HR: 08:35h
AN: B31A-03    [PDF]
TI: Simulations of Decadal-scale Climate Change Impacts on Agriculture: Attributing Trends in Regional Corn Yields to Physiological Effects Versus Adjusted Farm Management
AU: * Kucharik, C
EM: kucharik@wisc.edu
AF: Center for Sustainability and the Global Environment, University of Wisconsin-Madison, 1710 University Avenue, Madison, WI 53726 United States
AB: A recent study published in {\it Science} in early 2003 [by David Lobell and Gregory Asner, Dept. of Global Ecology, Carnegie Inst. of Washington] highlighted that little effort has been put forth to understand the impacts of previous decadal-scale climate changes on row-crop agriculture. The major obstacle to overcome in quantifying crop response to climate changes over large regions is deciphering between changes attributed to climate change versus technology, land-management and other factors. While the Lobell and Asner study concluded that regional temperature trends potentially contributed to corn and soybean yield trends from 1982-1998, a partitioning of the observed increases between direct physiological effects versus farmer management adjustment to climate was not a goal of their study. As part of this study, an agricultural version of the Integrated BIosphere Simulator (Agro-IBIS), was used to investigate how decadal-scale climate changes may have contributed to corn yield trends across the Mississippi Basin from 1948-2001. The primary objective was to investigate the relative contributions of physiological effects and farmer adjustments in planting date and hybrid choice to long-term corn yield trends. The impacts of advancing technology on agriculture were removed from model simulations so that the impact of weather and farm management decisions (e.g., planting date and hybrid choice) could be separated from observed long-term trends in the USDA crop yield record. Scenarios were used that accounted for smart-farmers, where management adjustments (planting date and/or hybrid) were made in response to climate changes, and for business-as-usual-farmers who continued to plant the same hybrids on the same date during the study period. When average, optimum corn planting dates from the 1950s were compared with the 1990s, significant springtime warming in regions of the northern fringes of the cornbelt (e.g., North Dakota, Minnesota) over the past 40 years have caused optimum corn planting dates to retreat by 6 to 16 days, a trend of up to 0.4 days per year. This simulated result is in accordance with previous studies that confirmed an advance in the growing season in several regions of the globe of approximately 8-10 days during the last several decades. However, the degree of springtime warming appeared to decrease significantly towards the southern boundaries of the cornbelt, as optimum planting dates advanced by 1 to 5 days in a region from eastern Kansas through Missouri and central Illinois and Indiana. Not surprising was the response of a generic corn hybrid (1400 growing degree-days [GDD] to reach physiological maturity) for these two time periods. Clearly, the earlier planting dates over much of the northwest portion of the cornbelt region coincided with a simulated average 0.25 to 3 Mg ha$^{-1}$ increase in corn yield between the two decadal periods. We hypothesize that this region of the Northern Great Plains has likely seen significant changes and trends in management decision-making over the past several decades, most likely in the choice of corn hybrids (GDD to silking and maturity), planting dates, or a combination of the two in response to springtime warming. Therefore, in this region, farmer adaptation to climate trends may have taken precedence over the direct physiological effects of warmer temperatures towards increasing corn production.
DE: 1600 GLOBAL CHANGE (New category)
DE: 1615 Biogeochemical processes (4805)
DE: 1620 Climate dynamics (3309)
DE: 1851 Plant ecology
SC: Biogeosciences [B]
MN: 2003 Fall Meeting