Analysis of Spatio-temporal Change Characteristics of Water Environment in Nansi Lake Based on Satellite Remote Sensing Monitoring

. Lakes play a crucial role in the preservation and distribution of surface water resources, making them integral components of environmental protection. This study utilizes remote sensing data obtained from Landsat-8 to investigate the temporal changes in water area, land use patterns, and distribution of cyanobacteria in Nansi Lake between 2015 and 2023. Furthermore, it conducts an extensive analysis of the environmental condition of Nansi Lake, considering the escalating environmental crises resulting from industrial and agricultural development in recent decades. By adopting a multidimensional approach, this article offers a comprehensive examination of the environmental situation in the Nansi Lake region. Notably, the findings of this study exhibit enhanced accuracy and practicality when compared to conventional environmental analysis methods. Consequently, this research contributes novel ideas and methodologies for the monitoring and safeguarding of the ecological environment of lakes through the application of remote sensing techniques.


Introduction
Aquatic ecosystems are an important foundation for the existence of life on Earth.In recent decades, human activities have had some impacts on freshwater lakes and their surrounding environment.Compared to traditional sampling analysis methods, conducting a series of environmental analysis on a certain research area based on satellite image data is a fast, effective, and costeffective method.
In the past decade or so, due to the rapid development of domestic industry and agriculture, the phenomenon of water pollution in China has become increasingly apparent.This article conducted a series of environmental analyses on Nansi Lake from 2015 to 2023 based on Landsat-8 data.This study conducted a two-year cycle to monitor changes in water area, analyse land use types, and extract and analyse cyanobacteria from the selected five time periods in the Nansi Lake area.
Compared to previous research on remote sensing of lakes, this article conducts a more objective and rigorous analysis of the environmental situation of Nansi Lake from multiple dimensions.It can accurately reflect the current environmental situation of Nansi Lake, providing new ideas and theoretical basis for using aerial remote sensing technology to monitor changes in the lake environment on a large scale.

Research area
Nansi Lake is the largest freshwater lake in northern China, located in Weishan County, Jining, Shandong Province.It plays a crucial role in the East Route of the South to North Water Diversion Project [1][2] .Since the beginning of the 21st century, due to the development of heavy industry along the lake coast and other reasons, the water pollution in this area has been severe.The study area of Nansi Lake is shown in the Figure 1.

Data and used methods
This article uses remote sensing images provided by Landsat-8 satellite as the data source and analyses the environmental situation of Nansi Lake through a combination of three methods [3][4][5][6].
This study utilized a New Normalized Difference Water Index(NNDWI=ρ(RED+NIR)/ρ(GREEN*C))) to extract the water body of Nansi Lake.Support vector machine classification [7][8] and cluster analysis were used to study the land use types in the Nansi Lake area.The Normalized Difference Vegetation Index (NDVI=ρ(NIR-RED)/ ρ (NIR+RED)), which eliminated outliers, and parameter assignment were used to extract and analyse cyanobacteria in the water body of Nansi Lake.And it is shown in Table 1.

Results & discussion
This article uses Landsat-8 image data from 2015 to 2023 as the data source to conduct three different studies on the Nansi Lake area.The current research on extracting water based on remote sensing technology is generally divided into two directions: enhancing water information and suppressing non-water information.This article selects the method of enhancing water information.Compared with other water indices, NNDWI has a lower probability of classification errors and higher accuracy in extracting results [9][10] .In order to accurately extract the boundary of the Nansi Lake water area, this paper further refined the above results using threshold method based on the NNDWI results.The threshold range is 0.0000001 to 0.8.The final water area changes in the Nansi Lake area from 2015 to 2023 are shown in Figure 2.
Due to the high classification accuracy of the support vector machine classification method compared to the minimum distance classification method and the maximum likelihood classification method in this research area.So, this article uses SVM to divide the Nansi Lake area into four categories: water area, arable land, residential land, and other land.The classification results of land use types in Nansi Lake based on SVM are shown in Table 2.
In order to further optimize the extraction results, this study excluded the interference of suspended impurities in water through threshold method based on NDVI, and combined with real research data in the Nansi Lake area, outliers in the results were removed through parameter assignment, which can more intuitively reflect the distribution of Cyanobacteria in the water.Figure 3-7 show the distribution of Cyanobacteria in Nansi Lake.

Conclusion
Analysing the changes in the water area of Nansi Lake from 2015 to 2023, the water area in this area first increased and then decreased.As of 2023, the total decreased area was 85.428 square kilometres.Through the study of land use types in the Nansi Lake region, this article found that during this period, the proportion of water area in the Nansi Lake region decreased from 49% in 2015 to 44% in 2023.With the decrease in the proportion of water bodies, the proportion of arable land in the region has increased, and the changes in the two are extremely similar.Through the extraction and analysis of cyanobacteria in this area, it can be concluded that the distribution of cyanobacteria in Nansi Lake is random, and the frequency of occurrence is periodic.The situation of cyanobacteria is most severe in the southeast region.
Through a comprehensive analysis of the Nansi Lake area, the following conclusions can be drawn: between 2015 and 2023, the water area of Nansi Lake had some fluctuations, but overall it was still within the selfregulation range of the lake; The proportion of arable land in the region has increased, while the proportion of water bodies has decreased, and there is a correlation between the two; The problem of cyanobacteria in the Nansi Lake area has been weakening over time, and the data from 2023 can reflect that the vast majority of water bodies are not covered by cyanobacteria.In summary, during this period, environmental protection in the Nansi Lake region achieved phased results, and environmental issues were greatly reduced.The local government has made effective efforts in protecting the environment.
In the field of environmental research in Nansi Lake, this article uses remote sensing technology to combine water area, land use status, and cyanobacteria analysis.By comparing and analysing multiple dimensions, the analysis of the regional environment is more intuitive, accurate, and efficient.This article uses NNDWI for the first time to extract the water bodies of Nansi Lake.Compared to other water body indices, this index has significant advantages in accuracy and reliability.In terms of cyanobacteria extraction and analysis, this article uses a combination of NDVI and assignment parameter method to accurately study the distribution of cyanobacteria in Nansi Lake.In the future, if the relationship between lake water quality parameters and cyanobacteria reflectance spectra can be identified, and combined with on-site sampling data, it will be more efficient to complete large-scale quantitative analysis of water pollution.

Table 1 .
Parameter table for assignment

Year Parameters corresponding to cumulative area ratio less than 5% Parameters corresponding to cumulative area ratio greater than 90%
Fig. 1.Research area of Nansi Lake.

Table 2 .
Land use classification based on SVM