Monitoring Of Chlorophyll-a Concentration And Evaluation Of Eutrophication Based On Remote Sensing Technology

Water is the cornerstone for the survival and reproduction of all living things in the biosphere,and freshwater is also the most important material for the steady development of human society.In recent years,The inland water is polluted extremely seriously in China,resulting in increasingly sharp contradictions between supply and demand.Eutrophication is the main manifestation of inland water pollution in China.Its evaluation results can not only directly reflect the quality and pollution status of water environment,but also provide theoretical basis and technical guidance for the prevention,management and decision-making of water pollution,Chlorophyll A is a basic component of algae,and its content can be used to characterize the biomass and productivity of phytoplankton in water.At the same time,it is also a basic index to measure the eutrophication of water body.The traditional monitoring method is to carry out laboratory analysis of water samples at home and abroad.Although it can accurately analyze and evaluate more water quality indicators,it is time-consuming and laborious,and the monitoring stations that can be collected are very limited.It can not reveal the distribution of water parameters in a wide range to meet the needs of real-time monitoring.Remote sensing technology can intuitively show the spatial and temporal distribution and change of water quality in a large area of water,and has the advantages of large amount of information,less geographical constraints and long-term monitoring.So far,it has been widely used in a large number of studies.This paper takes Baihua Lake in Guiyang City as the research area,combines the measured hyperspectral and multi-spectral remote sensing data,and based on the research foundation of remote sensing monitoring methods for water quality in other areas,explores remote sensing monitoring methods suitable for chlorophyll a in Baihua Lake,and establishes remote sensing evaluation model for eutrophication status of Baihua Lake.The main research contents and conclusions are as follows:(1)According to the spectral reflectance characteristics of Baihua Lake,seven models of band ratio,fluorescence peak location,first-order differential,reflection peak area,peak-valley distance,baseline fluorescence peak height and three-band were established respectively.The correlation analysis between the models and measured chlorophyll-a concentration showed that there was a good linear or non-linear correlation between the models and chlorophyll-a concentration.They were 0.79,0.92,0.81,0.86,0.75,0.94 and 0.88,respectively.These models can reflect the chlorophyll a concentration in Baihua Lake to a certain extent.(2)The regression analysis results of several models and chlorophyll-a concentration showed that the baseline fluorescence peak height model and the fluorescence peak position model had better estimation effect,and their determinant coefficients were 0.92 and 0.87,respectively.In addition,the test results based on the measured data show that the errors in estimating chlorophyll a concentration are small in Baihua Lake,and the baseline fluorescence peak height model is more accurate.Therefore,the baseline fluorescence peak height model is the best estimation model of chlorophyll a concentration in Baihua Lake.(3)Based on the chlorophyll-a concentration and measured spectral data of several other lakes(reservoirs)in different trophic states during this period,the chlorophyll-a estimation models suitable for different trophic states were established according to the above analysis methods.The effects of chlorophyll-a concentration estimation models in different trophic states were different.Medium-trophic and mild eutrophic water bodies,band ratio method and three methods were used to estimate chlorophyll-a concentration in different trophic states.The band model is superior to other algorithms,and the first-order differential model and baseline fluorescence peak height model are ideal for moderately and severely eutrophic water.(4)Combining HJ-1A CCD1 image data and measured spectral data,the relationship between chlorophyll a concentration and each band was analyzed.It was found that the sensitive bands of chlorophyll a concentration were blue and green bands.Among them,the blue and green band ratio(B2/B1)has a higher correlation with chlorophyll a concentration,and the correlation coefficient R2 is 0.84.Taking the ratio(B2/B1)as an independent variable and the measured chlorophyll a concentration as a dependent variable,a series of fitting models are established,and the accuracy of the corresponding models is analyzed and the sample area is tested.The inversion model with high precision is y=3.2425x10.269,and the model established by April data of the lake is applied to the actual data in December to verify,which also achieves better results,and further proves that the model has certain applicability.Within a certain period of time,it can be unrestricted.(5)According to the inversion model,the chlorophyll-a concentration of Baihua Lake showed a general trend of "high lakeside and low Lake center",and gradually increased from north to south.The spatial distribution of nutrient status index existed the phenomenon of "high inflow and low outflow" which was mainly affected by nutrient input.Among them,the eutrophication region accounts for 95% of the total area of the lake,and the middle trophic region accounts for only 5% of the total area of the lake.The water body is mainly eutrophic.