Wetland Species Mapping Using Advanced Technological Measurement DOI
Smrutisikha Mohanty, Prashant K. Srivastava, Prem Chandra Pandey

et al.

Aquatic Conservation Marine and Freshwater Ecosystems, Journal Year: 2024, Volume and Issue: 34(12)

Published: Dec. 1, 2024

ABSTRACT Wetlands are pivotal in supporting the natural ecosystem and maintaining biodiversity while being susceptible to anthropogenic activities climate change. However, monitoring wetlands over a large geographical temporal extent is challenging. Vegetation health can be considered good indicator of wetland conditions, measuring chlorophyll content will provide insight into vegetation health. Linking species mapping from spectral indices local regional conservation strategies could improve conservation. Here, we apply this Keetham Lake, India, using machine learning methods (relevance vector model) hyperspectral measurements. From 10 chlorophyll‐sensitive indices, identified four as best performing, particularly for: TVI + CCCI NDRE for calibration validation data. The least performing combinations were MCARI validation. Overall, that was best‐performing pair assessment implementation species. This approach allows precise species, providing data on their area they cover. By creating digital database, method enables long‐term changes species' numbers distribution, helping assess trends increase or decline freshwater ecosystems. Such vital both global efforts, offering insights forward‐looking, data‐driven preservation initiatives.

Language: Английский

Dynamics of the optical water quality parameters in the Lake Nokoué and Cotonou Channel complex (Benin) DOI Creative Commons

Romaric C.M. Hekpazo,

Metogbe Belfrid Djihouessi,

Béatrix Amen Tigo

et al.

Environmental Challenges, Journal Year: 2025, Volume and Issue: unknown, P. 101126 - 101126

Published: March 1, 2025

Language: Английский

Citations

0
Lake Surface Temperature Predictions under Different Climate Scenarios with Machine Learning Methods: A Case Study of Qinghai Lake and Hulun Lake, China DOI Creative Commons
Zhenghao Li, Zhijie Zhang, Shengqing Xiong

et al.

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(17), P. 3220 - 3220

Published: Aug. 30, 2024

Accurate prediction of lake surface water temperature (LSWT) is essential for understanding the impacts climate change on aquatic ecosystems and guiding environmental management strategies. Predictions LSWT two prominent lakes in northern China, Qinghai Lake Hulun Lake, under various future scenarios, were conducted present study. Utilizing historical hydrometeorological data MODIS satellite observations (MOD11A2), we employed three advanced machine learning models—Random Forest (RF), XGBoost, Multilayer Perceptron Neural Network (MLPNN)—to predict monthly average across scenarios (ssp119, ssp245, ssp585) from CMIP6 projections. Through comparison training validation results models both regions, RF model demonstrated highest accuracy, with a mean MAE 0.348 °C an RMSE 0.611 °C, making it most optimal suitable this purpose. With model, predicted reveals significant warming trend future, particularly high-emission scenario (ssp585). The rate increase pronounced ssp585, showing rise 0.55 per decade (R2 = 0.72) 0.32 0.85), surpassing trends observed ssp119 ssp245. These underscore vulnerability to provide insights proactive adaptation management.

Language: Английский

Citations

0
Uncertainty assessment of optically active and inactive water quality parameters predictions using satellite data, deep and ensemble learnings DOI
Bahareh Raheli,

Nasser Talabbeydokhti,

Vahid Nourani

et al.

Journal of Hydrology, Journal Year: 2024, Volume and Issue: unknown, P. 132091 - 132091

Published: Oct. 1, 2024

Language: Английский

Citations

0
Wetland Species Mapping Using Advanced Technological Measurement DOI
Smrutisikha Mohanty, Prashant K. Srivastava, Prem Chandra Pandey

et al.

Aquatic Conservation Marine and Freshwater Ecosystems, Journal Year: 2024, Volume and Issue: 34(12)

Published: Dec. 1, 2024

ABSTRACT Wetlands are pivotal in supporting the natural ecosystem and maintaining biodiversity while being susceptible to anthropogenic activities climate change. However, monitoring wetlands over a large geographical temporal extent is challenging. Vegetation health can be considered good indicator of wetland conditions, measuring chlorophyll content will provide insight into vegetation health. Linking species mapping from spectral indices local regional conservation strategies could improve conservation. Here, we apply this Keetham Lake, India, using machine learning methods (relevance vector model) hyperspectral measurements. From 10 chlorophyll‐sensitive indices, identified four as best performing, particularly for: TVI + CCCI NDRE for calibration validation data. The least performing combinations were MCARI validation. Overall, that was best‐performing pair assessment implementation species. This approach allows precise species, providing data on their area they cover. By creating digital database, method enables long‐term changes species' numbers distribution, helping assess trends increase or decline freshwater ecosystems. Such vital both global efforts, offering insights forward‐looking, data‐driven preservation initiatives.

Language: Английский

Citations

0