Vegetation cover and spatial pattern analysis using Landsat TM data /
The utility of Landsat TM data was investigated for landscape-level classification of rangeland vegetation and quantification of landscape pattern at three spatial scales namely Sonora Station pastures, Sonora Station and Sutton County. Two methods of ground verification, the GPS and PHOTO methods...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1996.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=743273431&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | The utility of Landsat TM data was investigated for landscape-level classification of rangeland vegetation and quantification of landscape pattern at three spatial scales namely Sonora Station pastures, Sonora Station and Sutton County. Two methods of ground verification, the GPS and PHOTO methods were used to classify the Landsat TM image and were compared for accuracy against an independent Line Intercept method. The GPS and PHOTO methods gave four woody cover classes on the Landsat TM image cover estimates of 10, 22, 46 and 54% for the GPS method, and 10, 20, 29 and 46% for the PHOTO method' The 20 and 22% woody classes from the PHOTO and GPS methods, respectively, were most dominant. Pasture woody cover estimates compared to the Line Intercept method were 82.7% and 70.5% accurate for the GPS and PHOTO methods respectively. The correlation coefficient between the GPS and PHOTO method was 0.98. The proportions of woody cover classes in various pastures, size of patches and heterogeneity of the distribution of woody classes could be quantified using the classified Landsat TM data. Livestock grazing resulted in a more heterogeneous landscape in terms of woody cover classes, with goats having the greatest effects on reducing woody cover. Patch shape, dominance and diversity indices were not affected by grazing management. Slope and aspect did not affect the distribution of woody cover classes at the Sonora Station level. At the Sutton County level the 22% woody class was most dominant. There were more bare areas compared to the Sonora Station, with an apparent classification confusion between bare areas and the 10% woody class. Dominance and diversity indices did not differ from those of the Sonora Station and average of pastures, suggesting no extent effect on the indices. There was no slope or aspect effect on the distribution of woody cover classes. Only the highest cover class seemed to have a greater distribution in the Buda limestone and Alluvium geology areas. It was concluded that Landsat TM could be used for woody vegetation cover classification, and to relate woody cover to grazing management and woody vegetation management practices. The 25m by 25m spatial resolution of the Landsat TM data was too course and of limited use in relating landscape indices to increased extent and to relate the distribution of woody cover classes to slope and aspect. |
|---|---|
| Item Description: | Vita. "Major Subject: Rangeland Ecology and Management". Landsat TM data map in pocket. |
| Physical Description: | xiii, 77 leaves : illustrations, maps ; 28 cm. Issued also on microfiche from University Microfilms Inc. |
| Bibliography: | Includes bibliographical references. |