H53B-1661
Alternatives to Crop Insurance for Mitigating Hydrologic Risk in the Upper Mississippi River Basin

Friday, 18 December 2015
Poster Hall (Moscone South)
John M Baker1, Timothy J Griffis2, Galen Gorski2,3 and Jeffrey D Wood2, (1)University of Minnesota Duluth, Duluth, MN, United States, (2)Univ Minnesota, Saint Paul, MN, United States, (3)University of California Santa Cruz, Santa Cruz, CA, United States
Abstract:
Corn and soybean production in the Upper Mississippi River Basin can be limited by either excess or shortage of water, often in the same year within the same watershed. Most producers indemnify themselves against these hazards through the Federal crop insurance program, which is heavily subsidized, thus discouraging expenditures on other forms of risk mitigation. The cost is not trivial, amounting to more than 60 billion USD over the past 15 years. Examination of long-term precipitation and streamflow records at the 8-digit scale suggests that inter-annual hydrologic variability in the region is increasing, particularly in an area stretching from NW IL through much of IA and southern MN. Analysis of crop insurance statistics shows that these same watersheds exhibit the highest frequency of coincident claims for yield losses to both excess water and drought within the same year. An emphasis on development of water management strategies to increase landscape storage and subsequent reuse through supplemental irrigation in this region could reduce the cost of the crop insurance program and stabilize yield. However, we also note that analysis of yield data from USDA-NASS shows that interannual yield variability at the watershed scale is much more muted than the indemnity data suggest, indicating that adverse selection is probably a factor in the crop insurance marketplace. Consequently, we propose that hydrologic mitigation practices may be most cost-effective if they are carefully targeted, using topographic, soil, and meteorological data, in combination with more site-specificity in crop insurance data.