A Study On The Remote Sensing Monitoring Of Submerged Aquatic Vegetation

The reestablishment of submerged aquatic vegetation has been recognized as a key measure for restoring eutrophied lakes or rivers. However, to map the distribution and monitor the growth and dynamics of SAV on a large scale is very labor intensive and timeconsuming due to the restriction of the water environment. The appropriate technique is required to monitoring the SAV distribution and growth dynamic on a large scale.This study was designed to investigate the spectral responses of several common submerged aquatic plants in Shanghai, using a FieldSpecTM Pro JR. Field Portable Spectroradiometer. A number of questions were considered.1) The spectral characteristics of a submerged plant Vallisneria spiralis with varied coverage/biomass were measured in the laboratory, and the functions between the coverage/biomass and reflectance were built.2) The spectral characteristics of a submerged plant V. spiralis with varied coverage/biomass were measured in the Middle Lake section of a field-scale constructed wetland, and the functions between the coverage/biomass/water depth and reflectance were built. Simultaneously, the spectral characteristics of different species SAV(Vallisneria spiralis, Elodea Canadensis, Myriophyllum spicatum and Potamogeton crispus) were analyzed to find the optimal bands to identify different species SAV.3) The spectral characteristics of a submerged plant M. spicatum with varied coverage were measured with a ground sensor/radiometer in the constructed lake of "Chongming International Wetland Park", Shanghai, in September 2005, November 2005 and April 2006. The Vegetable Indexes (NDVI and RVI), Spectral Indexes (EDFR and ddRE) and Spectral Parameters ("red edge" and "green peak") were calculated to build the function between the coverage/Vegetable Indexes (VI)/ Spectral Indexes (SI) and reflectance. The spectral characteristics of M. spicatum in different seasons were discussed by analyzing the changes of VI/SI position and Spectral Parameters.The results showed: 1) The reflectance rate of V. spiralis and M. spicatum increased with their increasing coverage/biomass and this were exhibited both at the visible band (500-650 nm) and the near infrared band (700-900 nm). 2) The spectral differences of different species were mainly showed in near infrared band (700-1000 nm). 3) A negative correlation (P < 0.05) was found at roughly the wavelength range of 400-900 nm, with the highest correlation coefficient around 574 nm(r = -0.9527). 4) RVI and NDVI of M. spicatum decreased with their decreasing coverage, and a clear linear relationship could be found for the NDVI (R~2=0.7785) and RVI (R~2=0.7073). 5) M. spicatum has the special spectral characteristics in different seasons, and EGFR, ddRE, "red edge" and "green peak" were changed regularly with season. 6) The primary differences in the spectral signature between the laboratory and field experiments were water environment and the fundus. The algal chlorophyll and other suspended contents in the field water could probably change the reflectance at the NIR.Simultaneously, the atmospheric correction of the QuickBird image of the constructed lake got in September 2005 was performed to realize the conversion of DN to ground reflectance. Using the results of function between the coverage of M. spicatum and the reflectance rate, the reflectance image was converted into the map of SAV coverage with 4 bands. The fourth band (760-900nm, NIR) was confirmed to be the best band to predict the coverage by comparing the predicted coverage of 20 sites that had not been selected to build the functions with the real measured coverage. On this foundation, the coverage of SAV was divvied into different grades, and the map of SAV coverage was turned into the distribution map of different SAV coverage grades by visual interpretation.In this research, the field spectral signature was combined with a QuickBird image, the function between the biophysical parameters of SAV and their spectral signature were used to estimate the SAV coverage on a large scale by selected and combined the optimal bands. The implications of this observation, in terms of the ability of hyperspectral remote sensing to estimate and monitor the distribution and dynamics of SAV on a large scale were discussed.