Estimation of leaf area index in natural deciduous broad-beaved forests using Landsat TM data
2001
Hoshi, N. (Niigata Univ. (Japan)) | Tatsuhara, S. | Abe, N.
The objective of this study is to estimate the leaf area index (LAI) in natural deciduous broad-leaved forests using Landsat-5 TM data. The LAI was measured by a litter trap method in the basins of Takigashira bog and Mt. Hisonokura in Niigata Prefecture. The correlations of LAI with the digital number of each band, brightness, greenness, and wetness of the tasseled cap concept, normalized difference vegetation index (NDVI), and ratio vegetation index (RVI) were examined. The highest correlation was observed between NDVI and LAI. The relationship between NDVI and LAI was showed by an exponential formula with a standard error of 6.6%. The relationship between LAI and dry weight was obtained by a simple linear regression. The correlation coefficient between estimates from NDVI with the two formulas and measurements was 0.576, which was significant at the 1% level. The LAI map in the study area was drawn with the LAI-NDVI relationship. The average LAI was 4.15 and 5.00 in Takigashira bog basin and Mt. Hisonokura basin, respectively.
Показать больше [+] Меньше [-]