AN OVERVIEW OF HURRICANE HORTENSE AND ITS AFTERMATH

Shawn P. Bennett*
National Weather Service

1. INTRODUCTION

Hurricane Hortense made landfall on the southwest corner of the island of Puerto Rico on september 10, 1996 at around 0530 utc. Hurricane Hortense drenched eastern Puerto Rico with up to 23 inches of rain as it battered the south coast of the island from ponce to cabo rojo with category one hurricane force winds. This paper will present a chronology of events based on surface observations, WSR-88D and satellite observations of Hurricane Hortense as it moved slowly through the northeastern caribbean causing a 100-year flood in Puerto Rico. Several photographs taken for the official damage survey are presented to illustrate the devastation wrought by the effects of the flood.

2. HURRICANE LANDFALL IN PUERTO RICO
Hurricane Hortense developed as tropical depression on September 3rd and into the second category four hurricane of the season on september 13th north of the Bahama Islands. figure 1 shows the track Hurricane Hortense took as it crossed Puerto Rico. Hugo in 1989 was the last hurricane to make landfall (i.e the eye crossing the coast at some point) in Puerto Rico.

Figure 1. Hurricane Hortense track map derrived from WSR-88D fixes on 10 sep 96 0400-0900 utc

In 1995, Hurricanes Luis and Marilyn came very close to Puerto Rico, as did Bertha in July, 1996, all of them swept the island with tropical storm force winds and rains as they tracked north of the northeastern portion of the commonwealth.

3. A turning point

Avila (1996) explains that as Hurricane Hortense moved into the eastern caribbean sea forecast track model guidance at the Tropical Prediction Center/National Hurricane Center (TPC/NHC) indicated that the storm should intensify into a hurricane, but brush by south of Puerto Rico and head toward landfall on the southeast coast of the Dominican Republic on the Island of Hispaniola. Avila (1996) showed an array of official forecast track models indicating a trajectory towards the southeast coast of the Dominican Republic. Most of the model tracks and official forecast tracks were to the left of the actual trajectory taken by Hurricane Hortense. Avila (1996) reported that, although none of the model or official forecast tracks captured the turn to the north toward Puerto Rico, which began about 0400 utc on 10 sep 96, the forecast error was much smaller than the most recent 10-year average of forecast errors.

4. Critical forecast decision made easier by WSR-88D observations

Fortunately, for Puerto Rico the WSR-88D located near cayey, Puerto Rico was operating during the passage of hurricane hortense and provided both the NWSFO San Juan and TPC/NHC Miami with observations that were crucial to amending the official hurricane watches and warnings. A hurricane watch and tropical storm warning was posted at 2100 utc for the British and U.S. Virgin Islands and Puerto Rico on 7 Sep 96, while Hurricane Hortense was over the Leeward Islands. However, the hurricane watch for Puerto Rico was discontinued on 8 sep 96 (although the tropical storm warning remained in effect) due to unfavorable wind shear conditions in the atmosphere. This led to a general sentiment in Puerto Rico (based on personal media interviews) that Hurricane Hortense would not pose a major threat to the island, although TPC/NHC had indicated low confidence in their intensity forecasts. Hurricane Hortense began to intensify over the eastern caribbean after the wind shear conditions became more favorable leading to a hurricane warning for Puerto Rico 14 hours before landfall. Figure 2 and figure 3 show Tropical Storm Hortense beginning to intensify into a hurricane. Both figures show spiral rain bands beginning to sweep over the islands.

Figure 2. Composite Reflectivity Image from the Puerto Rico WSR-88D located at Cayey, 2.2 NM pixel and 248 NM range resolution, on 8 Sep 96 at 1808 UTC.

Figure 3. Composite Reflectivity Image from the Puerto Rico WSR-88D located at Cayey, 2.2 NM pixel and 248 NM range resolution, on 8 Sep 96 at 1941 UTC.

The intensification of Hurricane Hortense is apparent when comparing figure 3 to figure 2. an increase in the core reflectivity (note: largest reflectivity signature seen SE of Puerto Rico) from 30 dbz (ref. fig. 2) to 45 dbz (ref. fig. 3) has occurred, as well as the formation of what appears to be an eye wall (note: arc shape to core reflectivity) on the north side of the center of circulation. a little over 24 hours later at 2246 UTC 9 sep 96 NWSFO San Juan made its first hurricane eye fix based on observations made by the WSR-88D. Figure 4 shows Hurricane Hortense just after landfall in Puerto Rico with the eye of the hurricane over the municipality of Guanica.

Figure 4. Base Reflectivity Image from the Puerto Rico WSR-88D located at Cayey, .54 NM pixel and 124 NM range resolution, on 10 Sep 96 at 0643 UTC.

5. Extremely heavy rains create devastating floods and mud slides

Hurricane Hortense drenched eastern Puerto Rico with over 24 inches of cumulative rainfall during its passage on 9-10 Sep 96. Figure 5 shows that 616 mm (24.6 inches) of rain fell near the headwaters of the Rio Grande de Loiza watershed. notice that much of eastern Puerto Rico received over 500 mm (20 inches).

Figure 5. 24-hour rainfall totals recorded at 73 USGS Stations during Hurricane Hortense, 9-10 Sep 96. Note: 600mm=24in, 500mm=20in, 400mm= 16in

These extremely heavy rains led to numerous mud slides and record flooding which resulted the loss of 18 lives, 3 persons are still missing, and hundreds of millions of dollars in damages. Much of eastern Puerto Rico was declared a federal disaster area. Figure 6 shows the first visible satellite image of Hurricane Hortense at 1115 UTC 10 Sep 96.

Figure 6. Goes-8 Visible Satellite Image of Hurricane Hortense at 715 am 10 Sep 96 on the morning after landfall near Guanica

The eye of the hurricane is in the mona passage between Puerto Rico and the Dominican Republic. A strong rain band can be seen entering the island near Ponce dumping its load of rain on top of an already saturated island exacerbating the flooding conditions.

Figure 7 shows a house near Guayama, Puerto Rico along the Guamani river in which fours persons from the same family lost their lives as flood waters raged through the windows and doors tossing its victims into the torrent and drowning them.

Figure 7. Four people in one family died at this site along the Guamani River near Guayama in southeast Puerto Rico, making it the location of the single worst loss of life during Hurricane Hortense. Photo: Israel Matos

7. Concluding Remarks

Hurricane Hortense was the worst natural disaster to affect Puerto Rico since Hurricane Hugo in 1989. The forecast and warning experience with Hurricane Hortense shows that with observations from the newly installed WSR-88D we had the incontrovertible evidence available to confirm that the Tropical Storm Hortense was intensifying into a hurricane and taking a turn to toward Puerto Rico. These data were of immense value in making the critical decision to post a hurricane warning for Puerto Rico at 1500 UTCon 9 sep 96 in the face of contrary model guidance. thus, the NWSFO San Juan and TPC/NHC Miami were able to provide the Commonwealth of Puerto Rico with more precise, accurate, and timely hurricane forecasts, warnings, and local statements because of the doppler radar observations from the Puerto Rico WSR-88D.

Without the observations from the WSR-88D the performance of our jobs would have been severely limited. the WSR-74S, which is located at the Luis Munoz Marin International airport about ½ mile from the northeastern coast of Puerto Rico, cannot provide the wealth of critical information the WSR-88D provides, such as wind velocity, precipitation, and storm morphology algorithms, nor can it observe very well storms which approach from the south and southeast because of beam blockage at low elevation angles.

REFERENCES

AVILA, L.A., 1996: HURRICANE HORTENSE, 3-16 SEPTEMBER 1996, PRELIMINARY REPORT, NOAA/NWS/TPC/NHC, MIAMI, FL, 19 PP.

For the Preliminary Report of Hortense from the National Hurricane Center click here

*CORRESPONDING AUTHOR E-MAIL ADDRESS: SHAWN P. BENNETT, SCIENCE AND OPERATIONS OFFICER, NATIONAL WEATHER SERVICE.