Construction Of Remote Sensing Retrieval Model Of Chlorophyll-a Based On Landsat8 Image And Its Application In Wuhan East Lake

In recent years,with the development of economy,China's water pollution problem is more and more serious,especially inland lakes eutrophication problem is particularly prominent.Chlorophyll is the most important pigment that affects photosynthesis and is an important indicator of water eutrophication assessment.With the development of remote sensing technology,satellite remote sensing has the advantages of low monitoring cost,wide range,fast cycle update and good dynamic continuity.It plays an increasingly important role in water environment monitoring.The inversion of chlorophyll concentration in inland water bodies with turbid eutrophication and regional differences has been a difficult problem in water color remote sensing inversion.How to quantitatively simulate a large amount of remote sensing information and the relationship between water quality concentrations with distinct geographical characteristics is the key to a breakthrough.Although there are many modeling inversion models that can be applied to the inversion of chlorophyll a concentration in inland lakes with the study of domestic and foreign scholars,these models may not apply because of the large regional differences in landscapes and the serious human impact In Wuhan East Lake in the city of inland lakes,therefore,need to build for the Wuhan East Lake chlorophyll a monitoring inversion model.Based on the study of the spectral characteristics of Wuhan East Lake water body,the chlorophyll a concentration data and the multi-spectral Landsat8 remote sensing images were used to analyze the spectral characteristics of the East Lake in Wuhan.Based on the study of the East Lake area in Wuhan,the chlorophyll was used as the research object.,Three-band empirical method and BP neural network technique were used to construct the inversion model of chlorophyll a concentration with the regional characteristics of Wuhan East Lake.The research work done in this paper is as follows:(1)Based on the inversion demand of chlorophyll a quantitative remote sensing,a series of pretreatment of Landsat8 remote sensing image was carried out,and the preparation of accurate extraction of remote sensing image information was completed.(2)By analyzing the correlation between the combination of each band and the chlorophyll a concentration of Landsat8 image,380 groups with strong sensitivity were selected as variables.SPSS software was used to determine the relationship between these variables and chlorophyll a concentration Quantitative relationship,the establishment of the R~2 is 0.739,RMSE is1.721,the average error of 0.1208 band combination model.(3)By analyzing the spectral characteristics of Wuhan East Lake water body,this paper chooses?1 as near infrared band,?2 and?3 are red band and mid-infrared band respectively,and the structure of R~2is 0.712,RMSE is 3.106,the average error is 0.1993 Band model.(4)In order to further improve the accuracy of chlorophyll a in Wuhan East Lake,this paper compares the training results of different implied nodes by using R and MSE as indicators.A BP neural network model of 6-8-1 structure,According to Landsat8 image of the chlorophyll is more sensitive to the 2-7 band reflectance as the input value,chlorophyll a concentration measured value as the output value,according to the inversion results calculated that the model R~2 is 0.942,RMSE 1.47.(5)By comparing the R~2 and RMSE of the three models,The BP neural network is suitable for finding the water quality monitoring inversion model suitable for Wuhan East Lake area.Reliable basis and method for the use of remote sensing technology to quickly and large-scale,accurate monitoring of the East Lake waters of Wuhan,the concentration of chlorophyll a provides a scientific technical support.Based on the characteristics of the measured data,this paper uses the 2015 data to construct the model and uses the three model inversions to obtain the chlorophyll a concentration profile of 12 East Lake in Wuhan from March to August,Spatial Distribution Characteristics and Time Variation of Chlorophyll-a Concentration in Wuhan East Lake Area.The results show that the concentration of chlorophyll a in the lake area is not uniform,and the concentration difference between each sub-lake is large,and it has obvious spatial distribution characteristics and general time variation.The space is roughly the fruit lake,the concentration of chlorophyll in the lake near the residential area is relatively large,and the concentration of chlorophyll in the lake area is small,the concentration of lake shore is larger and the concentration of lake center is smaller.The change of chlorophyll a concentration in Wuhan East Lake was significant,but the change showed a wide range of continuous characteristics.The basic trend of concentration is summer>autumn>winter>spring?...