Drivers of Atmospheric Evaporative Demand during African Droughts

Seeking to advance the practice of famine early warning across sub-Saharan Africa we illuminate past drivers of high-impact droughts to gain a better understanding of the evaporative processes involved in drought dynamics. Atmospheric evaporative demand (ETo) is often used to estimate plant water balance and drought impacts to vegetation, and previously demonstrated linkages between precipitation, temperature, and ETo need to be better understood. This work is timely as new data streams will enable near-real-time monitoring of ETo and incorporation of ETo forecasts into seasonal outlooks for African growing seasons.

For historical droughts during major growing seasons in sub-Saharan Africa, we evaluate ETo and identify main drivers for drought cases--identified based on below-normal precipitation during the wettest three months of the growing season--and contrast these with the ETo drivers that dominate in wetter years (we also consider droughts triggered by above normal ETo). Our focus is on regions of Africa where adequate precipitation is important for productive agriculture and pastoral activities and where evaporative demand might exacerbate moisture limitations, e.g., the Sahel and semi-arid East and Southern Africa. It is expected that important ETo drivers are partly connected with precipitation-related processes (e.g., cloud cover and radiation, temperature, humidity) but that there are variations between regions and events. Factors less directly related to precipitation processes (e.g. wind speed) could also play an important role. The goal here is to provide a generalized understanding of what aspects of evaporative demand historically have posed an additional hazard to plant stress and how precipitation outcomes are responsible for the ETo drivers. In addition, we explore whether there have been discernible changes through time in regard to ETo drivers during below-normal precipitation seasons.

Upper and lower terciles of CHIRPS precipitation are used to identify anomalous dry and wet cases. The ETo dataset spans the 1980-near present period and is calculated following ASCE's formulation of Penman-Monteith method driven by daily temperature, humidity, wind speed, and solar radiation from NASA’s MERRA-2; this data is the basis for the Evaporative Demand Drought Index (EDDI). For this analysis, daily ETo and drivers are aggregated to monthly and seasonal time steps. ETo drivers are identified with the decomposition method from Hobbins et al. (2016), which considers the anomalies in meteorological variables and the sensitivity of ETo to each of them.

Hobbins, M. T., Wood, A., McEvoy, D. J., Huntington, J. L., Morton, C., Anderson, M., & Hain, C. (2016). The evaporative demand drought index. Part I: Linking drought evolution to variations in evaporative demand. Journal of Hydrometeorology, 17(6), 1745-1761.