Over the last decade, a specific type of multiple linear regression has been used by Dr. Gray and his team to make forecasts separately for each of several parameters of tropical storm activity in the northern Atlantic Basin (Gray et al. 1992). The least absolute deviation (LAD), rather than the least squared deviation (where "deviation" represents the error), is used as the criterion to develop the prediction model for these forecasts. LAD regression is discussed in Gray et al. (1993). Forecasts for each year's summer/fall storm season are made in late November of the preceding calendar year, in April, June, and finally in early August. Forecast parameters include (1) named storms, (2) named storm days, (3) hurricanes, (4) hurricane days, (5) intense hurricanes, (6) intense hurricane days, (7) hurricane destruction potential, (8) net tropical cyclone activity, and (9) maximum potential destruction. The predictors used, and the skill expected, at each of the three forecast times change as opportunities for higher skill emerge. For example, prediction of the ENSO condition for the storm season has been incorporated into the regression equation (Gray et al. 1994b), as has a set of Atlantic Ocean regional predictors (Landsea et al. 1997). The following list identifies the predictor clusters, their expected influence on Atlantic tropical storm activity, and their status regarding the 1997 storm season. More detail on these influences is found in Gray et al. (1993, 1994a).

(1) The Quasi-Biennial Oscillation (QBO) at 50 and 30 mb in the northern tropics expected at the onset of the hurricane season: Westerly QBO winds enhance storm activity, while easterly winds suppress it (Gray et al. 1992). During fall 1997 the QBO will be westerly, which is expected to enhance storm activity.

(2) El Niño/Southern Oscillation (ENSO): Warm east-central equatorial Pacific sea surface temperature, (SST), or the warm phase of ENSO, reduces storm activity, while anomalously cool SST enhances it. Going into fall we currently have a strong positive SST anomaly in the Niño 3.4 region. This is expected to suppress hurricane activity.

(3) African rainfall: Intense hurricane activity is enhanced when the Western Sahel receives above normal rainfall in June-July immediately preceding the current hurricane season, and when the Gulf of Guinea region had been wet during the August-November period of the previous year. Conversely, activity is suppressed when precipitation in those two regions /periods is below average (Landsea et al. 1993). This June-July 1997 features weak drought conditions (-0.4 standard deviations, or SDs) in the western Sahel. Gulf of Guinea rainfall for August through November of 1995 was also slightly below average (-0.4 SDs). The net effect of these indicators is an inhibiting influence on intense hurricane activity in 1997.

(4) West Africa west-to-east surface pressure and temperature gradients: Above average west-to-east surface pressure and (often associated) east-to-west surface temperature gradients from February to May are associated with enhanced hurricane activity later that year. For February through May of this year, the combined index of these two gradients was 0.6 SDs, indicating an enhancing influence on storm activity.

(5) Caribbean basin sea level pressure anomaly (SLPA) and upper tropospheric (200 mb) zonal wind anomaly (ZWA): Negative anomalies of either one of these in Jun-Jul weakly imply enhanced storm activity, while positives weakly associate with reduced activity. For Jun-Jul 1997 the SLPA was very near its average, while ZWA was somewhat higher than normal (suppressing storm activity) The net result is a slight suppression of 1997 storms.

(6) Atlantic Ocean Regional Predictors (MR, SSTA-MATL, SSTA-TATL, SSTA-SATL): When the northeast Atlantic (30-40N, 20-30W; near Azores) subtropical ridge (MR) is weaker than normal in March, seasonal hurricane activity is enhanced (Landsea et. al 1997). The SLP anomaly in March this year was negative (-.85 SDs), indicating enhanced storm activity for 1997. The north Atlantic SST anomalies in two regions (MATL: 30-50N, 10-30W; TATL: 6-22N, 18-80W) in the May to June period is positively correlated with storm activity several months later through the enhancement of deep oceanic convection. This year the anomalies in these regions were positive (0.76C and 0.30C, respectively), encouraging storm activity for 1997. The south Atlantic SST anomalies during May to June in the region 0-22S, 10E-35W, called SATL, is negatively correlated with storm activity. This year a -0.50C anomaly was observed, indicating enhanced activity. Thus, all four regional Atlantic predictors point to an active 1997 season.

The LAD multiple regression predictions made in early August are shown in Table 1. The forecasts are shown both as objective results of the LAD regression equations (middle column) for the period beginning in August, and as a final forecast (second to last column) which includes a qualitative human/intuitive adjustment to the August-onward objective forecast before adding in the storm counts already observed before August. The mean values over the 1950-95 period are shown in the last column. The first column shows expected skill for the forecasts in terms of percent variance explained. The first coefficient is based on the development (dependent) sample, and is thus an overestimate of the skill expected for a forecast for a future season. The second coefficient, based on resampling simulations, is an estimate of the skill expected on truly independent data (Mielke et al. 1995). The number of predictors used ranges from 3 to 8 out of the pool of 14 candidates (composing the broad predictor clusters described above), depending on the predictand. By economizing on the number of predictors, the penalty for overfitting on the development sample is minimized.

Near-average tropical storm activity is predicted for 1997. This is a slight drop from the three earlier forecasts for 1997 (all of which were consistent), based largely on the unexpectedly strong, 1982/83-like El Niño conditions that developed this spring and summer. The best analog years to 1997 are 1953, 1957, and 1969.

Table 1. Predicted Atlantic tropical storm historical hindcast skills and forecasts for the 1997 season, as of early June and early August 1997. Skill expected for independent (real-time) forecasts made in August (shown after the slash in bold in column 1) has been estimated using resampling simulations. These are lower than the hindcasting skill levels (shown before the slash) which were obtained using a partial cross-validation design. Some human judgement was used to modify the "After 1 Aug" objective forecast to the "Observed to Aug 1" to obtain the total 1997 final forecast. The June forecast is shown in column 4 on the left side of the > sign; the August forecast is shown after the > sign.
ATLANTIC TROPICAL CYCLONE PARAMETER	1 Skill (% variance)	2 Observed to 1 Aug	3 After 1 Aug Objctv Fcst	4 Total 1997, Final Fcst	5 1950-90 Mean
Named storms	.47/.36	4	8.5	11>11	9.3
Named storm days	.63/.42	10	51.2	55>45	46.9
Hurricanes	.51/.37	2	6.6	7>6	5.8
Hurricane days	.69/.48	2	26.4	25>20	23.7
Intense hurricanes	.64/.50	0	2.7	3>2	2.2
Intense hurricane days	.58/.48	0	8.2	5>3	4.7
Hurricane destruction potential	.72/.54	4	88	75>60	70.6
Net tropical cyclone activity (% avg)	.71/.54	18	96	110>100	100%
Maximum potential destruction	.68/.53	12	58.5	70>60	61.7

 
 
 
 
 
 
 

Gray, W.M., C.W. Landsea, P.W. Mielke, and K.J. Berry, 1992: Predicting Atlantic seasonal hurricane activity 6-11 months in advance. Wea. Forecasting, 7, 440-455.

Gray, W.M., C.W. Landsea, P. Mielke and K. Berry, 1993: Predicting Atlantic basin seasonal tropical cyclone activity by 1 August. Wea. Forecasting, 8, 73-86.

Gray, W.M., C.W. Landsea, P. Mielke and K. Berry, 1994a: Predicting Atlantic basin seasonal tropical cyclone activity by 1 June. Wea. Forecasting, 9, 103-115.

Gray, W.M., J.D. Sheaffer, P.W. Mielke, K.J. Berry and J.A. Knaff, 1994b: Predicting ENSO 9-14 months in advance. Proceedings of the 18th Annual Climate Diagnostics Workshop, Boulder, Colorado, November 1-5, 1993, 390-393.

Landsea, C.W., W.M. Gray, P.W. Mielke and K.J. Berry, 1993: Predictability of seasonal Sahelian rainfall by 1 December of the previous year and 1 June of the current year. Preprints, 20th Conference on Hurricane and Tropical Meteorology, AMS, San Antonio, Texas, 473-476.

Landsea, C.W., W.M. Gray, K.J. Berry and P.W. Mielke, Jr., 1997: Revised Atlantic basin seasonal tropical cyclone prediction methods for 1 June and 1 August forecast dates. Wea. Forecasting, 12, in preparation.

Mielke, P.W., K.J. Berry, C.W. Landsea and W.M. Gray, 1995: Artificial skill and validation in meteorological forecasting. Wea. Forecasting, 11, 153-169.