T138 Integrated Approaches to Modeling Geochemical, Hydrological, and Ecological Processes in Watersheds II

Monday, 1 November 2010
1:30 PM - 5:30 PM
Room 207

LANDSLIDE DISTURBANCE AND SCALE IN WATERSHED STUDIES OF FLUVIAL CARBON AND NUTRIENT YIELDS

Paper number: T138-6
Program: Technical Sessions
Time: 2:50 PM – 3:05 PM

Authors

Abstract

To understand the influence of spatial and temporal scales remains a great challenge in the field of hydrology and watershed studies. We seek to learn how watersheds function as we examine the relationship between landslide disturbance and stream biogeochemistry. We are working across multiple scales in sets of watersheds all located in the Sierra de las Minas, Guatemala, a mountain range flanked by active fault systems, on the north by the Polochic system and on the south, the Motagua fault system, a left lateral transform fault that forms the boundary between the North American and Caribbean plates. Our study examined trends in chemical yields and stream discharge for different watershed size classes (sub-basin and local catchments) and within seven sets of nested watersheds of the Sierra de las Minas. We sampled river water, soils, sediments, and suspended matter and gauged streams in watersheds at a variety of sizes and scales as part of a larger ecological study of the role of landslides in mobilizing carbon and controlling chemical yields of nutrients and major elements in watersheds. In June and July 2009, we conducted field work in three river basins on the southern slope of the mountain range where extensive landsliding occurred after Hurricane Mitch (1998). Samples were analyzed for major ions, dissolved organic carbon, nutrients, particulate organic carbon and total nitrogen, and stable isotopes of carbon.

A strong positive correlation exists between watershed area and measured stream discharge. For all sites (n = 27), R2 was 0.870 (Pearson’s correlation coefficient, p < 0.001). Analysis of just the 12 sub-basin watersheds (2.4–24 km2) also showed a strong positive correlation with discharge (R2 = 0.714, p < 0.001). The correlation was less strong for the 9 local basins (0.05–3.37 km2) but still significant (R2 = 0.428, p = 0.04). At the local scale, heterogeneity confounds any significant trend between watershed area and landslide density or chemical yields. At the sub-basin level, chemical yields had stronger positive correlations with landslide density than with watershed area. However, any effect of landsliding is obscured by increases in watershed size within the nested watersheds as the strong negative trend between landslide density and watershed area define the trends in correlations with chemical yields.