Fugui Wang and Yi Jun Xu
Hurricanes frequently cause extensive damage to forests. Such disturbances not only affect the forest industry and wildlife habitats directly, but cause long-term influences on forest succession, site productivity and site drainage.
Hurricane Katrina, the third strongest storm to hit the U.S. coast during the past 100 years, made its landfall on August 29, 2005, near the border of Louisiana and Mississippi. With sustained winds of 121 miles per hour and a storm surge level up to 23 feet, the hurricane caused catastrophic damage in the coastal region of these two states.
Large areas of the region’s forests across the wind swath were severely damaged. It is important to accurately identify the disturbed areas, disturbance severities and the factors that have contributed to the spatial patterns caused by the hurricane. This information can be especially crucial for resources managers and researchers to take short-term actions on salvage harvesting, wood industry and habitat protection and to assess the long-term environmental impacts and forest ecosystem recovery. With much of the Katrina-affected region under permanent or seasonal inundation on the coastal lowlands, however, a full-scale, ground-based assessment was technically difficult and economically challenging.
Using a combination of remote sensing techniques and geographical information system (GIS) applications, LSU AgCenter scientists conducted a comprehensive assessment of Hurricane Katrina damage to forests in the Lower Pearl River Valley – Washington and St. Tammany parishes in Louisiana and Hancock and Pearl River counties in Mississippi. They analyzed the relationships among hurricane disturbance gradient and forest characteristics and site conditions and developed models to predict vulnerabilities of the forests to wind damage. The biotic and environmental factors studied included forest type, forest coverage, stand density, soil great group, elevation, slope, exposure and stream buffer zone.
Katrina damage and spatial distributions
Before Hurricane Katrina, the percentage of coverage and spatial distribution of forests in the Lower Pearl River Valley region varied with forest types (Figure 1). Wetland hardwood forests and upland evergreen forests were two major forest types. The upland forests spread out across the entire region, while the wetland forests concentrated on the Pearl River and Bogue Chitto River floodplains (Figure 2). Urban forests were mainly distributed around towns and cities along the north shore of Lake Pontchartrain and Bay St. Louis. The shrub/scrub land cover appeared in large patches in St. Tammany Parish.
Hurricane Katrina damaged 60 percent of the total forest land in the Lower Pearl River Valley, creating a mosaic of landscape patches including undisturbed, moderately disturbed and highly disturbed forests in the region. A large portion of the damaged forests were clustered in the Lower Pearl River Valley and south Hancock County, Miss. (Figure 3). Lightly disturbed forests were randomly scattered as small patches in the region, whereas undisturbed forests clustered in south St. Tammany Parish and northwestern Hancock County.
Forest stand conditions and disturbance intensities
Among five forest types, wetland forests suffered the greatest damage with 73 percent of the area disturbed (Figure 4). The upland evergreen forests and shrub/scrub vegetation showed the lowest area of disturbance (less than 40 percent) with a significantly lower disturbance severity. Urban forest cover in the Lower Pearl River Basin region was small (Figure 1) but suffered great damage under Katrina’s wind swath.
In wetland forests, more than half of the stands were composed of water tupelo, swamp tupelo, blackgum, sweetgum, baldcypress and various oaks. Among these species, cypress and tupelo appeared to be more resistant to strong winds because of the presence of buttressed boles. The shallow rooting of oak trees may have contributed to their susceptibility to hurricane wind. In addition to wind damage, Hurricane Katrina pushed Gulf waters up toward the inner Pearl River, causing extensive flooding and saltwater intrusion in the area for several weeks. Unlike the upland forests, the bottomland hardwoods on the floodplains were affected by both the hurricane’s wind and the flooding caused by the storm surge.
Site conditions and disturbance intensities
Researchers found a close association between the hurricane disturbance intensities and the local site conditions. Forests growing on moderately developed wet soils were found to be most susceptible to hurricane damage, with a forest damage rate of more than 80 percent. By contrast, forests growing on well-developed wet soils appeared to be most resistant to Katrina’s winds, with more than half of the total forests on this soil group undisturbed. It was also found that forests on weakly developed wet soils were susceptible to the hurricane winds, whereas forests on well-developed, low-fertility soils suffered only moderate and light disturbance.
Elevation in the region seemed to have had an effect on hurricane disturbance. Except for the coastal areas at or below 24 meters (79 feet) in elevation, the percentage of disturbance was generally lower at higher elevations (Figure 5). The greatest disturbance occurred at an elevation range between 24-48 meters (79-158 feet). Forests in the areas facing northeast to southwest showed a higher probability of the hurricane disturbance than those in the areas facing west to northwest. Greater exposure to the hurricane winds resulted in higher chances of forest damage.
Forest damage intensity differed among slopes. Forests on 1-degree slopes appeared to be highly susceptible to the hurricane wind. Percentages of highly disturbed forests decreased from 42 percent on flat land to 8.7 percent at a 6- degree slope. At a zero-degree slope, almost the same amount of forest land areas was highly and moderately disturbed. However, as the slope increased, the percentage of the highly disturbed forests was considerably lower than moderately disturbed.
Forests adjacent to the streams and rivers were found more severely damaged by the hurricane (Figure 6), indicating a high susceptibility of bottomland forests to the hurricane damage. Comparing the decreasing trend of percentages of the forests highly disturbed with increasing distance to the river channels, percentages of the forests moderately and lightly disturbed displayed slightly escalating trends.
Beyond assessment on Katrina’s damage, this study elucidates the usefulness of remote sensing and GIS technique combined with statistics modeling in assessing the large-scale risks of hurricanes to coastal forests. The model developed in the study can ultimately be used to predict potential wind damage to forests in South Louisiana.
(This article was published in the winter 2008 issue of Louisiana Agriculture.)