Derived data and code for: Sensitivity of primary production to precipitation across the United States (Ecology Letters)

This collection contains derived data products and code from the publication "Sensitivity of primary production to precipitation across the United States" published in Ecology Letters in 2019. Reproduction of results in the article is possible by placing all datasets here in a data directory and processing the data using the R code and Jupyter notebooks in the "Code and documentation" repository. The "Code and documentation" repository also provides detailed instructions, explanations of scripts and data products, and metadata.

Please direct any questions about the analysis, code, or data products to the corresponding author (gregmaurer@gmail.com).

Primary production, a key regulator of the global carbon cycle, is highly responsive to variations in climate. Yet, a detailed, continental-scale risk assessment of climate-related impacts on primary production is lacking. We combined 16 years of MODIS NDVI data, a remotely-sensed proxy for primary production, with observations from 1218 climate stations to derive values of ecosystem sensitivity to precipitation and aridity. For the first time, we produced an empirically-derived map of ecosystem sensitivity to climate across the conterminous United States. Over this 16-year period, annual primary production values were most sensitive to precipitation and aridity in dryland and grassland ecosystems. Century-long trends measured at the climate stations showed intensifying aridity and climatic variability in many of these sensitive regions. Dryland ecosystems in the western U.S. may be particularly vulnerable to reductions in primary production and consequent degradation of ecosystem services as climate change and variability increase in the future.