| Accurate and quantitative assessment of the impact of natural environmental changes and human activities on total suspended solids(TSS)concentration is one of the important contents of water environment protection and pollution prevention,and it can provide important support for water resources protection.As a typical trans-provincial basin(Guangdong-Guangxi),the Hedi Reservoir Basin in western Guangdong is an important source of drinking water.It undertakes the mission of supporting the sustainable development of the Leizhou peninsula.It is a veritable‘life water’and‘economic water’which has important strategic significance for the ecological protection of the basin.Due to the poor validation accuracy of the previous TSS models developed from coastal or inland waters,a new three-band remote sensing model of TSS was built by regression analysis for the inland reservoir of south China using the synchronous field spectral data,water quality samples(N=49)and remote sensing data(Landsat OLI and Sentinel-2,29 scenes)in the trans-provincial Hedi reservoir in Guangdong-Guangxi provinces of south China(Drinking water source of Guangdong Province).The temporal and spatial distribution of TSS for 2014-2018 was mapped to probe its spatial-temporal distribution characteristics in seasonal and interannual variations.Besides,we also attempted to analyze the driving factors and mechanisms that affect the changes in TSS.Combined with the Markov model for analyzing the trend of random events,the continuous change process of the TSS in the study area was quantitatively evaluated by calculating the progress degree of TSS.Also,this correlation between the progress degree of TSS and the level of TSS concentration was built,directly reflecting the variation of the TSS concentration and the water quality under China’s River and Lake Chief System.The main results and findings of this study are as follows:(1)The optical characteristics of the water body in the Hedi Reservoir are remarkable,and its spectral curve shows significant characteristics of the Four-Peak shape.According to the band sensitivity analysis,it is known that the three-band factor of remote sensing composed of 557nm,655nm and 697nm are significantly correlated with TSS concentration(r=0.87).Therefore,a three-band remote sensing estimation model for TSS based on Landsat OLI data was calibrated using the regression analysis(R~2=0.81,N=34,P value<0.001),and the three-band model based on the OLI sensor explained about 82%of the TSS concentration variation with an acceptable validation accuracy(RMSE=6.24mg/L,MRE=18.02%,N=15).(2)The results of TSS mapping indicate that in the upstream reservoir zone(Shijiao Town)of Hedi reservoir from 2014 to 2018 there is always the high-risk area of TSS that influences the drinking water security of the reservoir(up to 2.1 km south),and the TSS in the whole reservoir area has a spatial distribution characteristic of‘high upstream and low downstream’,where the average concentration of TSS at 31.54mg/L in the upstream(the reservoir area of Shijiao Town)is two-three times that of the downstream(12.55mg/L)of the reservoir.Besides,the reservoir area of Xin’an Town was an important risk area in 2017 with an average TSS of 31.84mg/L due to the cultivation of orchards.(3)Rainfall is an important driving factor affecting the short-term and seasonal dynamics of TSS concentration.From 2014 to 2018,the TSS concentration in upstream(Shijiao Town)was more affected by rainfall in the dry season(with an average TSS of 33.66 mg/L)than in the flood period(with an average of 22.45 mg/L).The main reason is that the reservoir area of Shijiao Town is located in the upstream runoff input area,where is the deposition area of TSS by the upstream Jiuzhou River flow and the surrounding load runoff.During the dry season,the water level is relatively low.In addition to the surface runoff caused by rainfall,it is also easy to resuspend the sediment at the bottom of the water,which eventually leads to an increase in TSS.In contrast,the water level is relatively high and the TSS concentration is relatively less disturbed by rainfall in the flood season.At this time,the TSS is lower and less affected.(4)The Markov dynamic process analysis indicates the water quality improvement of the upstream reservoir area under China’s River and Lake Chief System after 2016with an annual average TSS reduction of 3.76 mg/L,from 32.66mg/L in 2014 to13.84mg/L in 2018.Especially in 2018,the improvement of the TSS level reached 3levels which the TSS level transferⅠfromⅣ(the average TSS was 33.66 mg/L in2018),with the highest progress value(up to 10.23),followed by 2014(2.49),2017(2.04),2016(0.81)and 2015(0.19),which indicates the effectiveness of implementing the River Chief System of China science September of 2017. |
PDF Full Text Request