Three remote sensing studies in Shasta Valley, California.

Abstract

The three remote sensing studies presented in this thesis were undertaken to determine how Landsat imagery could be utilized in the study of land-use in the Shasta Valley. Landsat Multispectral Scanner (MSS), Thematic Mapper (TM), and Enhanced Thematic Mapper Plus (ETM+) imagery was used to observe the effects of the 1987-1992 drought on water coverage in the Shasta Valley, to create a land-cover map of the farming community of Gazelle, California, and to examine the impact of logging on Goosenest, California. Boolean processing of three MSS datasets showed that the 1974 subscene, selected to illustrate long-term average climatic conditions, had the highest water coverage at 8.4%; the 1992 subscene, acquired at the height of the drought, had the lowest water coverage at 2.3%; and the 1985 subscene, acquired after the short-term 1976-1977 drought and before the start of the 1987-1992 drought, had a water coverage of5.6%. A thematic map of the farming region around Gazelle, California was created using unsupervised classification of ETM+ bands 3, 4,5. An attempt was made to predict which parcels of land had been logged by Weed Lumber Company/Long-Bell in Townships 44-45N, Range 3W on Goosenest. After determining that this was impossible due to the complex logging history of the region, change detection study of the state of vegetation on Goosenest was conducted using three years of MSS data. It was found that cleared lands (or lands with a low vegetation cover) recovered between the years 1974-1985 while more vegetated lands were cleared (or had a reduction in vegetation) during the years 1985-1992. Much of the change in vegetative cover was seen to be a product of the climate. However, logging patterns could clearly be detected. A more thorough investigation of past logging on Goosenest might improve interpretation of change detection using satellite imagery.