Greater Horn of Africa Climate Outlook Forum
Forecast Interpretation
Chris Funk and Gideon Kinyodah Galu, USGS/FEWS NET
Greg Husak and Joel Michaelsen, University of California,
Santa Barbara
As a demonstration of the newly developed Forecast Interpretation
Tool developed by Greg Husak and Joel Michaelsen at the University
of California, Santa Barbara, we vectorized the
GHACOF9
consensus forecast map ( Figure
1
). This vectorized map and a set of gamma distribution parameters
for the March-April-May season for the Greater Horn of Africa were
then combined (Figure
2
) with the Forecast Interpretation Tool to prepare a new
set of conditional gamma distribution parameters. The climatological
and conditional gamma distribution parameters were then used to generate
maps of climatological and conditional (forecast) median rainfall
(Figures 3
and 4
). The difference between these two fields (
Figure 5
) represents the likely change in rainfall implied by the COF9 forecast.
The conditional distribution fields were then used to examine three
questions applicable to the livestock, water resources and health sectors.
The first question was: “what is the chance, given the COF9 consensus,
of receiving more than 200 mm of rainfall in March-April-May?”
Figure 6
attempts to answer this question.
Figure 7
presents similar results, but looking at the chance of rainfall exceeding
700 mm. We leave the interpretation of these products to regional
experts.
The Forecast Interpretation Tool and gridded monthly rainfall
dataset may obtained free of charge by emailing Greg Husak (husak@geog.ucsb.edu),
Chris Funk (chris@geog.ucsb.edu) or Gideon Galu (ggalu@fews.net).
Figure 3 - Climatological Median Rainfall - March-April-May accumulation
[mm/season]
Figure 4 - Forecast Median Rainfall - March-April-May accumulation
[mm/season]
Figure 5 - Forecast Rainfall Anomaly - March-April-May accumulation
[mm/season]
Figure 6 - Forecast Chance of Exceeding 200mm - March-April-May
accumulation [%]
Figure 7 - Forecast Chance of Exceeding 700mm - March-April-May
accumulation [%]