Simulating the Effects of Agricultural Land Use Change on the Climate of the Northern North American Great Plains

The northern North American Great Plains (NNAGP) have undergone extensive land use changes. Of these, the agricultural practice of summer fallow was widespread in the 1970s, but has largely disappeared to the benefit of soil conservation. The adoption of alternate cropping systems and no-till agriculture has led to nearly 200,000 km² of land that is now planted instead of left fallow across the NNAGP. Increased evapotranspiration and higher albedo from vegetation has led to a reduction in sensible heat flux and lower Bowen ratios that have decreased from ca. 2 to ca. 1 across parts of the NNAGP since the 1970s increasing the likelihood of rainfall in the region. The NNAGP have cooled at -0.2 ° C decade ^-1 during May and June across this time period, nearly opposite the global warming trend. Near surface humidity has increased and the vapor pressure deficit has decreased by -0.2 hPa decade^-1 during May and June. Precipitation has similarly increased during the same time, exceeding 8 mm decade^-1 and accompanied by a 40% increase in convective likelihood during May and June. Global climate models have been unable to correctly simulate these changes, likely due to the coarse resolution and inaccurate land cover data.

Here, we use the Weather Research and Forecasting (WRF) model with convection-permitting grid spacing to test what proportion of the observed changes can be attributed to shifts in land use compared to internal climate variability. WRF is run at 4 km horizontal grid spacing with ERA5 as initial and lateral boundary conditions. The model’s land cover data is improved by integrating a more accurate representation of fallow captured by Landsat in the early 1980s and the mid 2010s. We perform two sets of 4-year long past and current climate simulations. One set with the original land cover and original meteorology and one set where the past land cover is used in current climate conditions and vise versa. The simulated changes in surface energy balance and precipitation dynamics will be discussed.