Remote-sensing retrieval of total phosphorus in Tiande Lake by optimizing an XGBoost machine learning model DOI

Aimin LI,

Xuan Kang,

Xiangyu Yan

et al.

Published: Nov. 15, 2023

Traditional models for total phosphorus (TP) retrieval from remote-sensing images generally show low accuracy. Additionally, parameter adjustment and selection of combinations machine learning (ML) exert significant influences on the regressive prediction effect model performance. To solve these problems, this research proposed an extreme gradient boosting (XGBoost) optimized by Bayesian optimization (BO), that is, BO-XGB. The optimal parameters are sought automatically a small sample size through BO, which shortens training time. Taking Tiande Lake in Zhengzhou City (Henan Province, China) as region interest, BO-XGB TP is established based GF1-WFV satellite data water-quality data. Moreover, accuracy compared with those other four ML methods, namely, XGBoost model, k-nearest neighbors (KNN), multilayer perceptron (MLP) random forest (RF). Compared models, demonstrates highest accuracy, coefficients determination R2 , root mean square error (RMSE), relative (MRE) separately 0.923, 2.15 × 10-3 mg/L, 1.81%. Finally, adopted to retrieve spatial distribution concentration Lake. results optimizing using BO can significantly improve algorithm more suitable retrieving mass findings have implications inverse modelling non-optical water quality such nitrogen(TN).

Language: Английский

Multi-sensor and multi-platform retrieval of water chlorophyll a concentration in karst wetlands using transfer learning frameworks with ASD, UAV, and Planet CubeSate reflectance data DOI
Bolin Fu,

Sunzhe Li,

Zhinan Lao

et al.

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 901, P. 165963 - 165963

Published: Aug. 4, 2023

Language: Английский

Citations

33
Meta-Analysis of Satellite Observations for United Nations Sustainable Development Goals: Exploring the Potential of Machine Learning for Water Quality Monitoring DOI Open Access
Sabastian Simbarashe Mukonza, Jie‐Lun Chiang

Environments, Journal Year: 2023, Volume and Issue: 10(10), P. 170 - 170

Published: Oct. 2, 2023

This review paper adopts bibliometric and meta-analysis approaches to explore the application of supervised machine learning regression models in satellite-based water quality monitoring. The consistent pattern observed across peer-reviewed research papers shows an increasing interest use satellites as innovative approach for monitoring quality, a critical step towards addressing challenges posed by rising anthropogenic pollution. Traditional methods have limitations, but satellite sensors provide potential solution that lowering costs expanding temporal spatial coverage. However, conventional statistical are limited when faced with formidable challenge conducting recognition analysis geospatial big data because they characterized high volume complexity. As compelling alternative, deep techniques has emerged indispensable tool, remarkable capability discern intricate patterns might otherwise remain elusive traditional statistics. study employed targeted search strategy, utilizing specific criteria titles 332 journal articles indexed Scopus, resulting inclusion 165 meta-analysis. Our comprehensive provides insights into trends, productivity, impact It highlights key journals publishers this domain while examining relationship between first author’s presentation, publication year, citation count, factor. major findings highlight widespread including MultiSpectral Instrument (MSI), Ocean Land Color (OLCI), Operational Imager (OLI), Moderate Resolution Imaging Spectroradiometer (MODIS), Thematic Mapper (TM), Enhanced Plus (ETM+), practice multi-sensor fusion. Deep neural networks identified popular high-performing algorithms, significant competition from extreme gradient boosting (XGBoost), even though XGBoost is relatively newer field learning. Chlorophyll-a clarity indicators receive special attention, geo-location had optical classes. contributes significantly providing extensive examples in-depth discussions code, well highlighting cyber infrastructure used research. Advances high-performance computing, large-scale processing capabilities, availability open-source software facilitating growing prominence applications artificial intelligence monitoring, positively contributing

Language: Английский

Citations

19
The normalized difference yellow vegetation index (NDYVI): A new index for crop identification by using GaoFen-6 WFV data DOI Creative Commons
Yanbing Wei,

Miao Lu,

Qiangyi Yu

et al.

Computers and Electronics in Agriculture, Journal Year: 2024, Volume and Issue: 226, P. 109417 - 109417

Published: Sept. 7, 2024

Language: Английский

Citations

7
Long-term monitoring chlorophyll-a concentration using HJ-1 A/B imagery and machine learning algorithms in typical lakes, a cold semi-arid region DOI Creative Commons

Jianhua Ren,

Haoyun Zhou,

Zui Tao

et al.

Optics Express, Journal Year: 2024, Volume and Issue: 32(9), P. 16371 - 16371

Published: April 9, 2024

Chlorophyll a (Chl-a) in lakes serves as an effective marker for assessing algal biomass and the nutritional level of lakes, its observation is feasible through remote sensing methods. HJ-1 (Huanjing-1) satellite, deployed 2008, incorporates CCD capable 30 m resolution has revisit interval 2 days, rendering it superb choice or supplemental sensor monitoring trophic state lakes. For long-term regional-scale mapping, both imagery evaluation machine learning algorithms are essential. The several typical algorithms, i.e., Support Vector Regression (SVR), Gradient Boosting Decision Trees (GBDT), XGBoost (XGB), Random Forest (RF), K-Nearest Neighbor (KNN), Kernel Ridge (KRR), Multi-Layer Perception Network (MLP), were developed using our in-situ measured Chl-a. A cross-validation grid to identify most hyperparameter combinations each algorithm was used, well selected optimal superparameter combinations. In Chl-a mapping three R2 GBDT, XGB, RF, KRR all reached 0.90, while XGB also exhibited stable performance with smallest error (RMSE = 3.11 μg/L). Adjustments made align spatial-temporal patterns past data, utilizing HJ1-A/B images algorithm, which demonstrates stability. Our results highlight considerable effectiveness utility A/B cold arid region, providing application cases contribute ongoing efforts monitor water qualities.

Language: Английский

Citations

6
Retrieval performance of mangrove tree heights using multiple machine learning regression models and UAV-LiDAR point clouds DOI Creative Commons
Bolin Fu,

Linhang Jiang,

Hang Yao

et al.

International Journal of Digital Earth, Journal Year: 2024, Volume and Issue: 17(1)

Published: Aug. 19, 2024

Mangroves are vital coastal ecosystems that provide crucial links between land and sea. Tree height is a key indicator for assessing mangroves' health status. Currently, there still numerous challenges in estimating mangrove tree height. In this study, multiple deep learning shallow machine regression models were developed to accurately estimate using multi-dimensional Light Detection Ranging (LiDAR) point clouds their derivatives. We constructed novel CNN_RepMLP model mapping. also further verified the applicability of different types heights, explored influence LiDAR-derived features on inversion accuracy heights. The results indicated following: (1) displayed satisfactory performance exhibited better robustness generalization ability than convolutional neural network (CNN) model. (2) Among feature combinations, combining variables with intensity can not only mitigate negative impact models, but enhance accuracy. (3) ensemble framework ExtraTrees as meta-model make use differences complementarities single base trees compared other models. (4) Multiple based UAV-LiDAR point-cloud-derived suitable outperformed CNN stacking had more detailed differentiation terms Its prediction realistically reflect spatial characteristics

Language: Английский

Citations

5
Water quality parameters retrieval and nutrient status evaluation based on machine learning methods and Sentinel- 2 imagery: a case study of the Hongjiannao Lake DOI
Ying Liu, Zhixiong Wang, Hui Yue

et al.

Environmental Monitoring and Assessment, Journal Year: 2025, Volume and Issue: 197(5)

Published: April 15, 2025

Language: Английский

Citations

0
Temperature Is a Key Factor Affecting Total Phosphorus and Total Nitrogen Concentrations in Northeastern Lakes Based on Sentinel-2 Images and Machine Learning Methods DOI Creative Commons

Hua Feng Qin,

Chong Fang, Ge Liu

et al.

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(2), P. 267 - 267

Published: Jan. 13, 2025

Nitrogen and phosphorus are limiting nutrients in freshwater ecosystems, the remote estimation of total (TP) nitrogen (TN) eutrophic waters is great significance. This study utilized machine learning algorithms based on Sentinel-2 satellite imagery for TP TN concentrations Lake Xingkai, Chagan Songhua. Results indicate that random forest (RF) XGBoost regression perform better. The performance GBDT algorithm was slightly lower than RF algorithms, BP had overfitting, SVR poor fitting performance. showed concentration inversion model highest accuracy (R2 = 0.98, RMSE 0.09, MAPE 19.74%). Extreme Gradient Boosting (XGB) also performed well, though less accurately 0.97, 0.14, 20.67%). For concentration, XGB model’s 0.82, 0.08, 24.89%) comparable to 0.07, 29.55%). applied all cloud-free images these typical lakes northeastern China during non-glacial period from 2017 2023, generating spatiotemporal distribution maps concentrations. Between Songhua increasing, decreasing, initially decreasing then increasing patterns, respectively. A positive correlation between temperature observed, as higher temperatures enhance biological activity. In contrast, a negative found with promote phytoplankton growth reproduction. not only offers new method monitoring eutrophication but provides valuable support sustainable water resource management ecological protection goals.

Language: Английский

Citations

0
PGIS as a Tool for Reservoir Health Assessment: Community Insights Validated by Laboratory Analysis and Remote Sensing DOI

Sami Ur Rahman,

Yingxin Chen, Yuping Su

et al.

Ecohydrology, Journal Year: 2025, Volume and Issue: 18(2)

Published: March 1, 2025

ABSTRACT This study employed an innovative participatory geographic information system (PGIS) approach to evaluate the health of reservoirs and their socioecological importance communities within Shanzai sub‐catchment. The participation rate was 100% in all five communities, with 53% participants were women. Statiscial analysis shows that algal bloom negatively correlate less fish productivity positively unhealthy reservoir indicators. In contrast, clean water correlates healthy indicators, while blooms consistently show negative correlations indicators reservoir. These findings current health, 41% recognizing as healthy, 23% 36% responded moderate. Laboratory identified 30 phytoplankton genera, Cyanophyta dominant group. Highest density observed May, followed by June April, providing crucial insights into seasonal dynamics ecosystems. Sentinel‐2 imagery further highlighted fluctuations, extent particularly increased during May 2023, supporting quality measurements validating algae a community‐identified indicator. underlines value accuracy community‐driven environmental monitoring. alignment mapping, laboratory analyses, remote sensing demonstrates efficiency PGIS managing freshwater resources. By fostering knowledge exchange, this promotes sustainable resource monitoring conservation. represent significant contribution advancement underscore prioritizing initiatives research.

Language: Английский

Citations

0
Prediction Dynamics in Cotton Aphid Using Unmanned Aerial Vehicle Multispectral Images and Vegetation Indices DOI Creative Commons
Jiang Ping-an, Xuelin Zhou, Tonglai Liu

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 5908 - 5918

Published: Jan. 1, 2023

Cotton harvest can be increased by having real-time information on the state of cotton aphid populations. However, traditional monitoring relies ground sample methods supported models such as linear regression, resulting in low forecast accuracy. Therefore, this paper purposes to enhance precision remote sensing prediction model investigating construction approach. We explored effectiveness XGBoost algorithm combined with GWO and SVR method for relying vegetation indices derived from UAV multispectral photography. Originally, 12 related aphids were calculated reflectance. Additionally, optimal index combination pest was determined utilizing analysis correction two-way ANOVA, algorithm. Furthermore, a prevalence constructed via methodology associated catalog combination, optimized using Compared seven algorithms, experimental results demonstrate that MSE MAE XGBoost-GWO-SVR are reduced 90.20% 70.36% (SVR), 90.14% 70.26% (XGBoost-SVR), 7.47% 0.14% (XGBoost-GA-SVR), 5.80% 0.11% (XGBoost-PSO-SVR), 12.06% 58.95% (LR), 84.77% 89.22% (BPNN), whereas $R^{2}$ is 22.5% (SVR XGBoost-SVR), 0.3% 12.51% (BPNN). The XGBoost-SVR GWO, PSO, GA not significantly different. Among these models, obtained highest 0.980 lowest 2.838.

Language: Английский

Citations

9
Retrieval of Water Quality Parameters in Dianshan Lake Based on Sentinel-2 MSI Imagery and Machine Learning: Algorithm Evaluation and Spatiotemporal Change Research DOI Creative Commons
Lei Dong,

Cailan Gong,

Hongyan Huai

et al.

Remote Sensing, Journal Year: 2023, Volume and Issue: 15(20), P. 5001 - 5001

Published: Oct. 18, 2023

According to current research, machine learning algorithms have been proven be effective in detecting both optical and non-optical parameters of water quality. The use satellite remote sensing is a valuable method for monitoring long-term changes the quality lake water. In this study, Sentinel-2 MSI images situ data from Dianshan Lake area 2017 2023 were used. Four methods tested, optimal detection models determined each parameter. It was ultimately that these could applied analyze spatiotemporal variations distribution patterns Lake. Based on research findings, integrated algorithms, especially CatBoost, achieved good results retrieval all parameters. Spatiotemporal analysis reveals overall uneven, with significant spatial variations. Permanganate index (CODMn), Total Nitrogen (TN), Phosphorus (TP) show relatively small interannual differences, generally exhibiting decreasing trend concentrations. contrast, chlorophyll-a (Chl-a), dissolved oxygen (DO), Secchi Disk Depth (SDD) exhibit inter-year differences. Chl-a reached its peak 2020, followed by decrease, while DO SDD showed opposite trend. Further indicated significantly influenced climatic factors human activities such as agricultural expansion. Overall, there has an improvement study demonstrates feasibility accurately even without measured spectral data, using reflectance data. presented paper can provide new insights into resource management

Language: Английский

Citations

9