Font Size: a A A

Analysis Of Spatial And Temporal Patterns Of Underwater Light Field And Its Driving Forces In Dongting Lake,China:An Optics And Remote Sensing Approach

Posted on:2019-05-29Degree:DoctorType:Dissertation
Country:ChinaCandidate:Z B ZhengFull Text:PDF
GTID:1480305489951459Subject:Geographical environment remote sensing
Abstract/Summary:PDF Full Text Request
As the second largest freshwater lake in China,Dongting Lake is the largest throughput lake over the Yangtze River and undertakes important storage and transportation functions in the Yangtze River Basin.In recent years,with the anthropogenic reclamation,the construction of water conservancy projects such as the Three Gorges Project and the gradual escalation of sand mining in the lake area,the decay of underwater light field and the increase of water turbidity due to the resuspension of sediment have been becoming the main problems towards the region.Therefore,improving the underwater light field conditions in the lake and increasing the transparency of the water so as to restore the lake aquatic plants are the significant procedure for the ongoing water environment management of Dongting Lake.Nonetheless,if this is a case that an urgent task should improve the conditions of underwater light field and enhance the transparency of water bodies,the research of spatial and temporal patterns of underwater light field and its driving forces over Dongting Lake becomes the crucial thing at present.Remote sensing technology can be used to quickly and accurately monitor the temporal and spatial pattern of underwater light field characterized by diffuse attenuation coefficient(Kd)and Secchi disk depth(ZSD)of water bodies in Dongting Lake.The theory and technology used here can be provided the fundamental of the remote sensing estimation of Kd and ZSD over the lake waters,expecially in the middle and lower reaches of the Yangtze River area.Taking Dongting Lake as the case,this reseach analyzes its underwater light field information,explores its variational rules,establishes the empirical model and semianalytical model to estimate the Kd and ZSD information,evaluates the accuracy and applicability of the estimation model,determines the best algorithms to estimate the spatial and temporal patterns of underwater light field and its variational rules over Dongting Lake,and then reveals the driving mechanism of spatial and temporal patterns of underwater light field in Dongting Lake,such as large-scale water conservancy projects,sand mining activities and climate change,realize the whole process remote sensing monitoring of the underwater phenomenon of Dongting Lake from the natural phenomenon-estimation model-spatio-temporal patterns-variational rule-driving factors,which is of the great significance for monitoring the underwater light field of Dongting Lake,and can provide reference for the water environment and water ecological restoration of the lakes in the middle and lower reaches of the Yangtze River in China.The main research contents and conclusions of this paper are as follows:(1)Study on the underwater light field and its influence factors of Dongting Lake based on in situ measurements data.In terms of water body composition,the concentration of total suspended matter(TSM)in Dongting Lake is higher overall,while the concentration of chlorophyll-a(Chla)and colored dissolved oraganic matter(CDOM)are lower.At the same time,the proportion of inorganic suspended matter(ISM)in TSM is more than 70%,which is typical of TSM dominated waters.The absorption contribution of each component does not show significant difference at different wavelengths.The water absorption of most samples is dominated by nonpigment particles throughout the visible range.The absorption contribution of CDOM is small,and is almost zero at 675 nm.The contribution of absorption comes from the spatial heterogeneity of the water optical properties of the sample due to the difference in water body components at each sample point.(2)Semi-analytical model construction of remote sensing estimation for underwater light field over Dongting Lake.A serious overestimation has been appeared when we estimated the Kd and ZSD using the QAA algorithm directly in Dongting Lake,with a MAPE of 341.7%and 253%respectively.The overestimation phenomenon is mainly caused by the tremendous error of bb(490)and a(490).On the contrary,using the 715nm as the reference wavelength for re-parameterization of QAA algorithm,the results of estimated Kd(490)and ZSD were several times lower than that of the original algorithm.The MAPE dropped to 70.8%and 64.5%respectively,and the error dropped by more than 80%,howeve,it's still larger than the criterion of IOCCG(<30%);the result shows that the original assumption of the QAA algorithm is invalid in turbid lakes of China,and the reference band must be re-adjusted and reparameterized..(3)The empirical model construction and applicability analysis of remote sensing estimation for underwater light field in Dongting Lake.Compared with the semianalytical model,the performance of the band-ratio empirical model of Kd(nearinfrared and green band)and ZSD(red and green band)were better.The coefficients of determination of Kd(490)and ZSD model(R2)were 0.808 and 0.814 respectively.The MAPE and RMSE were 20.85%and 0.515 m-1,21.68%and 0.076 m,respectively.The accuracy of the proposed method was fitting well with the criterion of water color remote sensing.The Kd(490)band-ratio algorithms can be well applied to the estimation of the diffuse attenuation coefficient of the combination of Landsat ETM+/TM/MSS,which provides a theoretical basis for the future retrieval and dynamic monitoring of the underwater light field of Dongting Lake based on long-time series of remote sensing images;The results of the noise equivalent ZSD(NEZSD)method show that the proposed algorithm has the best applicability to OLI images compared with the commonly used color remote sensing images(ETM+/OLI/MSI/OLCI/MODIS/HJ-1/GF-1),NEZSD is 0.072 m.(4)Study on temporal and spatial patterns of underwater light field and its variational rules over Dongting Lake.In the view of patterns of Kd(490)and ZSD,it has significant spatio-temporal heterogeneity.The water bodies between Dongting Lake and the Yangtze River around Chenglingji site have the opposite time-period regularity,and their monthly variations can be parameterized.It is then found that the Kd(490)and ZSD values of the water body near the Chenglingji Hydrologic Station had a significant clear-turbid inversions during the year.The relationship between Kd(490)and ZSD in Dongting Lake and the Three Gorges reservoir is similar with water body composition and optical parameters.The fluctuations of TSM dominated by ISM is the main factor affecting the variations of underwater light field in Dongting Lake.The Three Gorges reservoir area has the same covariation features and foundation with the underwater light field of the main channel of the Yangtze River in the middle and lower reaches of the Yangtze River,providing an important basis for revealing the driving forces of the temporal and spatial distribution of the underwater light field in Dongting Lake.(5)The main driving forces of temporal and spatial variations of underwater light field in Dongting Lake are revealed:the clear-turbid inversions between Dongting Lake and the main channel of the Yangtze River around Chenglingji site is due to the different driving forces behind themselves.The fluctuations of the underwater optical field of the main channel of the Changling River around Chenglingji is mainly due to the changes of surface runoff and sediment discharges of the Yangtze River's middle and lower reaches driven by the outbound runoff of Three Gorges Dam and Gezhouba Reservoir,whereas the main driving force of the temporal and spatial heterogeneity of the underwater light field in Dongting Lake can be explained by the variations of surface runoff driven by local precipitations,and the interference caused by sand mining activity as well.
Keywords/Search Tags:Dongting Lake, underwater light field, Landsat, estimation model, spatio-temporal patterns, driving forces
PDF Full Text Request
Related items