A Study On Eutrophication Evaluation Of Lake Based On Hyper-spectral Remote Sensing

Water is the source of human life and is the requirement for nearly all living feature to survival in the biosphere.There are nearly half of the world's people rely on lake,reservoir,river,for drinking,transportation,fishing,and tourism.In recent years,most of lakes and reservoirs are appeared different eutrophication,especially lakes and reservoir which with algae eutrophication and near cities such as lake Taihu and Chaohu are more hyper-trophic.So,Accurate and effective evaluation of eutrophication is of great significance to water quality research and treatment.Prediction of water quality parameters in lake water is an important condition for water quality assessment.Nitrogen and phosphorus are essential nutrients in the eutrophication process of water bodies.Nitrogen is an essential element in the formation of amino acids and other organic substances.Phosphorus provides continuous material support for cell growth and division:phosphorus is a component of nucleic acid proteins which is the important place to store genetic material and Phosphorus is a component of phospholipids which is the component of cell biofilms,and adenosine triphosphate that provides energy for plant growth and other processes.Therefore,the concentration of phosphorus has a great influence on the nutritional state of water bodies.Chlorophyll a is one of the representative substances that characterize phytoplankton:chloroplast a is the embodiment of bloom,which the concentration of chloroplast a determines to a certain extent the amount of phytoplankton content.In this paper,the measured spectral data,traditional measured concentration data and remote sensing image data are chose to establish the evaluation model of the eutrophication of Hongfeng Lake.The nutritional status of Hongfeng Lake is studied.And the contents are as follows:(1)Combining the MLP neural network model to predict the water quality parameters of Hongfeng Lake,it was found that the predicted values of some water quality parameters(total nitrogen,total phosphorus,chlorophyll a,permanganate index,dissolved oxygen,etc.)and the measured values The correlation coefficient is very high.This shows that when operating the MLP neural network to make predictions,its effect is very effective.However,because the concentration of chloroplast a in Hongfeng Lake varies greatly at different times,it is found that the average error between the predicted and the measured values of chloroplast a in Hongfeng Lake is relatively large when analyzing all sampling points at all times.Overall,MLP neural network model prediction results are relatively good and can be used for water quality parameter prediction.(2)A model for estimating the total phosphorus concentration was established based on the measured spectral data and the traditional measured data of Baihua Lake.The spectral data of other reservoirs such as Aha and Xiaoguan Reservoir were used as verification,and then the model was applied to Hongfeng Lake.It is showed that high accuracy of quadratic fitting model of the first-order differential spectrum at 823 nm with correlation coefficient R~2 at0.9433.When verified,the total phosphorus concentration verified by most sampling points is within the standard value range.The distribution map of total phosphorus concentration in Hongfeng Lake is drawn,and the results are not much different from the traditional measured data.Therefore,it shows that the model can be used to estimate total phosphorus concentration to some extent.(3)Using GF-1 WFV2 data after radiation correction,atmospheric correction,and geometric correction to invert the concentration of chloroplast a in Hongfeng Lake,it was found that the concentration of chloroplast a in Hongfeng Lake was generally higher in summer and lower in winter.The concentration of chlorophyll a in the lake area with a large population such as Beihu Lake District and Houwu District is higher than in other areas.(4)When analyzing the nutritional status of Hongfeng Lake using the comprehensive nutritional status index method,the results are as follows:In summer,Hongfeng Lake has a high degree of eutrophication,almost half of the lake area is in a mild eutrophication state,and the other half of the lake area is in a middle nutrition state.Afterwards,the degree of eutrophication in Hongfeng Lake has been eased.Most of the lake areas are in the state of moderate nutrition,and only a small part is in the state of mild eutrophication.This shows that the water quality of Hongfeng Lake is gradually improving.On the other hand,There are also reasons why weather changes cause changes in nutrient concentrations in the lake area.