Luquillo Experimental Forest
Recovery from Hurricanes
Hurricanes account for much of the spatial and temporal variation in the characteristics of light penetration, air and soil temperature, soil moisture, forest structure, composition, productivity and soil organic matter pools in the Luquillo Experimental Forest (LEF), Puerto Rico (Lugo, 2000; Lodge and McDowell, 1991; Scatena and Larsen, 1991; Tanner et al., 1991; Zimmerman et al., 1996). The degree of hurricane damage to forests is determined by the characteristics of the hurricane (i.e., intensity, frequency and scale), landscape topography, soils (e.g., texture, structure), and ecosystem structure and composition both above and below ground (e.g., age/size combination, susceptibility of various species) (Boose et al., 1994; Everham, 1996; Tanner et al, 1991; Zimmerman et al., 1996). There are, however, few studies that have examined spatial patterns of damage to forests by hurricanes of various frequency, intensity and scales, and forest recovery (structure, composition and biomass) after hurricanes over the entire Luquillo Mountains. In this study, we used a spatially-explicit hurricane disturbance and recovery model, RECOVER (Everham, 1996), to simulate hurricane-induced mortality in the Luquillo Mountains, Puerto Rico.
Our results (table 1) indicate that 1) the simulated maximum tree mortalities in the Luquillo Experimental Forest (LEF) are 26.15 % (category 5), 18.82 % (category 4, e.g., Hurricane Hugo in 1989), 14.12 % (category 3), 9.41 % (category 2) and 4.18 % (category 1), respectively, with the mean mortalities for the entire forest as 17.76 %, 8.11 %, 6.09 %, 4.40 % and 2.41 %, respectively; 2) ridge tops tend to expose to hurricanes and suffer more damage than valleys; 3) hurricanes occur more often in lowlands, north part of the forest and track to the south, and northwest direction of the LEF; and 4) given a hurricane that hits through the entire forest, higher elevations tend to have a higher mortality than lower elevations (e.g., the correlation coefficient between mortality and elevation for simulated hurricane #1 is 0.7631).