17 January 1997 Improvements of plant parameter estimations with hyperspectral data compared to multispectral data
Author Affiliations +
The information content of optical remote sensing data of broad band multispectral sensors is well known. Many investigations were conducted which analyze the correlations between TM-bands or indices derived from these bands with plant parameters like LAI, biomass and water content. Hyperspectral data of imaging spectrometers assess the reflectance properties of plant canopies in much more detail. To which extent this additional spectral information improves the determination of plant parameters is still a task of intensive research. This task is analyzed on the basis of different imaging spectrometer data, field spectrometer measurements, and agricultural ground truth acquired over a farming area in the Upper Rhine valley. In addition to multispectral approaches further spectral analyses, which parameterize the red edge position and water absorption features at 1.0 micrometers , are performed and the results are compared to NDVI. It is shown (1) that in comparison to the NDVI the red edge position is more sensitive to biomass changes, (2) that hyperspectral data are needed for the determination of the red edge position (3) that modeling of the water absorption allows the extraction of information on plant water content, (4) that the relations between NDVI and the water absorption feature are species dependent, and (5) that absorption features of cellulose in the SWIR can be connected with different maturing stages of vegetation.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Heike Bach, Heike Bach, Wolfram Mauser, Wolfram Mauser, } "Improvements of plant parameter estimations with hyperspectral data compared to multispectral data", Proc. SPIE 2959, Remote Sensing of Vegetation and Sea, (17 January 1997); doi: 10.1117/12.264281; https://doi.org/10.1117/12.264281

Back to Top