<!doctype html public "-//w3c//dtd
html 4.0 transitional//en">EXPERIMENTAL FORECAST OF EAST AFRICAN RAINFALL FOR
OCTOBER-DECEMBER 2003
Seasonal
Forecasting Group, Ocean Applications, Met Office, Bracknell, UK
INTRODUCTION
The Met Office is conducting research into the
effects of sea surface temperatures and other climatic variables on tropical
rainfall. As part of this research, experimental seasonal rainfall forecasts
have been made for the Sahel and adjacent regions in tropical NW Africa since
1986, and for the Nordeste region of Brazil since 1987. Using similar
statistical methods, forecasts for tropical East Africa October-December
rainfall (the 'short rains') have been issued since 1994 and appear in previous
September issues of this bulletin .
The region covered by the East Africa prediction is
between 5N and 15S and between 30E and the Indian Ocean coast. Forecasts for E
Africa were produced using statistical methods and dynamically using the Met
Office's GLObal SEAsonal (GLOSEA) forecast model. This year, there is strong disagreement between the statistical
and dynamical forecast and hence low forecast confidence so we have decided not
to include forecasts for higher resolution (2.5o latitude x 3.75
o longitude rectangular) sub-regions as we have done in recent years.
The statistical forecast is made by using linear
regression and discriminant analysis techniques, with three indices of global
sea surface temperature (SST) anomaly patterns (Shown in Colman 2002 http://grads.iges.org/ellfb/Sep02/Colman/colman.htm
). The forecast model is derived from historical rainfall and SST information.
GLOSEA is a version of the HadCM3 coupled
ocean-atmosphere general circulation model
modified for making seasonal predictions. The GLOSEA forecast was
produced as a 40-member ensemble of predictions initialised with atmosphere and
oceanic conditions observed in early August.
GLOSEA hindcasts produced as part of the DEMETER project (see www.ecmwf.int/research/demeter
for more about DEMETER) were used to correct the GLOSEA forecast output for
model bias and convert the forecast into probability format.
Forecasts are expressed as probabilities for 5
equi-probable categories (quints) for the whole region and as probabilities of
3 equi-probable categories (terciles) for the sub-regions. The categories are equi-probable over the
1961-1990 climatology period.
The category boundaries (as percentages of mean rainfall)
are:
|
VeryDry/Dry |
Dry/Average |
Average/Wet |
Wet /Very Wet |
|
74% |
86% |
102% |
124% |
FORECAST SKILL
PERFORMANCE OF TRIAL FORECASTS
The statistical and dynamical forecasts were tested
using trial forecasts over the period 1948 to 1997 and 1959 to 2000
respectively. The assessment measure used is correlation.
Statistical linear regression forecasts were assessed
using a method where a trial forecast is made for each year using a regression
equation calculated using data for the remaining years. This assessment
provides a good measure of forecast skill from minimal data.
Statistical forecast skill correlation=0.50
DEMETER GLOSEA skill correlation=0.38
These correlations are statistically significant at
the 5% level.
PERFORMANCE OF REAL TIME EMPIRICAL
FORECASTS
Forecasts have been made for this region since 1994. The
forecasts for 1994 and 1995 were strongly influenced by above and below average
SST in the NW Pacific respectively and the forecasts for 1997 and 1998 where
influenced by the 1997 El Nino and the 1998 La Nina events. The observed
rainfalls for 2000 and 2001 were quite close to the dry-average and very
dry-dry boundaries respectively so the forecasts for these seasons were as
accurate as they could be.
|
Year |
1994 |
1995 |
1996 |
1997 |
1998 |
1999 |
2000 |
2001 |
|
Forecast Category |
Very Wet |
Dry |
Average |
Wet |
Dry or Average |
Dry |
Dry or Average |
Dry or Very Dry |
|
Observed Category |
Wet |
Dry |
Very Dry |
Very Wet |
Very Dry |
Average |
Dry |
Very Dry |
Note: The categories used for the 1994-1998 forecasts
are based on a 1951-1980 climatology. For the 1999 and later forecasts,
categories based on the 1961-1990 climatology are used as 1961-1990 is the
accepted WMO standard climatology period and is used by most forecasters. The
1961-1990 rainfall average is 104% of the 1951-1980 average.
FORECASTS
FOR THE 2003 SEASON
STATISTICAL FORECAST
Below average SST in the NW Pacific near Japan and in
the tropical E Pacific are favouring below average rainfall in E Africa this
year. The regression forecast is % of the 1961-1990 average and is in
the AVERAGE category
The discriminant analysis technique gives the following
probabilities for the 5 (1961-1990 based) categories:
|
Very
Dry |
Dry |
Average |
Wet |
Very Wet |
|
0.30 |
0.36 |
0.22 |
0.04 |
0.09 |
GLOSEA DYNAMICAL FORECAST
Based on the performance of DEMETER GLOSEA hindcasts of rainfall from 1959 to 2000,
the dynamical forecast is presented as
probabilities of 5 (1961-1990 based) observed rainfall categories which are:
|
Very
Dry |
Dry |
Average |
Wet |
Very Wet |
|
0.15 |
0.08 |
0.08 |
0.13 |
0.56 |
OVERALL BEST ESTIMATE: This
year, the dynamical forecast is favouring the VERY WET category whilst the
empirical forecasts favour the DRY category, a clear disagreement. Both forecasts give relatively low
probabilities to the AVERAGE category so we have ruled out a compromise. Our
best estimate forecast is for the DRY category which is favoured by the
slightly more skilful statistical forecast. However, given the disagreement
between forecasts, confidence in this best estimate is very low.
References:
Mutai, C.C., Ward, M.N and Colman, A.W. Prediction of
East Africa seasonal "short rainfall" rooted in evidence for
widespread SST-forced variability during October-December. I.J.Climatol.18
975-997 (1998).
Colman, A.W.
Experimental Forecast of East African Rainfall for October-December 2002
Experimental Long Lead Bulletin Vol. 11 No 3 (2002) (http://grads.iges.org/ellfb/Sep02/Colman/colman.htm
)
Acknowledgements:
Thanks to Sarah Ineson, Pete Mclean and Richard Graham
for supplying dynamical forecast output.