A new global reference evapotranspiration dataset for FEWS NET: Opportunities in monitoring food security and famine early warning

In traditional hydroclimatological monitoring of food security, drought analysis has been bedevilled by poor treatment of atmospheric evaporative demand (the “thirst of the atmosphere”), which physically represents the demand side of drought. In general, either long-term mean evaporative demand has been used, leading to underestimation of the severity of drought events, or it has been poorly parameterized by temperature-based models, which oversimplify forcing and lead to the appearance of ever-increasing “droughtiness” with warming. Similar complications arise in the use of such products in monitoring crop water use.

To improve the treatment of evaporative demand by the Famine Early Warning Systems Network (FEWS NET), NOAA has developed a new global reference evapotranspiration (ET₀) reanalysis that uses the UN Food and Agriculture Organization formulation (FAO-56) of Penman-Monteith ET₀ and is forced by MERRA phase 2 (MERRA2) meteorological and radiative drivers.

The NOAA ET₀ reanalysis provides daily data from January 1, 1980 to the near-present at a resolution of 0.5° latitude x 0.625° longitude. The reanalysis is verified against station-based data across southern Africa, a region presenting both significant challenges regarding hydroclimatic variability and observational quantity and quality and significant potential benefits to food-insecure populations. We also verified globally against spatially distributed ET₀ derived from two reanalyses–the Global Data Assimilation System and Princeton Global Forcing, and these verifications produced similar results, yet demonstrated wide regional and seasonal differences.

In this presentation, we place the new reanalysis within the context of FEWS NET operations, and present use cases that verify the operational applicability of the reanalysis in long-established drought, famine, crop- and pastoral-stress metrics, and in predictability assessments of drought forecasts. We also describe how new users can access the dataset.