论文部分内容阅读
The aim of this paper is to investigate the feasibility of using Landsat TM data to retrieve leaf area index (LAI). To get a LAI retrieval model based ground reflectance and vegetation index, detailed field data were collected in the study area of eastern China, dominated by bamboo, tea plant and greengage. Plant canopy reflectance of Landsat TM wavelength bands has been inversed using software of 6S. LAI is an important ecological parameter. In this paper, atmospheric corrected Landsat TM imagery was utilized to calculate different vegetation indices (VI), such as simple ratio vegetation index (SR), shortwave infrared modified simple ratio (MSR), and normalized difference vegetation index (NDVI). Data of 53 samples of LAI were measured by LAI-2000 (LI-COR) in the study area. LAI was modeled based on different reflectances of bands and different vegetation indices from Landsat TM and LAI samples data. There are certainly correlations between LAI and the reflectance of TM3, TM4, TM5 and TM7. The best model through analyzing the results is LAI = 1.2097*MSR + 0.4741 using the method of regression analysis. The result shows that the correlation coefficient R2 is 0.5157, and average accuracy is 85.75%. However, whether the model of this paper is suitable for application in subtropics needs to be verified in the future.
The aim of this paper is to investigate the feasibility of using Landsat ™ data to retrieve leaf area index (LAI). To get a LAI retrieval model based ground reflectance and vegetation index, detailed field data were collected in the study area of eastern China, dominated by bamboo, tea plant and greengage. Plant canopy reflectance of Landsat ™ wavelength bands has been inversed using software of 6S. LAI is an important ecological parameter. In this paper, atmospheric corrected Landsat ™ imagery was utilized to calculate different vegetation indices (VI such as simple ratio vegetation index (SR), shortwave infrared modified simple ratio (MSR), and normalized difference vegetation index (NDVI). Data of 53 samples of LAI were measured by LAI-2000 area. LAI was modeled based on different reflectances of bands and different vegetation indices from Landsat ™ and LAI samples data. There are certainly correlations between LAI and the reflectance of TM3, TM4, TM5 and TM7. The best model through analyzing the results is LAI = 1.2097 * MSR + 0.4741 using the method of regression analysis. The result shows that the correlation coefficient R2 is 0.5157 and the average accuracy is 85.75%. However, whether the model of this paper is suitable for application in subtropics needs to be verified in the future.