Thursday, 13 July 2006
56-8

Classification of the Owens Dry Lake Playa Surface Using Satellite Imagery and Unique Surface Characterization Methods.

Mica H. Heilmann1, Brad L. Inman1, Joel E. Kimmelshue1, Brian Schmid1, John B. Dickey1, Richard Coles1, and Richard Harasick2. (1) CH2M HILL, INC, 2485 Natomas Park Drive, Suite 600, Sacramento, CA 95833, (2) LADWP, P.O. Box 111, Los Angeles, CA 90051

The historic Owens Lake, located in the Great Basin Desert east of the Sierra Nevada Mountains in California, receded after the Owens River was diverted upstream, first to irrigate farmland, and later into the Los Angeles Aqueduct. Over time, the exposed Owens Lake playa (Playa) has become the leading source of PM10 (particulate matter < 10 microns in diameter) in the United States. In cooperation with Great Basin Unified Air Pollution Control District (GBUAPCD), the Los Angeles Department of Water and Power (LADWP) has constructed the majority of a 7,770-hectare dust control area, to be completed by the end of 2006. As LADWP approaches completion of the project, focus is turning to evaluating emissions from remaining, uncontrolled Playa surfaces. The majority of exposed Playa soils have a crusted surface horizon consisting primarily of evaporite salts. This crust is vulnerable to erosive forces of wind and saltating sand. Crust characteristics vary widely across the Playa and also vary seasonally with weather cycles and relative humidity. Understanding how the crust varies across the Playa, and how it responds to environmental conditions, is critical to determine emission patterns. As a result, a surface survey was conducted on 7,690 hectares of the remaining, uncontrolled Playa to specifically classify crust conditions that affect surface vulnerability to wind erosion. This survey also developed a baseline condition, from which monitoring of seasonal changes in crust conditions will be conducted. Remotely sensed images were a principal resource for this effort. This survey was an ideal application for remote sensing because (1) a conventional survey of the vast project area was not feasible due to accessibility constraints in some areas; (2) conditions of the exposed surface were the focus of the survey; (3) topography was nearly flat and few objects existed on the Playa that would obstruct view of the surface or provide benchmarks during field observations; (4) a library of historical, high-resolution satellite imagery was available; and (5) several characteristics, including overflow areas, abrupt changes of crust condition or reflectance, are often visible on satellite images and are significant (in terms of erodibility) and repeatable (across images and through time) surface mapping indicators. Preliminary survey polygons were delineated based on visible range spectra in QuickBird and Landsat satellite imagery. Delineations were based on twenty-two images, dating from 1994 to 2005. QuickBird imagery was the primary resource for surface polygon delineation because of its higher image resolution (sub-meter pixel). Landsat images (30-meter pixel) were used to assess changes in large-scale features over time (such as major overflow areas as a result of stormflow onto the Playa). Delineations were verified and refined by extensive ground observation and characterization. Once polygons were established, surfaces within them were characterized and classified in the field. A unique methodology was specifically developed for description of Playa surface conditions. Descriptive parameters were selected to support surface erodibility evaluation. Surfaces were classified into surface mapping units based on the prevalence of each suite of conditions on the Playa, anticipated wind erodibility and logical relationships among characteristics. In general, satellite delineations were consistent with transitions visible on the ground. Despite known temporal variations in crust conditions, results suggest that mapping units generally remain distinct from one another over the long term. The resulting surface condition map is now being employed in the assessment of Playa emission rates.

Back to 1.5A Diffuse Reflectance Spectroscopy, Soil Sensing, Remote Sensing and Image Analysis - Theater
Back to WCSS

Back to The 18th World Congress of Soil Science (July 9-15, 2006)