Research On Spectral Reflectance Of Suspended River Water

It has been recognized and accepted as a high-tech monitoring method to use remote sensing technology to monitor water quality. Compared with the traditional method which sets water quality observation points to monitor water for the point to the surface, this new method is absolutely a faster and more efficient method to monitor large-scale water. But, at present, there are some limitations for traditional river water quality monitoring by satellite data because of the restrictions on geometric resolution and spectral resolution, and using satellite data to water quality parameters inversion rarely consider the influence of the depth of water, so this paper argues it a shortcoming.Considering the limitations of the general satellite data to monitor river water quality, this paper takes ChengDu JinJiang river as the research site, the suspension as the research object, to explore the influence of different concentration of suspended solids and different water depth on water spectrum feature with water spectral data collected by imaging spectrometer. The main research content and research results are as follows:(1) To collect suspended solids concentration data and spectral dataAccording to water quality sampling points setting by the experiment, collect water sample on site and measure suspended solids concentration in the laboratory. In this paper, the innovative whiteboard method is presented to collect suspended solids spectrum data, The spectral data of water has been collected in different concentration of suspended solids and different water depth.(2) To process water spectral dataWe select ten points on the white ceramic of the imaging spectrometer image. Reflectivity value of ceramic board of reflectance on the full wave imaging spectrometer is obtained through mean value treatment. Considering the influence of noise, finally, we choose the 400-900 nm band reflectance values. Because of imaging spectrometer with high spectral resolution, the noise interfere the white ceramic reflectance value, we conducted a data compression. Finally, the smooth spectral curve of water in different concentration and depth of suspended solids has been got during 400-900nm wavelength.(3) Spectral reflectance characteristics under different concentrations of suspended solidsWith the increase of concentration of suspended solids in water whiteboard reflection are generally reduced, but the damping is different. With bimodality reflectance curve, when suspended matter concentration is low, the first reflection peak in the green band, with the increase of the concentration, the first reflection peak move to the red band, The second reflection peak position has not changed basic. On the infrared wavelengths, suspended solids concentration high reflectivity was lower than those of suspended matter concentration decreases slowly.(4) Spectral reflectance characteristics of suspended solids under different depthWith the increase of depth, white reflectance values gradually decrease; different concentrations of suspended solids, the sensitivity to depth is inconsistent. The most sensitive is high concentration, and the least sensitive is low concentration. Different concentrations’ sensitivity to the depth is different on different band.(5) To establish the. model between different suspension concentration and depth and reflectivityBased on the single band model, band ratio model and differential spectral reflectance model, we did correlation analysis of reflectance and suspended sediment concentration, to determine the most sensitive bands of their respective,The model were simulated according to the sensitive bands, results show that the best fitting effect under different conditions of the two items; The same method,we did correlation analysis of reflectance and the depth. Finally, we fit a empirical formula between the depth and reflectivity. The results show that:for high concentration and medium concentration of water, the best fitting effect is power function, low concentration of water the best fitting effect is linear function.