Nonlinear Ecosystem Responses to Variable Precipitation Timing and Magnitude Using a Low-dimensional Modeling Framework

The Kansas River Basin in the central Great Plains of the United States is a unique region characterized by an east-west precipitation gradient and a complex assemblage of ecosystems exhibiting spatial and temporal heterogeneity. It is one of the centers of agriculture in North America and may be greatly impacted by climate change. To better characterize the nature of variable precipitation and the ecosystem response, we present a low-dimensional framework for assessing the impacts of precipitation timing and magnitude on water and carbon fluxes, with particular focus on temporal sensitivity to inputs, seasonality in water flux dynamics, and the nonlinear interactions between ecosystem functioning and soil moisture, evapotranspiration, and carbon assimilation. Results show variation in seasonal sensitivity to precipitation, with varying relationships between soil moisture, potential evapotranspiration, and precipitation timing and magnitude, in addition to the daily mean and variance of water and carbon fluxes. In addition, we examine the IPCC GCM A1B climate change scenario forecasted precipitation and temperature patterns and relate these to implications for grassland, forest, and agriculture land classes. This is important for understanding the nonlinearity inherent to water and carbon cycling in the Kansas River Basin ecosystems and provides assessment of the magnitude and extent that climate change may have on the functionality of different locations.