For intense humcane strikes along the Gulf 

 Coast, the single largest factor of influence 

 is the phase of the Quasi-Bieimial 

 Oscillation (QBO; Gray 1984, Shapiro 

 1989). The QBO is an east-west oscillation 

 of stratospheric-level winds that encircle 

 the globe near the equator. In the west 

 phase, Atlantic hurricane activity is 

 enhanced, but is diminished in the east- 

 phase QBO years. The QBO is followed in 

 importance by the 200-millibar zonal winds 

 and sea-level pressure anomalies, such that 

 westedy winds and high pressures in the 

 Caribbean favor more tropical cyclone 

 landings (Knaff, 1997). With regard to 

 interannual variation, normalized United 

 States' damages are well related to ENSO, 

 such that significantly less damage occurs 

 during El Nino events and than dvuing 

 La Nina events. 



JET "^ . 



Figure 10: Winter features for La Nina. 



Lack of hurricane variation consistency in the Gulf of Mexico compared to East Coast occurrences is likely 

 caused by the dominance of local conditions over basin-wide factors that can lead to intense hurricane 

 development (Landsea et al, 1999). An example is tropical storm development along a stationary frontal 

 boundary of strongly contrasting conditions. This occurred in the case of hurricane Alicia in 1983. The 

 prominent role of local conditions is likely why there is no distinct multi-decadal variation of intense hurricanes 

 in the Gulf of Mexico (Landsea et al, 1992). 



Tropical cyclone activity in the western Gulf of Mexico during the three periods of interest to the Rincon 

 Project somewhat followed the causal/correlated patterns discussed previously. Table 1 gives tropical cyclone 

 occurrences by ENSO type and baseline (non-ENSO) years. Classification of years was according to PieLke and 

 Landsea (1999). More tropical cyclones impacted the western Gulf of Mexico in La Nina years than El Nifio 

 years, generally following the script. However, table 1 shows that six tropical cyclones impacting the western 

 Gulf occurred during the relatively low Adantic basin activity period of the 1970-80s. In the first Rincon study 

 period (1940-57), only hurricanes' Alice and Audrey affected the western Gulf of Mexico. During the second 

 study period (1958-81), 11 tropical cyclones impacted the Gulf coast. In the third study period (1982-99), four 

 hurricanes (Arlene, Charley, Frances and Bret) occurred, three during El Niiio years and the fourth (Bret) 

 during the La Nina year of 1999. 



WATERSHED PRECIPITATION ANOMALIES AND ENSO PHASE 



Table 1 shows twice as many tropical cyclones occurring in the western Gulf of Mexico during La Nina years as 

 El Nino years. The question arises as to whether Nueces River watershed precipitation is sensitive to ENSO 

 phases. For example, do La Nina months record more precipitation baseline-year months. To investigate this 

 issue, monthly precipitation for 1950-99 for Corpus Christi WSO AP, BeeviUe 5 NE, and Sabinal were stratified 

 into baseline (16 years and 4 tropical cyclones). El Nino (18 years and 4 tropical cyclones), and La Niiia years 

 (16 years and 8 tropical cyclones). All months for a particular category (ENSO, baseline) were used in analyses 

 to capture the effects on precipitation of nontropical storms as well as tropical cyclones. For each precipitation 

 gauge, the 75 percentile precipitation was determined in the base years to serve as the benchmark of 

 comparison. The analysis technique applied, similar to Gershunov and Bamett (1998), determined the 

 proportion of months of record with a precipitation amount that equaled or exceeded the benchmark. The 

 proportion determined is compared to 0.25 and the quotient is reduced by 1.0 and converted to percent 

 anomaly. 



Appendix D ♦ D-7 



