Climate impacts on net primary productivity trends in natural and managed ecosystems of the central and eastern United States

The central and eastern United States is an important food and timber production region and could potentially be a large-scale carbon sink. These characteristics are functions of the human management of the landscape and favorable soil and climate conditions. Large-scale assessments of carbon uptake and trends in net primary productivity (NPP) have been previously reported for this region using satellite observations, but they cannot quantify the contributions from changes in management and technology independently from climate effects. To address this shortcoming, we used the Agro-IBIS dynamic ecosystem model to examine the relationship of temperature and precipitation trends to NPP changes from 1950 to 2002 and 1982 to 2002 over both natural and managed ecosystems. In order to isolate the vegetation growth response to climate trends, we minimized the representation of management for agroecosystems and forested ecosystems by removing nitrogen stress and irrigation from the model. Simulated NPP trends were larger during 1982-2002 than during 1950-2002 and were positive across most of the domain. For the 1982-2002 period, corn had the largest NPP trend of 6.43 g C m^-2 year^-2 (p = 0.22), followed by soybean, spring wheat, deciduous forest, and grassland with 4.06 (p = 0.30), 4.05 (p = 0.42), 1.98 (p = 0.17), and 0.84 g C m^-2 year^-2 (p = 0.73), respectively. Winter wheat had a trend of -0.64 g C m^-2 year^-2 (p = 0.55) and evergreen needleleaf forest had a negligible NPP trend. Trends in NPP in grasslands, deciduous forest, spring wheat, and winter wheat were associated with trends in precipitation, while trends in corn and soybean were attributed to a combination of longer growing periods, decreased summer average temperatures, and increased precipitation. We found that 19% of corn and 11% of soybean NPP trends could be explained by temperature trends while 23% of corn and 44% of soybean trends could be explained by precipitation trends. Our results provide further evidence supporting observational results that suggest 20-25% of recent crop yield trends can be explained by changing climate, and suggest that over the past several decades climate changes have favored increased crop productivity in most agroecosystems of the central United States with the exception of winter wheat.