Recently, DSE and the University of Arkansas released a study in the Journal of Data Science, Statistics, and Visualization, along with an open-source tool aimed at outlining necessary reforms for federal crop insurance programs that could support farmers, agricultural businesses, and the environment.
The open-source tool leverages artificial intelligence to forecast potential scenarios for the U.S. corn sector, highlighting the escalating economic uncertainties that growers and insurers may encounter.
“Crop insurance premiums have surged by 500% since the early 2000s, and our projections indicate they could double again by 2050,” stated Sam Pottinger, the study’s lead author and a Senior Research Data Scientist & Software Engineer at DSE. “This alarming trend is largely attributed to an anticipated increase in extreme weather phenomena, posing significant risks to both farmers and insurance providers.”
The AI-driven model simulates growing conditions from 2020 to 2050 across various scenarios, including a “baseline” projection reflecting current agronomic practices. It then compares this baseline against predictions made by the Intergovernmental Panel on Climate Change, which foresees more frequent and severe weather events linked to climate change.
The research indicates that if current growing conditions persist, federal crop insurance companies may expect steady claims rates over the next three decades. However, in light of climate change, claims are projected to rise by 22% by 2030 and by 29% by 2050.
“Our projections reveal not only an increase in claims frequency but also a significant rise in the severity of these claims,” explained Lawson Conner, co-author of the research and an Assistant Professor in the Department of Agricultural Economics and Agribusiness at the University of Arkansas. “Insurance companies could face a rise of up to 19% in the average amount covered per claim by 2050.”
The newly launched web tool allows users to engage interactively with the study’s outcomes. Users can investigate how crop insurance rates are calculated by selecting the “rates” option, or assess the implications of different coverage structures for growers by accessing the “claims” section. The tool also features a “neighborhoods” tab that highlights areas likely to be most affected, and it provides access to the complete dataset for download.
“We chose to release this open-source tool alongside our study to promote transparency and extend our findings beyond the confines of academia,” remarked Maya Weltman-Fahs, Senior Program Manager at DSE. “This tool is particularly beneficial for policymakers and decision-makers, enabling them to take actionable steps.”
Federal crop insurance is administered by the United States Department of Agriculture (USDA) and acts as a vital source of economic support for farmers and agricultural entities in the U.S. Typically, federal coverage is available to all U.S. farmers, though some may opt for additional insurance from private providers. Currently, the primary determinant for federal crop insurance coverage, as established by the Farm Bill, is a farmer’s annual average crop yield.
To enhance protection for growers amid future challenges and minimize financial exposure for insurers, the study’s authors propose two key pathways for reform. Firstly, a minor modification to the Farm Bill could incentivize farmers to adopt or receive proper recognition for sustainable farming practices, such as cover cropping and crop rotation.
While these practices might yield lower average outputs in any given year, farmers are increasingly implementing them to enhance long-term crop resilience. Secondly, the authors highlight the potential for similar adjustments through 508(h), a mechanism that allows private companies to propose alternative or supplementary insurance products for USDA’s evaluation.
“As we witness more severe droughts, extended heat waves, and catastrophic flooding, the need for our recommendations becomes increasingly urgent,” noted Timothy Bowles, co-author and Associate Professor at UC Berkeley’s Department of Environmental Science, Policy, and Management. “Our insights aim to aid growers and insurance providers in navigating the impacts of extreme weather.”
“We are directly communicating our findings to advocates and policymakers to support informed decision-making regarding federal insurance policies.”
For those interested in a broader discussion on climate and agricultural policy, read more about our initiatives here.
This marks our second collaborative publication, following the autumn 2024 release of a Science paper and a complementary open-source tool, along with an interactive visualization detailing strategies for reducing global plastic pollution by up to 91% by 2050.
More information:
A Samuel Pottinger et al, Climate-driven doubling of U.S. maize loss probability: Interactive simulation with neural network Monte Carlo, Journal of Data Science, Statistics, and Visualisation (2025). doi.org/10.52933/jdssv.v5i3.134. On arXiv: DOI: 10.48550/arxiv.2408.02217
Source
phys.org