Land Surface Modeling Applications for Famine Early Warning

Abstract:

AGU 2015 Fall Meeting

Session ID#: 7598

Remote Sensing Applications for Water Resources Management

Land Surface Modeling Applications for Famine Early Warning

James Verdin, USGS EROS

Christa Peters-Lidard, NASA GSFC

Amy McNally, NASA GSFC, UMD/ESSIC

Kristi Arsenault, NASA GSFC, SAIC

Shugong Wang, NASA GSFC, SAIC

Sujay Kumar, NASA GSFC, SAIC

Shrad Shukla, UCSB

Chris Funk, USGS EROS

Greg Fall, NOAA

Logan Karsten, NOAA, UCAR

Famine early warning has traditionally required close monitoring of agro-climatological conditions, putting them in historical context, and projecting them forward to anticipate end-of-season outcomes. In recent years, it has become necessary to factor in the effects of a changing climate as well. There has also been a growing appreciation of the linkage between food security and water availability. In 2009, Famine Early Warning Systems Network (FEWS NET) science partners began developing land surface modeling (LSM) applications to address these needs. With support from the NASA Applied Sciences Program, an instance of the Land Information System (LIS) was developed to specifically support FEWS NET. A simple crop water balance model (GeoWRSI) traditionally used by FEWS NET took its place alongside the Noah land surface model and the latest version of the Variable Infiltration Capacity (VIC) model, and LIS data readers were developed for FEWS NET precipitation forcings (NOAA’s RFE and USGS/UCSB’s CHIRPS). The resulting system was successfully used to monitor and project soil moisture conditions in the Horn of Africa, foretelling poor crop outcomes in the OND 2013 and MAM 2014 seasons. In parallel, NOAA created another instance of LIS to monitor snow water resources in Afghanistan, which are an early indicator of water availability for irrigation and crop production. These successes have been followed by investment in LSM implementations to track and project water availability in Sub-Saharan Africa and Yemen, work that is now underway. Adoption of LSM and data assimilation technology has enabled FEWS NET to take greater advantage of remote sensing observations to robustly estimate key agro-climatological states, like soil moisture and snow water equivalent, building confidence in our understanding of conditions in data sparse regions of the world.