Climate-driven variations in suspended particulate matter dominate water clarity in shallow lakes

Monitoring phycocyanin in global inland waters by remote sensing: Progress and future developments

Cyanobacterial blooms are increasingly becoming major threats to global inland aquatic ecosystems. Phycocyanin (PC), a pigment unique to cyanobacteria, can provide important reference for the study of cyanobacterial blooms warning. New satellite technology and cloud computing platforms have greatly improved research on PC, with the average number of studies examining it having increased from 5 per year before 2018 to 17 per year thereafter. Many empirical, semi-empirical, semi-analytical, quasi-analytical algorithm (QAA) and machine learning (ML) algorithms have been developed based on unique absorption characteristics of PC at approximately 620 nm. However, most models have been developed for individual lakes or clusters of them in specific regions, and their applicability at greater spatial scales requires evaluation. A review of optical mechanisms, principles and advantages and disadvantages of different model types, performance advantages and disadvantages of mainstream sensors in PC remote sensing inversion, and an evaluation of global lacustrine PC datasets is needed. We examine 230 articles from the Web of Science citation database between 1900 and 2024, summarize 57 of them that deal with construction of PC inversion models, and compile a list of 6526 PC sampling sites worldwide. This review proposed the key to achieving global lacustrine PC remote sensing inversion and spatiotemporal evolution analysis is to fully use existing multi-source remote sensing big data platforms, and a deep combination of ML and optical mechanisms, to classify the object lakes in advance based on lake optical characteristics, eutrophication level, water depth, climate type, altitude, population density within the watershed. Additionally, integrating data from multi-source satellite sensors, ground-based observations, and unmanned aerial vehicles, will enable future development of global lacustrine PC remote estimation, and contribute to achieving United Nations Sustainable Development Goals inland water goals.

Satellite remote sensing of turbidity in Lake Xingkai using eight years of OLCI observations

In the context of global climate change and land use change, both of which significantly affect lake ecosystems, rainfall and wind conditions play a crucial role in lake mixing processes. Furthermore, land use changes impact water quality through modifications in runoff and sediment inputs. These factors exert a profound influence on lake ecosystems, thereby necessitating further investigation into their extent and ultimate consequences. In this study, we developed a band ratio model for estimating the turbidity concentration of Lake Xingkai (XKH), a border lake between China and Russia, using data from the Sentinel-3 Ocean and Land Color Instrument (OLCI). The model demonstrated a high degree of accuracy, with an R2 of 0.84, a root mean square error (RMSE) of 27.08 NTU, a mean absolute error (MAE) of 16.58 NTU, and a mean absolute percentage error (MAPE) of 21.44% within the turbidity range of 20-400 NTU. The model was applied to 1240 cloud-cleared OLCI images from 2016 to 2023. The following findings were identified: (1) The optimal band for turbidity estimation was identified, and a robust model was developed based on the spectral response of turbidity in XKH; (2) monthly and annual analyses revealed a distinct upward trend in turbidity from July to October in the sub-region of XKH influenced by the Muling River tributary, differing from other areas of the lake. (3) By integrating meteorological and land use data, we investigated the influence of land use change on turbidity, uncovering the formation of persistent, distinctive river plume during periods of minimal climate impact. (4) Subsequent analysis revealed a correlation between turbidity and the occurrence of algal blooms. Therefore, monitoring turbidity changes can serve as an early warning for algal bloom events, offering valuable insights into the combined effects of climate and environmental changes on lake ecosystems.

A novel chlorophyll-a retrieval model based on suspended particulate matter classification and different machine learning

Chlorophyll-a (Chla) in inland waters is one of the most significant optical parameters of aquatic ecosystem assessment, and long-term and daily Chla concentration monitoring has the potential to facilitate in early warning of algal blooms. MOD09 products have multiple observation advantages (higher temporal, spatial resolution and signal-to-noise ratio), and play an extremely important role in the remote sensing of water color. For developing a high accuracy machine learning model of remotely estimating Chla concentration in inland waters based on MOD09 products, this study proposed an assumption that the accuracy of Chla concentration retrieval will be improved after classifying water bodies into three groups by suspended particulate matter (SPM) concentration. A total of 10 commonly used machine learning models were compared and evaluated in this study, including random forest regressor (RFR), deep neural networks (DNN), extreme gradient boosting (XGBoost), and convolutional neural network (CNN). Altogether, 41 basic bands and 820 band ratios between the 41 bands were filtered by measuring their correlation with Ln(Chla) and several bands brought into different machine learning models. Results demonstrated that the construction of Chla concentration remote estimation model based on SPM classification could significantly improve the correlation between Ln(Chla) and 41 basic spectral band combinations, the correlation between Ln(Chla) and 820 band ratios, and the model verification R2 from 0.41 to 0.83. Furthermore, B3, B20, and B32 were finally selected based on correlation with SPM to classify SPM and the classification accuracy could reach 0.9. Finally, we concluded that RFR model performed best via comparing the R2, RMSE, and MAPE. By comparing the relative contribution of input bands in different groups, B3 contributed most to three groups. The model constructed in this study has promising prospects for promotion and application in other inland waters, and could provide systematic research reference for subsequent research.

A novel total phosphorus concentration retrieval method based on two-line classification in lakes and reservoirs across China

Phosphorus is widely recognized as a nutrient that restricts growth and is the primary contributor to eutrophication in 80 % of water bodies. Consequently, the Chinese government has consistently prioritized monitoring and controlling total phosphorus (TP) levels. The remote estimation of TP in lakes and reservoirs at a national scale is a challenging task due to TP being a non-optically active parameter. Currently, there is a lack of developed TP inversion models specifically designed for lakes and reservoirs in China. For solving this problem, a novel two-line classification method drawn on scatter plots based on the natural logarithm of TP (Ln(TP)) and B33/B9 was proposed and used to classify 1211 measured samples obtained from field cruises in 105 lakes and reservoirs across China from 2012 to 2022 into three categories, Class 1, Class 2, and Class 3. Results demonstrate that the proposed classification method has the ability to enhance the correlation between Ln(TP) and 43 basic potential single band and band combinations. Specifically, the correlation range improved from (-0.31,0.15) to (-0.77,0.24) in Class 1, (-0.81, 0.36) in Class 2, and (-0.74, 0.52) in Class 3. Additionally, the classification method also improved the correlation range between Ln(TP) and 820 band ratios, from (-0.32, 0.32) to (-0.83, 0.82) in Class 1, (-0.86, 0.86) in Class 2, and (-0.86, 0.86) in Class 3. These datasets were subsequently utilized as input for eXtreme Gradient Boosting (XGBoost) models. Finally, well performing XGBoost models in Class 1 (R2 = 0.76, RMSE = 0.3, MAPE = 12 %), Class 2 (R2 = 0.84, RMSE = 0.49, MAPE = 38 %), and Class 3 (R2 = 0.74, RMSE = 0.46, MAPE = 14 %) were used to map TP of 563 large lakes and reservoirs (≥20 km2) across China using MODIS images from 2005, 2010, 2015, and 2020. This study presents a novel approach for estimating non-optically active parameters through remote sensing on a national scale.