Validity evaluation of a machine-learning model for chlorophyll a retrieval using Sentinel-2 from inland and coastal waters

Kim, Young Woo; Kim, TaeHo; Shin, Jihoon; Lee, Dae‐Seong; Park, Young‐Seuk; Kim, Yeji; Cha, YoonKyung

doi:10.1016/j.ecolind.2022.108737

Cited by 41 publications

(19 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent studies have used machine learning to accurately model and predict aquatic carbon cycling [31,32,65]. Machine learning methods are particularly useful when using surface reflectance over inland waters, where atmospheric corrections can sometimes cause artifacts that lead to nonsensical results in approaches that predict using band ratios, band linear combinations, or other empirical or analytical models [66][67][68]. For example, gradient boosting models, such as used here, were effective in predicting Chl-a using Sentinel 2 and 3 imagery of inland waters [69,70].…”

Section: Statistical Modeling and Scalingmentioning

confidence: 99%

Scaling waterbody carbon dioxide and methane fluxes in the arctic using an integrated terrestrial-aquatic approach

Ludwig

Natali

Schade

et al. 2023

Environ. Res. Lett.

View full text Add to dashboard Cite

In the Arctic waterbodies are abundant and rapid thaw of permafrost is destabilizing the carbon cycle and changing hydrology. It is particularly important to quantify and accurately scale aquatic carbon emissions in arctic ecosystems. Recently available high-resolution remote sensing datasets capture the physical characteristics of arctic landscapes at unprecedented spatial resolution. We demonstrate how machine learning models can capitalize on these spatial datasets to greatly improve accuracy when scaling waterbody CO2 and CH4 fluxes across the YK Delta of south-west AK. We found that waterbody size and contour were strong predictors for aquatic CO2 emissions, attributing greater than two-thirds of the influence to the scaling model. Small ponds (<0.001 km2) were hotspots of emissions, contributing fluxes several times their relative area, but were less than 5% of the total carbon budget. Small to medium lakes (0.001 – 0.1 km2) contributed the majority of carbon emissions from waterbodies. Waterbody CH4 emissions were predicted by a combination of wetland landcover and related drivers, as well as watershed hydrology, and waterbody surface reflectance related to chromophoric dissolved organic matter. When compared to our machine learning approach, traditional scaling methods that did not account for relevant landscape characteristics overestimated waterbody CO2 and CH4 emissions by 26-79% and 8-53% respectively. This study demonstrates the importance of an integrated terrestrial-aquatic approach to improving estimates and uncertainty when scaling C emissions in the arctic.

show abstract

Section: Statistical Modeling and Scalingmentioning

confidence: 99%

Scaling waterbody carbon dioxide and methane fluxes in the arctic using an integrated terrestrial-aquatic approach

Ludwig

Natali

Schade

et al. 2023

Environ. Res. Lett.

View full text Add to dashboard Cite

show abstract

“…The type of land cover can influence variations in Chl-a [33]. As discussed in Section 4.2.3, the distribution of high Chl-a concentrations in the main lake area in three seasons was mainly located in the southeast part of the lake.…”

Section: Ecological Variations Across Lakes Between 2020 and 2021mentioning

confidence: 97%

“…Level-2A products derived from the associated Level-1C products have been systematically generated by the ground segment over Europe since March 2018, and their production was extended to the global scale in December 2018. L1C products can be processed by the Sen2cor [33] and CR2CC processors [24], a process which is somewhat complicated. TOA reflectance has been applied for Chl-a retrieval [10].…”

Section: Validation Of In Situ and Sentinel-2 Reflectancementioning

confidence: 99%

See 1 more Smart Citation

Retrieval of Chlorophyll-a Concentrations Using Sentinel-2 MSI Imagery in Lake Chagan Based on Assessments with Machine Learning Models

Shi

Jiang

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

Chlorophyll-a (Chl-a) is an important characterized parameter of lakes. Monitoring it accurately through remote sensing is thus of great significance for early warnings of water eutrophication. Sentinel Multispectral Imager (MSI) images from May to September between 2020 and 2021 were used along with in-situ measurements to estimate Chl-a in Lake Chagan, which is located in Jilin Province, Northeast China. In this study, the extreme gradient boosting (XGBoost) and Random Forest (RF) models, which had similar performances, were generated by six single bands and six band combinations. The RF model was then selected based on the assessments (R2 = 0.79, RMSE = 2.51 μg L−1, MAPE = 9.86%), since its learning of the input features in the model conformed to the bio-optical properties of Case 2 waters. The study considered Chl-a concentrations in Lake Chagan as a seasonal pattern according to the K-Nearest-Neighbors (KNN) classification. The RF model also showed relatively stable performance for three seasons (spring, summer and autumn) and it was applied to map Chl-a in the whole lake. The research presents a more reliable machine learning (ML) model with higher precision than previous empirical models, as shown by the effects of the input features linked with the biological mechanisms of Chl-a. Its robustness was revealed by the temporal and spatial distributions of Chl-a concentrations, which were consistent with in-situ measurements in the map. This research was capable of revealing the current ecological situation in Lake Chagan and can serve as a reference in remote sensing of inland lakes.

show abstract

“…The interpretability of features is provided by SHAP both globally and locally and considers the interaction synergy between variables while considering the impact of individual variables. Given the excellent interpretability of SHAP for ML models, it has seen extensive use in interpreting disaster susceptibility and ecological environment domains [59]. The purpose of this research was to develop a landslide susceptibility model utilizing the RF and XGBoost algorithms, which was then interpreted and analyzed using the Shapley value estimation method from the SHAP theory of treeSHAP.…”

Section: Shapley Additive Explanationsmentioning

confidence: 99%

Investigation of Landslide Susceptibility Decision Mechanisms in Different Ensemble-Based Machine Learning Models with Various Types of Factor Data

Lu,

Ren,

Yue

et al. 2023

Sustainability

View full text Add to dashboard Cite

Machine learning (ML)-based methods of landslide susceptibility assessment primarily focus on two dimensions: accuracy and complexity. The complexity is not only influenced by specific model frameworks but also by the type and complexity of the modeling data. Therefore, considering the impact of factor data types on the model’s decision-making mechanism holds significant importance in assessing regional landslide characteristics and conducting landslide risk warnings given the achievement of good predictive performance for landslide susceptibility using excellent ML methods. The decision-making mechanism of landslide susceptibility models coupled with different types of factor data in machine learning methods was explained in this study by utilizing the Shapley Additive exPlanations (SHAP) method. Furthermore, a comparative analysis was carried out to examine the differential effects of diverse data types for identical factors on model predictions. The study area selected was Cenxi, Guangxi, where a geographic spatial database was constructed by combining 23 landslide conditioning factors with 214 landslide samples from the region. Initially, the factors were standardized using five conditional probability models, frequency ratio (FR), information value (IV), certainty factor (CF), evidential belief function (EBF), and weights of evidence (WOE), based on the spatial arrangement of landslides. This led to the formation of six types of factor databases using the initial data. Subsequently, two ensemble-based ML methods, random forest (RF) and XGBoost, were utilized to build models for predicting landslide susceptibility. Various evaluation metrics were employed to compare the predictive capabilities of different models and determined the optimal model. Simultaneously, the analysis was conducted using the interpretable SHAP method for intrinsic decision-making mechanisms of different ensemble-based ML models, with a specific focus on explaining and comparing the differential impacts of different types of factor data on prediction results. The results of the study illustrated that the XGBoost-CF model constructed with CF values of factors not only exhibited the best predictive accuracy and stability but also yielded more reasonable results for landslide susceptibility zoning, and was thus identified as the optimal model. The global interpretation results revealed that slope was the most crucial factor influencing landslides, and its interaction with other factors in the study area collectively contributed to landslide occurrences. The differences in the internal decision-making mechanisms of models based on different data types for the same factors primarily manifested in the extent of influence on prediction results and the dependency of factors, providing an explanation for the performance of standardized data in ML models and the reasons behind the higher predictive performance of coupled models based on conditional probability models and ML methods. Through comprehensive analysis of the local interpretation results from different models analyzing the same sample with different sample characteristics, the reasons for model prediction errors can be summarized, thereby providing a reference framework for constructing more accurate and rational landslide susceptibility models and facilitating landslide warning and management.

show abstract

Validity evaluation of a machine-learning model for chlorophyll a retrieval using Sentinel-2 from inland and coastal waters

Cited by 41 publications

References 67 publications

Scaling waterbody carbon dioxide and methane fluxes in the arctic using an integrated terrestrial-aquatic approach

Scaling waterbody carbon dioxide and methane fluxes in the arctic using an integrated terrestrial-aquatic approach

Retrieval of Chlorophyll-a Concentrations Using Sentinel-2 MSI Imagery in Lake Chagan Based on Assessments with Machine Learning Models

Investigation of Landslide Susceptibility Decision Mechanisms in Different Ensemble-Based Machine Learning Models with Various Types of Factor Data

Contact Info

Product

Resources

About