Skip to main content
Improving Numerical Weather Forecasts with More Representative Land Surface Modeling: Implementation of Updated Satellite-Derived Land Cover and Land Use Change Maps
Project Start Date
05/05/2025
Project End Date
05/05/2028

Team Members:

Person Name Person role on project Affiliation
Matthew Johnson Principal Investigator NASA Ames Research Center , Moffett Field , USA
Jia Jung Co-Investigator Bay Area Environmental Research Institute, NASA Ames Research Center  Research Scientist , Moffett Field, USA
Justin Fain Co-Investigator Bay Area Environmental Research Ins,  Moffett Field     , USA
Weile Wang Co-Investigator NASA Ames Research Center, Moffett Field, United States
Taejin Park Collaborator Bay Area Environmental Research Institute, Sunnyvale,
Jessica McCarty Collaborator Miami University, Oxford, USA
Abstract

This project will evaluate the: a) compounding impacts of different landscape transformation processes driving land cover land use change (LCLUC) on local- to regional-scale meteorology and b) improvements in numerical weather forecast model accuracy when driven with more representative LCLU information derived from satellites. We will focus on three major anthropogenic and natural landscape transformation processes: 1) urban expansion, 2) agriculture and irrigation, and 3) wildfire and timber harvesting. Recent research has demonstrated that implementing satellite-derived, high spatiotemporal LCLU data into numerical weather forecast systems such as the Weather Research and Forecasting (WRF) model, replacing outdated and coarser spatial scale data sets, results in improved meteorological forecast skill. We hypothesize that lessons learned from more robust testing of numerical weather model forecasting skill when applying multiple high spatiotemporal resolution, satellite-derived maps of LCLU will create new knowledge necessary for developing NASA's Land-Earth System Digital Twin (L-ESDT). The improved representation of LCLU, and methods of how to quantify the impacts of LCLUC on meteorology, gained from this project can be applied in the L-ESDT development which will aid users in various applications and decision-making processes which have societal implications, such as the impact of LCLU on local- to regional-scale weather/climate (e.g., urban heat effects, humidity levels, precipitation perturbations). 

Nearly 30% of LCLU and land surface characteristics in the contiguous United States (CONUS) have been altered by anthropogenic activities largely from resource production (e.g., cropland, timber production and harvesting), built-up urban areas, irrigation, and lands used for recreation. Furthermore, wildland fires over the last few decades have significantly impacted land cover characteristics throughout the CONUS. LCLU, and associated short- and long-term changes, have direct impact on local- to regional-scale meteorology. However, the impacts of LCLU on weather and short-term climate are challenging to assess given there is no tool available which is specifically designed to test the impact of different LCLUC processes on weather. The L-ESDT being developed by NASA would be such a tool. A primary goal of this project is to demonstrate the increased accuracy of numerical weather forecasts when supplementing coarse spatial scale and outdated LCLU information currently used in numerical weather forecast models such as WRF with updated and improved satellite-derived maps. This will provide critical information for the development of NASA's L-ESDT in regard to what LCLU maps are applied in the system and the ability for users to test the impact of LCLUC on meteorology. The project will apply numerous LCLU data products from satellites (e.g., Landsat, MODIS, VIIRS, SMAP, Sentinel-1, GEDI) in high spatial resolution (1 km × 1 km) WRF simulations to demonstrate the best methods for: 1) improving forecast skill when using specific LCLU maps and 2) applying time series of LCLU maps, and altered states of these LCLU variables, for users of the L-ESDT to quantify the impact of LCLUC on weather and short-term climate.

This proposal is responsive to the ROSES-24 A.2 Land Cover Land Use Change solicitation and will provide critical information for the development of NASA's L-ESDT. This proposal will also address some of the Most Important Science and Applications Questions and Objectives in the Decadal Survey 2017-2027 and NASA's Earth Science Plan 2020-2024. The proposal team is interdisciplinary and has extensive experience in numerical weather modeling, satellite-derived LCLU data products, and the implementation of high spatial resolution LCLU maps in numerical weather models to improve forecast skill.