🚀 go-pugleaf

     EARLY APRIL ATLANTIC BASIN SEASONAL HURRICANE PREDICTION FOR 1996
      (A year of expected somewhat above average hurricane activity)

    (Prepared for issue at the 18th Annual National Hurricane Conference,
                       Orlando, Florida, April 5, 1996)

     By

     William M. Gray (Professor of Atmospheric Science)
     Christopher W. Landsea (Post-doctoral fellow at NOAA/HRD Lab., Miami,  FL)
     Paul W. Mielke, Jr. and Kenneth J. Berry (Professors of Statistics)
     and John A. Knaff (Advanced Ph.D. Student)


   (This forecast is based on ongoing research by the authors, together
         with meteorological information through March 1996)
 

   [A more complete explanation of this forecast with figures and tables  
    is available on the World Wide Web at this URL:

        http://tropical.atmos.colostate.edu/forecasts/index.html ]  
    --- also

   Thomas Milligan, Colo. State Univ., Media Representative  
   (970-491-6432) is available to answer various questions about this forecast

 

                Department of Atmospheric Science  
                   Colorado State University  
                    Fort Collins, CO  80523   

ABSTRACT

This paper presents the authors' forecast of Atlantic tropical cyclone
activity expected to occur during 1996.  This forecast is based on a couple
of new forecast schemes recently developed by the authors.  One scheme is  
based on meteorological information through November 1995, and the other,
a newer scheme, is based on somewhat different meteorological data through
March 1996.  Both forecast schemes predict nine individual seasonal tropical
cyclone parameters.  Each scheme chooses the best of five to seven (those
resulting in the most skillful prediction) of ten forecast parameters as
predictors.  These ever evolving forecast schemes use the following as
possible predictors: forward extrapolations of the stratospheric
Quasi-Biennial Oscillation (QBO) of equatorial zonal winds, two measures
of West African rainfall, an extended range forecast of summertime El Nino
conditions, an extended range forecast of Western Sahel monsoonal rainfall,
and the October through November and March  Atlantic subtropical ridge
strength.  In addition, other forecast parameters from the North Atlantic,
the Pacific Ocean and from the Asia-Australia regions add additional
qualitative assessments.  We combine the results of both of these forecasts
along with qualitative information not explicitly contained in either
scheme to produce the final forecast values.  

Information obtained through March 1996 indicates that 1996 Atlantic
hurricane activity is likely to be somewhat greater than the average
season with 7 hurricanes (average 5.7), 11 named storms (average 9.3),
55 named storm days (average 46), 25 hurricane days (average 23), 2 intense
(category 3-4-5) hurricanes (average 2.1), 5 intense hurricane days
(average is 4.5) and a hurricane destruction potential (HDP) of 75
(average 71).  Collectively, net tropical cyclone activity is expected
to be 105 percent of the long term average.  The 1996 season should be much
less active active than the 1995 season, but more active than the average
season of the last 25 years.  

This early April forecast anticipates a greater amount of 1996 hurricane
activity than indicated by our late November forecast.  

1.  Background

A 10-page paper titled ``Discussion of Atlantic Basin Seasonal  
Hurricane Activity in 1995 and Prospects for 1996" was written on 15  
March 1996 for inclusion in the Annual National Hurricane Conference,  
Orlando Florida, pre-conference material.  This paper gives you  
background information for our new updated 5 April 1996 seasonal  
forecast.  This report gives the final number of our early April  
forecast.  A more complete explanation of this forecast with figures  
and tables is available on the World Wide Web.  See cover sheet for  
access information.  Thomas Milligan, Colo. State Univ., Media  
Representative (970-491-6432) is usually available to answer various  
questions about this forecast.

2.  Early April Atlantic Basin Hurricane Forecast for 1996

We have recently developed a new Atlantic Basin seasonal hurricane  
forecast for issuance in early April.  This early April forecast  
includes meteorological data through March.  This gives four more  
months of information than is included in our 30 November 1995  
forecast.

About half of the predictors contained in our early April forecast  
are similar to the forecast parameters of our early December forecast  
and about half are different.  Both schemes are based on 46 years of  
hindcast information (1950-1995).  The skills of the two schemes are  
comparable, and we expect to obtain the best prediction by making a  
qualitative assessment using both of these forecasts.  This early  
April assessment thus makes use of both our 30 November 1995 forecast  
and our recently developed early April forecast.  


Table 1 shows our various 1996 hurricane season forecasts.  The  
second column gives our corrected 30 November 1995 forecast.  More  
recent and improved analysis of global conditions through November  
1995 along with global circulation feature changes since November   
indicate that our 30 November 1995 forecast of storm activity for  
1996 was too low.  We found an error in one of the input parameters  
to our 30 November 1995 forecast.  We have also been able to update  
some of the other information used in this forecast. 


Table 1:  Updated statistical and adjusted Atlantic basin hurricane  
forecasts for 1996.
                                Corrected   Early    
                                30 Nov 95   April       Adjusted  
Forecast               1950-90  Statistical Statistical 5 April  
Parameter              Average   Forecast   Forecast    Forecast   
----------------------------------------------------------------
Named Storms (NS)        9.3       10.1      12.8         11  
Named Storm Days (NSD)  46.6       52.5      62.7         55  
Hurricanes (H)           5.8        6.7       9.2          7  
Hurricane Days (HD)     23.9       29.0      19.7         25   
Intense Hurricanes (IH)  2.3        2.7       2.4          2   
Intense Hurricane 
   Days (IHD)            4.7        7.7       4.6          5  
Hurricane Destruction       
   Potential (HDP)      71.2       58.9      95.6         75  
Net Tropical Cyclone       
   Activity (NTC)       100       100.8     118.2        105   
Maximum Potential 
   Destruction (MPD)   66.0        73.2      77.7         75   
-----------------------------------------------------------------

Our 5 April 1996 statistical prediction is given in the third column.   
This early April forecast also indicates a greater amount of  
hurricane activity than was originally anticipated.  Our final  
qualitative adjusted forecast is shown in the right column of Table  
1.  


We now expect a somewhat above normal hurricane season for 1996.   
This is noteworthy considering that 1995 was one of the most active  
hurricane seasons on record.  Usually there is reduced hurricane  
activity following very active seasons.  Our update is based on more  
recent data and the projections of our newly developed early April  
forecast scheme. 


Table 2 lists results of our corrected 30 November 1996 forecast and  
our newly developed early April forecast.  The last column on the  
right gives our adjusted 5 April 1996 hurricane forecast which is a  
compromise between our corrected 30 November 1995 forecast, our early  
April statistical forecast and from other qualitative information  
relating to global circulation statistics which are not explicitly  
addressed in either scheme.  This information includes Atlantic Ocean  
surface pressure and sea surface temperature anomalies, Pacific and  
Asian regional meteorological data such as the current state of the  
El Nino, Singapore 100 mb temperature anomalies, and other relevant  
information.   These will be discussed in the first author's talk on  
Friday morning. 


Table 2:  1996 Atlantic basin hurricane statistical and average  
forecast made on 30 November 1995 and revised early April forecast  
and the differences between the two.

Forecast                   30 Nov. 95   5 April 96   Increase  
Parameter                  Forecast     Forecast 
--------------------------------------------------------------
Named Storms (NS)             8            11           3  
Named Storm Days (NSD)       40            55          15  
Hurricanes (H)                5             7           2  
Hurricane Days (HD)          20            25           5   
Intense Hurricanes (IH)       2             2           0   
Intense Hurricane Days (IHD)  5             5           0  
Hurricane Destruction     
    Potential (HDP)          50            75          25  
Net Tropical Cyclone     
    Activity (NTC)           85           105          20   
Maximum Potential     
    Destruction (MPD)        55            75          20   



Thus, from the data analyzed through the end of March, we anticipate  
that 1996 will experience somewhat above average hurricane activity  
in comparison with the activity over the last 45 years.  This 1996  
season should be distinctly more active than the recent inactive  
hurricane seasons of 1991-1994 or that of the average hurricane  
season of the last 25 years.


3.  Coming Updated Forecasts 


Later in 1996 updated forecasts will be issued on 


a)  Thursday 6 June 1996} just after the official start of the  
hurricane season. 

b) Wednesday 7 August 1996} just before the start of the most active  
part of the hurricane season.


These later forecasts utilize data closer in time to the hurricane  
season and should be somewhat more accurate than our late November  
and early April forecasts.


4.  Cautionary Note

It is important that the reader appreciate that these seasonal  
forecasts are based on statistical schemes which will fail in some  
years.  These forecasts also do not specifically predict where within  
the Atlantic basin storms will strike.  Even if 1996 should prove to
be a somewhat average hurricane season, there are no assurances that
many hurricanes will strike along the US or Caribbean Basin coastline
and do much damage.  
-- 
##############################################################################
# Bill Thorson                # Research Coordinator                         #
# Dept of Atmospheric Science # Tropical Meteorology Project                 #
# Colorado State University   #                                              #