Development and Evaluation of Local Season-Ahead Precipitation Predictions in the Blue Nile Basin, Ethiopia

Background
Ethiopia’s predominately rain-fed agricultural society is greatly impacted by seasonal and inter-annual variability in precipitation, with 75% of precipitation occurring during one June-September rainy season (JJAS). Skillful season-ahead predictions conditioned on local and large-scale hydro-climate variables can provide valuable knowledge to farmers and reservoir operators, enabling informed water resource allocation and management decisions.

Objective
In Ethiopia, the potential for advancing agriculture and hydropower management, and subsequently economic growth, is substantial. Yet, evidence suggests a weak adoption of prediction information by sectoral audiences. To address common critiques of skill, scale, and uncertainty, probabilistic forecasts are developed at various temporal and spatial scales for the Finchaa hydropower dam and the Koga agricultural scheme in an attempt to promote uptake and application.

Initial Results
Results indicate that prediction skill is possible at the local scale, particularly for statistical models. While the use of dynamical models is potentially appealing from an operational perspective, the modest skill and model complexity are clear drawbacks. PCR methods have been utilized for decades in seasonal climate prediction, given the ability to condition relationships to variables of interest at the local scale using the historical record without the need to fully capture complex climate dynamics at the global or regional scale. In a region with many teleconnections to global climate phenomena, statistical methods may be better positioned to capture smaller-scale inter-annual factors that modulate variability. 

Dissemination
Alexander, S, Block, P., and Wu, S. Development and Evaluation of Season-Ahead Precipitation Predictions for Sectoral Management in the Blue Nile Basin, Ethiopia. American Geophysical Union (AGU) Fall Meeting, New Orleans, LA, December 2017.

Figure 1. Local statistical model framework

Figure 2. Observed JJAS precipitation and statistically downscaled NMME (dynamic model) predicted precipitation issued June 1.

Figure 3. Observed JJAS precipitation (black) and probabilistic JJAS precipitation predictions results from locally-tailored statistical PCR model issued June 1.