An assessment of the value of

Issued May 2002*

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 forecasts have been made of seasonal rainfall for the Sahel (region 1) for each year from 1986 onwards. Since 1992, forecasts of seasonal rainfall have also been made for a slightly redefined rectangular Sahel (region 2, 15W to 37.5E and 12.5N to 17.5N), for an area south of the Sahel (region 3, 7.5W to 33.75E, 10N to 12.5N), and for an area extending further south to the coast (region 4, approximately 7.5W to 7.5E, 5N to 10N). The four regions are labelled in figure 1a.

The statistical forecasting techniques are based on March and April sea surface temperature (SST) anomaly patterns. Further details can be found in Folland et al, 1991. Several forecasts have been made using different versions of each technique, and they have been averaged together with dynamical forecasts produced using the Met Office Atmosphere Global Circulation Model (AGCM) and persistence forecasts (observed rainfall for last year’s season) to obtain the forecasts shown below in figure 1.

The forecast period for regions 2-4 is July-September. For region 1 annual rainfall is predicted, though most of the rain in this region falls during July- September. For forecasting purposes, the predicted rainfall indices are categorised into quints which are equi-probable over 1961-1990. The 5 quints are referred to as Very Dry, Dry, Average, Wet and Very Wet. In table 1 the quints are defined as percentages of 1961-1990 average

This year we have used recently run AGCM hindcasts for 1982-2000 to help calibrate our dynamical forecasts.

SEA SURFACE TEMPERATURE ANOMALIES

The SST indices used to predict rainfall in N Africa represent regional and global scale anomaly patterns. Most important are tropical Pacific and Atlantic anomalies, and interhemispheric differences in anomalies.

Warm SST in the Tropical NW Pacific and the north-west Indian Ocean favours above average rainfall in regions 1,2 and 3 this year. Warm SST in the South Atlantic near the African coast favours above average rainfall in region 4. The interhemispheric contrast in SST is weak favouring near average rainfall this year in regions 1 2 and 3.

THE PREDICTION SYSTEM

The forecasts are weighted combinations (table 4) of statistical forecasts, dynamical forecasts and persistence (last year’s observed seasonal rainfall). The statistical best estimate forecasts are produced by linear regression with SST indices as predictors. Statistical probability forecasts are calculated from the same SST indices using linear discriminant analysis.

Prior to 2001, only predictors (a) and (b) were used. Adding predictors (c), (d) and (e) was found to improve trial forecast skill over 1951-2000 (Table 3). In particular adding predictor (e) improved skill in predicting region 1 and 2 rainfall variability between 1981-2000. Predictors (a) and (b) are poor at predicting variability over this period. The trial forecasts referred to in table 3 were produced using the jackknife method in which data for the forecast year and the next two subsequent years are excluded when calculating prediction equations.

The statistical forecast is a correlation skill weighted combination of methods a to e. Predictors a to e are approximately weighted 0.125, 0.125, 0.125, 0.125 and 0.5 respectively. Predictor e has a higher weight than predictors a-d since this predictor is much better at predicting the 1981-2000 seasons than predictors a-d and since forecasts from predictors a-d are quite highly correlated with each other.

The dynamical forecast was produced using the 19 level HADAM3 version of the Met Office AGCM. The forecast is based on an ensemble of 9 AGCM runs each initialised with slightly different atmospheric conditions observed over the period May 8-9th and forced using the mean SST anomalies observed over the preceding 4 weeks which are assumed to persist throughout the forecast period. The AGCM ensemble was run to 5 months ahead (up to 30th September). Further information about dynamical ensemble forecasts at the Met Office can be found on our website at www.metoffice.com/research/seasonal.

The dynamical forecast output is expressed as both deterministic forecasts and probability forecasts for the 5 quint categories. The model forecasts were calibrated using 9 member ensemble hindcasts for 19 years (1982-2000) and SST forced 10 member ensemble simulations for 1951-1981. The deterministic forecasts are produced by correcting the ensemble mean forecast for model bias as observed in model simulations and hindcasts for 1961-1990. To evaluate the dynamical forecast probabilities for 5 observed quint categories , 5 frequency distributions of observed quint categories are evaluated for sets for years when the model simulates or predicts the same category. The forecast probabilities are proportional to the mean of these frequency distributions for the 5 categories predicted by the 9 forecast members.

The forecasts are weighted to reflect the reliability of the different inputs. The ratio of weights for the statistical forecast/dynamical forecast/persistence is about 2:1:1 for each of regions 1,2 and 3. Persistence is not used for the region 4 forecast, as persistence skill is negligible for this region. For region 4 the statistical /dynamical weighting ratio is 3:2.

LAST YEAR

Last year, the DRY category was observed in regions 1,2 and 4 and the AVERAGE category was observed in region 3.

FORECAST SUMMARY

Forecasts for regions 1-4 are shown in figure 1. Weighted average deterministic forecasts are shown as percentages of the 1961-1990 average in figure 1a. In Figure 1b, the forecasts are expressed as percentage standardised units (e.g. standardised values of +100 indicate rainfalls one standard deviation above average) relative to 1961-1990 (NB. 1901-1980 for region 1 for compatibility with previous publications by the Met Office and Nicholson (1984). Quint categories are indicated in figure 1c. The skill of these weighted forecasts is indicated in fig. 1d by the trial forecast correlations with observed rainfall in the period 1951-2000. The correlations are well above the 5% significance level for all 4 regions. Probability forecasts for the 5 quint categories are shown in figure 1f-j respectively. The Relative Operating Characteristic (ROC) skill in figure 1e is a measure of the performance of these probability forecasts over the period 1951-2000. ROC scores above 60% are considered to indicate significant (5% level) skill.

There are considerable differences between the forecasts for 2002 provided by the different methods. Persistence favours the DRY category in regions 1,2 and the AVERAGE category in region 3. The statistical and dynamical methods both favour the WET category for regions 1,2 and 3 and the VERY WET category for all region 4. Confidence is MODERATE due to the agreement between dynamical and statistical forecasts.

Our best estimate forecasts are:

Region 1: WET
Region 2: WET
Region 3: WET
Region 4: VERY WET

Hence, rainfall is expected to be greater in 2002 than during the past 2 years in all regions. There is an above chance probability of a “VERY WET” category rainfall season in regions 1,3 and 4 (fig 1j).

IMPACT OF SST CHANGES BETWEEN 1^st MAY and 14^th JUNE

SST anomalies associated with north African rainfall have not changed since April in a way that would significantly alter the forecast for N African rainfall. (Northern hemisphere SST is still slightly warmer than Southern hemisphere SST, and a substantial develpment of EL Nino conditions in the Pacific is not expected during the period of this forecast). Hence our forecast remains the same as issued in May . An updated forecast will be issued to relevant National Met services and published on our website in July.

REFERENCES:

Folland, C.K., Owen, J., Ward, M.N and Colman, A.W. 1991: Prediction of seasonal rainfall in the Sahel region using empirical and dynamical methods. Journal of Forecasting, 10, 21-56.

Folland, C.K., Parker, D.E., Colman, A.W. and Washington,R. 1999: Large scale modes of Ocean Surface Temperature since the late nineteenth century. In Beyond El Nino, decadal and Interdecadal variability. Ed. A Navarra, Springer pp 75-102.

Nicholson, S.E. 1985: Sub-Saharan rainfall 1981-84. J. Clim. Appl. Met., 24, pp 1388-1391.

Figure Captions:

Table 1 Quint Boundaries (% 1961-90 Average) :

REGION	VERY-DRY / DRY	DRY/ AVERAGE	AVERAGE /WET	WET/ VERY WET
1	75	97	109	121
2	81	93	102	117
3	88	99	104	112
4	82	94	106	115

Table 2: Statistical Forecast Predictors:

Predictors	Training period	Reference
(a)Time Indices of 3 Global Scale SST EOFS	1901-2000,1951-2000	Folland et al 1991
(b)Time indices of 2 EOFs of South Atlantic SST and 1 EOF of Pacific SST	1901-2000,1951-2000
(c)Time indices of 3 Global scale EOFs of filtered SST	1901-2000,1951-2000	Folland et al. 1999
(d) May-June indices of global filtered EOFS calculated using predicted SST	1901-2000,1951-2000
(e)Time index of correlation field between March-April SST and rainfall with correlations not significant at 5% level set to 0	1981-2000

Table 3: Performance Of Trial Forecasts Using Combinations Of Predictors, 1951-2000 Measured Using Correlation Between Forecast And Observed.

Predictors	REGION 1	REGION 2	REGION 3	REGION 4
a+b (as in previous in years)	0.48	0.47	0.49	0.40
a+b+c	0.59	0.53	0.48	0.40
a+b+c+e	0.69	0.62	0.49	0.43
a+b+c+d+e	0.71	0.63	0.52	0.42
a+b+c+d+e+ dynamical	0.73	0.67	0.57	0.56
a+b+c+d+e+ dynamical+persistence	0.74	0.66	0.57	N/A

Table 4 Forecast Weights

Region	Statistical	Dynamical	Persistence
1	0.52	0.23	0.25
2	0.50	0.25	0.25
3	0.49	0.26	0.25
4	0.60	0.40	0.00

Figure 1: Predictions For 2001 And Prediction Skill For 4 North African Region. Probabilities, Skill And Regression (Standardized Units) Forecasts Are Percentages