Skip to main content
Land Cover and Land Use Feedbacks to Development of Flash Drought
Project Start Date
05/01/2025
Project End Date
05/05/2028
Grant Number
24-LCLUC24_2-0029

Team Members:

Person Name Person role on project Affiliation
Nicholas Parazoo Principal Investigator Jet Propulsion Laboratory, Pasadena , USA
Mahmoud Osman Co-Investigator NASA GSFC / UMD ESSIC, Greenbelt    , USA
Benjamin Zaitchik Co-Investigator Johns Hopkins University, Baltimore    , USA
Gary Doran Co-Investigator Jet Propulsion Laboratory, Pasadena    , USA
Abstract

Flash droughts have been responsible for some of the most damaging droughts in the U.S. in the past two decades. The rapid emergence and onset of land drying and vegetation stress associated with flash drought is difficult to predict and monitor, and often results in significant damage to natural and managed vegetation, which has direct and immediate impacts to natural resources, food supplies, and the economy. Recent observational and modeling evidence indicates a stronger than expected influence of vegetation-atmosphere feedbacks on the development and amplification of flash drought, however the role of land cover and land use practices such as irrigation on flash drought severity have not been fully investigated and remain poorly understood. 

We propose a direct insertion of land cover type and irrigation data in Digital Twin numerical experiments to determine the impact of land cover and land use (LCLU) on vegetation-atmosphere feedbacks and flash drought development in the U.S. Great Plains. We will apply the NASA Unified Weather Research and Forecasting (NU-WRF) coupled regional model to simulate land-atmosphere feedbacks and conduct sensitivity experiments to determine the impact of land cover type, irrigation fraction, irrigation strategy, and initial conditions on the onset and amplification of flash drought. We will also apply random forest modeling with Shapley Additive exPlanations (SHAP) to rank the primary sources of drought predictability, providing quantitative insight on vegetation vs atmospheric mechanisms. We target a subset of flash drought case studies, including the 2011, 2012, and 2017 events in U.S. Great Plains. We will design our prototype Land Digital Twin to examine a range of scenarios for exploring land use impacts on short range weather conditions (atmospheric temperature, aridity, boundary layer growth, cloud development, and precipitation) and flash drought development. The proposed effort builds on recent work by the PI and Co-I by providing a more detailed examination of LCLU feedbacks to drought. Ultimately, we aim to better inform decision-making under forecasted drought by improving society drought preparedness through changes in land management strategies. 

Our main objectives are to (1) Develop a prototype land digital twin for LCLU impacts on flash drought, (2) Perform and evaluate LCLU sensitivity experiments, and (3) Identify dominant atmospheric and LCLU flash drought drivers across different case studies and irrigation scenarios. 

This project will determine LCLU impacts on short-range and sub-seasonal weather using direct insertion of land cover data from NASA (MODIS) and irrigation data from the Food and Agriculture Organization of the United Nations and a NASA coupled weather model (NU-WRF). As such, our project addresses the Weather element of the Land Use for Digital Twins (LU4DT) Sub-Element: Incorporation of land-use datasets as boundary conditions in regional short-range weather forecast models and evaluation of the impact on the quality of forecasts.