PDF Version of this ArticleNosing Out Future Landslides
by James T. Spartz
 |
| In 2004, a landslide at Peeks Creek in Macon County, NC, killed five people and destroyed 15 buildings. (photo by Leif Skoogfors, Federal Emergency Management Agency) |
In the mountains of the Southern Appalachian region, landslides are already common occurrences during major storms. Increases in the frequency of extreme rainfall events predicted under climate change could ramp up the risk of landslides in an area where development and roads steadily crawl up the steep hillsides, replacing trees. In western North Carolina, the North Carolina Geological Survey has documented over 2,700 landslides and landslide deposits since its landslide hazard-mapping program started in 2009.
Landslides aren't just about blocked roads and property damage. Case in point: Peeks Creek, North Carolina.
In late 2004, Hurricanes Frances and Ivan tracked over western North Carolina. Ivan dumped about 30 inches of rainfall in 10 days, triggering over 140 landslides. The largest landslide at Peeks Creek in Macon County, NC, killed five people, injured two, and destroyed 15 buildings.
Concern about these disasters sparked scientists at the SRS Coweeta Hydrologic Laboratory (Coweeta) to look more closely at the role of vegetation type and topographic position in relation to landslide incidents. They were already starting to suspect that the dominance of the woody evergreen shrubs rhododendron and mountain laurel in the subcanopy of Southern Appalachian forests, when in some topographic positions, might have a destabilizing effect on mountainside soils.
In 2008, SRS research ecologist and Coweeta project leader Jim Vose, along with SRS scientist Chelcy Ford and University of North Carolina at Chapel Hill colleagues T.C. Hales and Larry Band, set up an experiment at Coweeta to assess whether the presence of rhododendron, combined with topographic position, could play a part in heightened landslide risk.
In mountainous landscapes such as the Southern Appalachians, the roots of plants help reinforce the stability of soils, reducing landslide potential. Previous research found that root strength can vary greatly across plant varieties such as trees and shrubs and by topographic position—whether a group of trees or shrubs are on a convex hilltop (nose), steep slope, or concave valley floor (hollow).
"Rhododendron roots have been found to be especially weak, which reduces soil stability in the areas where they dominate," says Vose. "The 2004 Peeks Creek event, which covered about 11,960 square yards, or nearly 2.5 acres, began in an area dominated by rhododendron."
To investigate the relationship between root strength and soil stability, researchers dug pits downslope from 15 individual trees on the Coweeta site. The locations of trees varied from noses to hollows. Root and soil samples were taken below native oak, eastern hemlock, birch, tulip poplar, hickory, and other common tree species. The research team also tested areas below rhododendron thickets.
 |
| SRS researchers are studying whether the presence of rhododendron, combined with topographic position, plays a part in heightened landslide risk. (photo by Robert L. Anderson, USDA Forest Service, courtesy of Forestry Images) |
"We found that rhododendrons had the shallowest, weakest roots, suggesting that the presence of this species lowers the ability of soil to hold together under severe rain conditions in some hollows," says Vose. "Since debris flows usually start in the hollows, those dominated by rhododendron could represent a heightened hazard for landslides."
Since 2008, Coweeta scientists have extended their original experiment on root strength and are working with the North Carolina Geological Survey landslide risk-mapping project to incorporate their findings and climate change factors to improve landslide risk modeling.
"Landslide risk models have tended to simplify the factors they consider because the data are not available," says Vose. "This is especially true when it comes to soil and root strength properties." The new, dynamic landslide risk model will meld refined root cohesion and soil strength models with research-based simulations of water, carbon, and nutrient cycles.
The model will also integrate daily weather information, allowing researchers to better assess the vulnerability of the Southern Appalachian study region to alterations in precipitation under different climate change scenarios.
This information will be incorporated into the Regional Hydro-Ecologic Simulation System (RHESSys) previously developed for Coweeta and used to generate landslide risk estimates based on forest patterns and change over time combined with the circulation of soil and groundwater throughout the Coweeta catchment. This affords Coweeta scientists greater precision in testing the relationships between roots, topographic position, and landslide risk.
