The marine environment in the South Pacific Island Countries (SPICs) is sensitive and vulnerable to climate change. While large-scale changes in this region are well-documented, national-scale analyses that address management needs remain limited. This study evaluated the performance of satellite-derived datasets—including sea surface temperature (SST), sea surface salinity (SSS), Secchi disk depth (SDD), chlorophyll-a (Chl-a), net primary production (NPP), and sea level anomaly (SLA)—against in situ observations, and analyzed their spatial and temporal variability across 12 national Exclusive Economic Zones (EEZs) during 1998–2023. Validation results presented that current satellite datasets could provide applicable information for EEZ-scale analyses. In the past decades, the SPICs experienced a general increase in SST and SLA, accompanied by marked within-EEZ heterogeneity in Chl-a and NPP variations, with Papua New Guinea exhibiting the largest within-EEZ inter-annual variability. In addition to monitoring, satellite data would help to constrain the uncertainty of CMIP6 results in the SPICs, subject to the accuracy of specific products. By 2100, Nauru might experience the most vulnerable EEZ, while the marine environment in the French Polynesian EEZ can keep relatively stable among all 12 EEZs. Meanwhile, CMIP6 projections in the Southeastern EEZs are more sensitive to satellite-based constraints, showing pronounced adjustments. Our results demonstrate the potential of combining validated satellite data with CMIP6 models to provide national-scale decision support for climate adaptation and marine resource management in the SPICs.

Dissolved inorganic nitrogen(DIN)and dissolved inorganic phosphorus(DIP)are the two dominant nutrients influencing seawater quality.Due to the non-optically active nature of nutrients and their regional variability,relying solely on optical data is inadequate for achieving high-precision remote sensing retrieval in complex marine environments.We developed a novel remote sensing algorithm for DIN and DIP using MODIS remote sensing reflectance(Rrs)products and the XGBoost machine learning framework.Beyond optical inputs,our model integrates sea surface temperature(SST)and spatiotemporal information,including a shoreline-based pixel location descriptor,which significantly enhances model performance.We generated monthly average distributions of DIN and DIP concentrations across China's coastal waters from 2012 to 2022.The findings highlight extensive high-nutrient zones in the Bohai Bay and Changjiang River(Yangtze River)Estuary-Hangzhou Bay regions,with a notable declining trend in nutrient concentrations.The Zhujiang River(Pearl River)Estuary also exhibits elevated nutrient levels,albeit with minimal changes.This study pioneers the incorporation of dual-coordinate information in nutrient retrieval for complex marine environments,significantly improving model accuracy and addressing stripping artifacts associated with single-coordinate systems.Moreover,the results provide unprecedented spatiotemporal insights into nutrient distributions in China's coastal waters,offering valuable support for marine environmental management and policy-making.

Observations of mesoscale eddy structures rely heavily on satellite altimetry data. However, due to altimetry's coarse spatial resolution, the fine structure of eddy dynamics remains mysterious. Using high spatiotemporal resolution observations from the Geostationary Ocean Color Imager (GOCI), we reveal the fine structure and hourly dynamics of the eddy surface flow velocities, as well as the horizontal eddy transport processes. The sea surface flow field retrieved by the dense optical flow algorithm from the GOCI was consistent with the results derived from satellite altimetry data but had a much higher spatial resolution (500 m), which makes it feasible to capture the fine structure of eddy dynamics. Additionally, the hourly observations exposed rapid variations of the eddy kinetic energy and the horizontal advection transport of surface phytoplankton. These fine-scale and frequent GOCI observations increase the understanding of the dynamics and mass transport in mesoscale eddies.

Assessing sea surface nitrate (SSN) concentrations and dynamics is crucial for understanding marine ecosystem health, yet optical remote sensing of SSN remains challenging because of the lack of distinct spectral features. While various global-scale SSN regression and machine learning algorithms based on SSN-environment variable relationships have been developed, the prediction accuracy and spatiotemporal resolution of their applications continue to face limitations. Additionally, there has been relatively little reporting on the interannual variability of global SSN in previous studies. Here we aim to enhance the accuracy and spatial resolution of SSN retrievals by developing improved regression and machine learning models, enabling the generation of global daily similar to 8 km SSN products from satellite and model data. To construct the empirical regression models, the global ocean was divided into five regions on the basis of the relationship between sea surface temperature (SST) and SSN: 80 degrees S to 40 degrees N, the North Pacific, the North Atlantic, the Arabian Sea, and the eastern equatorial Pacific. After adding SSN-related physical variables, high-accuracy regional empirical models are developed, with root mean square deviations (RMSDs) of 1.641, 2.701, 1.221, 1.298, and 2.379 mu mol/kg for the studied regions. For the machine learning models, seven algorithms, namely, extremely randomized trees (ET), multilayer perceptron (MLP), stacking random forest (SRF), Gaussian process regression (GPR), support vector machine (SVM), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost) algorithms, were tested. After modeling, validation, and extensive tests using independent cruise dataset, the XGBoost model outperformed others (RMSD = 1.189 mu mol/kg) and bypassed the need for regional segmentation. Mechanistic analysis revealed the driving variables influencing SSN in both regional empirical and XGBoost models, improving interpretability. Comparative validation confirmed that our models surpass traditional approaches in accuracy and applicability, demonstrating their potential to advance global SSN monitoring. Using XGBoost-derived products, we find a slight weak decreasing trend in SSN over 23 years. The proposed robust and explainable SSN retrieval models have the potential to assist in ocean environmental management.

Excessive nitrogen concentrations may degrade coastal ecosystems, potentially causing harmful algae blooms and hypoxia. Total nitrogen (TN) and dissolved inorganic nitrogen (DIN) are key indicators for assessing external nitrogen loads and eutrophication. However, long-term, high-resolution data on coastal nitrogen dynamics remain limited because of sparse monitoring. To address this gap, Landsat-5/7/8/9 imagery (1986-2024) was integrated with in situ measurements to develop and compare multiple inversion models. A Gaussian process regression model (training/test root mean square error: TN = 0.16/0.14 mg/L; DIN = 0.09/0.09 mg/L) was selected to reconstruct nitrogen concentrations in Daya Bay. Throughout the study period, it revealed average TN and DIN concentrations of 0.50 and 0.23 mg/L respectively, with notable spatial clustering near estuaries and coastal waters. Change point analysis revealed abrupt shifts in 2013 (TN) and 1998 (DIN), but the Mann-Kendall test results indicated no significant long-term monotonic trends. Integrated analysis with meteorological data suggested that extreme rainfall events cause short-term spikes in nutrient levels, whereas seasonal regulation causes higher concentrations during the dry season. Comparative analyses of functional zones revealed that estuarine areas exhibited the highest nitrogen concentrations, with average TN and DIN concentrations of 0.75 and 0.44 mg/L, respectively. The variations among different zones were associated with changes in the coastline, land use, wastewater discharge, and fertilizer application. Attribution analysis based on the SHapley Additive exPlanations method indicated that domestic wastewater is the primary driver. This study provides a framework for assessing coastal nitrogen dynamics, with concurrent TN/DIN monitoring supporting future precision management of nitrogen sources in eutrophication control.

Monitoring the global ocean net primary productivity (NPP) has been the primary objective of satellite ocean color remote sensing since the launch of the world's first satellite ocean color sensor in 1978. However, considerable discrepancies persist in current satellite NPP estimations, and the use of limited in situ data is one of the key challenges for accurate NPP estimation. Here, a global depth‐resolved NPP profile (NPP re ) data set was constructed on the basis of 10 years of biogeochemical Argo (BGC‐Argo) measurements and a tuned carbon‐based productivity model (BGC_CbPM). On the basis of this data set and the XGBoost machine learning model, a global oceanic NPP re remote sensing inversion model (XGBoost_CbPM) was established for NPP estimation from Moderate Resolution Imaging Spectroradiometer (MODIS)/Aqua data. Validation with 14 independent samples revealed a coefficient of determination of 0.87 and a mean absolute percentage deviation of 12.52% between the model predictions and in situ measurements. In addition, the results obtained via the XGBoost_CbPM model suitably agreed with the in situ measurements at two time series stations, namely BATS and HOT. In particular, the time series changes in the NPP derived by the XGBoost_CbPM model at the BATS and HOT stations were better than those of the original satellite products based on the carbon‐based productivity model. More importantly, as a depth‐resolved model, the XGBoost_CbPM model can provide NPP profiles that are superior to those of traditional NPP models, which can be used to estimate only the water column‐integrated NPP. This study underscores the significant contribution of BGC‐Argo measurements in enhancing the satellite estimation of the global ocean NPP.

Recently, Randomly Wired Neural Networks (RWNNs) using random graphs for Convolutional Neural Network (CNN) construction have shown efficient layer connectivity, but may limit depth, affecting approximation, generalization, and robustness. In this work, we increase the depth of graph-structured CNNs while maintaining efficient pathway usage, which is achieved by building a feature-extraction backbone with a depth-first search, employing edges that have not been traversed for parameter-efficient skip connections. The proposed Efficiently Pathed Deep Network (EPDN) reaches maximum graph-based architecture depth without redundant node use, ensuring feature propagation with reduced connectivity. The deep structure of EPDN, coupled with its efficient pathway usage, allows for a nuanced feature extraction. EPDN is highly beneficial for processing remote sensing images, as its performance relies on the ability to resolve intricate spatial details. EPDN facilitates this by preserving low-level details through its deep and efficient skip connections, allowing for enhanced feature extraction. Additionally, the remote-sensing-adapted EPDN variant is akin to a special case of a multistep method for solving an Ordinary Differential Equation (ODE), leveraging historical data for improved prediction. EPDN outperforms existing CNNs in generalization and robustness on image classification benchmarks and remote sensing tasks. The source code is publicly available at https://github.com/AnonymousGithubLink/EPDN.

In this study, a remote sensing inversion model for marine nutrients was developed specifically for the coastal waters under the jurisdiction of Zhejiang Province, China, by integrating in situ measurements and satellite remote sensing data. Utilizing support vector machines (SVM), we constructed monthly average products of dissolved inorganic nitrogen (DIN) and active phosphate (PO4) concentrations based on Sentinel-3 imagery from 2018 to 2024. The model inputs included surface-layer nutrient data collected from fixed monitoring stations provided by the Zhejiang Marine Ecology Center, covering the years 2017 to 2022. Spatial analysis revealed a consistent nearshore-to-offshore gradient in nutrient concentrations, with significantly higher values concentrated in the Hangzhou Bay and Yangtze River Estuary regions. Temporally, both DIN and PO4 exhibited seasonal variability, peaking in winter and reaching minima in summer. Multi-source data analysis reveals that precipitation, river discharge, sea surface salinity, and wind conditions collectively influence the spatial distribution of nutrients in the marine environment. This work not only enhances the spatial and temporal coverage of nutrient monitoring in Zhejiang's coastal waters but also demonstrates the potential of satellite remote sensing as a valuable tool for large-scale marine environmental assessment. However, the applicability of the model to other sea regions remains limited due to regional specificity and the reliance on surface-layer nutrient indicators.

Acquiring a large number of in situ water spectral measurements is fundamental for constructing water color remote-sensing retrieval models and validating the accuracy of water color remote-sensing products. However, traditional manual site-based water spectral measurements are time-consuming and labor-intensive, resulting in an insufficient number of in situ water spectral samples to date. To resolve this issue, this study develops an unmanned aerial vehicle-based hyperspectral remote-sensing reflectance measurement system (UAV-RRS) capable of continuous on-the-move water spectral measurements. This paper provides a detailed introduction to the system components and conducts precise experiments on the correction and calibration of the spectral sensors. Using this system, an in situ–UAV–satellite multi-source remote-sensing reflectance comparison experiment was conducted in the middle reaches of the Qiantang River, East China, to evaluate the accuracy and reliability of UAV-RRS and extend the analysis to satellite data across different spatial scales. The results demonstrate that, in small-scale water bodies, UAV-RRS achieves higher spatial precision and spectral accuracy, offering a valuable solution for high-precision, low-altitude continuous water body observations.

The successful launch of Landsat-9 in 2020 ensures the continuity of Landsat Earth observation data. However, before Landsat-9 data can be effectively used in conjunction with Landsat-8, it is necessary to consider the differences between data from different sensors. This study utilized the Reduced Major Axis (RMA) regression to compare simulated reflectance data derived from the spectral response functions and spectral library data, showing minor spectral response differences between the two sensors (RMA slope close to 1, RMSD <= 0.0005); The consistency of more than 100 million pairs of top-of-atmosphere (TOA) and surface reflectance (SR) data from two sensors extracted based on Google Earth Engine was then evaluated using RMA and ordinary least squares (OLS) regression, and a transfer function developed using OLS regression was provided (R-2 > 0.84); The results showed that the atmospheric state has a significant effect on the continuity of the sensors, especially in the shorter wavelength bands; there are differences in the TOA and SR data, with the absolute mean difference |MD| <= 0. 0011, root mean square difference |RMSD| <= 0.0278 and mean relative difference |MRD| <= 0.47%; In addition, the influence of seasonal factors on the consistency of the two sensors was investigated, and the corresponding transformation functions were provided; The OLS transformation functions developed in this paper were validated on over 10 million samples in Africa, with R-2 > 0.86 for TOA, R-2 > 0.93 for SR, and R-2 > 0.91 for the SR seasonal transformations.

Accurate prediction of the spatiotemporal distribution of chlorophyll-a (Chl-a) is essential for evaluating marine ecosystem health and predicting ecological disasters. Current methods struggle to capture short-term variability and periodic trends in Chl-a, especially in noise-prone coastal regions. This study aims to enhance the prediction of marine Chl-a concentrations by introducing the chlorophyll-a concentration prediction model (ChlaPM), which was developed on the basis of a convolutional long short-term memory (ConvLSTM) network. The model integrates recent spatiotemporal feature extraction (RSTFE), periodic feature extraction (PFE), and denoising fusion (DNF) modules to effectively capture short-term spatiotemporal changes and periodic variations in Chl-a concentrations. In this study, the performance of ChlaPM in single-step and multistep predictions was evaluated using monthly average Chl-a remote sensing data spanning 1998–2023. The results indicate that compared with the RSTFE model, the ChlaPM model achieves substantial reductions in the root mean square error (RMSE) of 53.84%, 53.58%, and 49.70% for predicting Chl-a concentrations 1 month, 3 months, and 6 months into the future, respectively. These findings highlight the effectiveness of ChlaPM in addressing short-term variability and periodic trends and significantly enhances the accuracy of Chl-a prediction. Future work will focus on integrating additional relevant marine variables into the prediction model to further improve its prediction capabilities.

Satellite ocean color remote sensing plays a crucial role in monitoring marine environment at both regional and global scales. However, due to the reduced accuracy of atmospheric correction models under large solar zenith angles (>= 70 degrees), publicly available satellite ocean color products lack valid datasets for high-latitude oceans (>= 50 degrees S or >= 50 degrees N) during winter. Based on a neural network atmospheric correction model designed for high solar zenith angle observation environments (which used a Rayleigh scattering lookup table generated by PCOART-SA to compute Rayleigh scattering radiance and a neural network model to invert remote sensing reflectance from Rayleigh-corrected radiance), this study has established a monthly ocean color product dataset for high-latitude oceans, named NN-LAT50, covering the period from 2003 to 2020. We validated the accuracy of the ocean color products in NN-LAT50 dataset using multiple in situ datasets, and the results indicated that NN-LAT50 had more reliable and accurate retrievals compared to the NASA released ocean color products in high latitude oceans. Furthermore, during autumn and winter, coverage of the NN-LAT50 dataset far exceeds that of products released by NASA. For instance, during the winter in the Southern Hemisphere, the coverage rates are 3.02% for MODIS/Aqua, 21.59% for VIIRS, and 1.74% for OLCI, while the NN-LAT50 dataset maintains a coverage rate of 38.64%. This study is the first to establish a long-term (2003-2020) ocean color product dataset covering high-latitude oceans during winter, which can significantly enhance the observation of ecological changes in polar and subpolar oceans.

The spectral power exponent of the particulate backscattering coefficient (bbp), η, is a crucial parameter for assessing particle size distribution in the ocean, particularly for high-precision estimation of marine biogeochemical cycles. The development of remote sensing inversion models for η has been constrained by insufficient in situ data or faulty model mechanisms. In this study, biogeochemical Argo (BGC-Argo) data were effectively utilized to address the limitations of traditional observations, and the superior XGBoost method was employed to construct a global ocean η remote sensing inversion model (XGB model). In the 5-fold cross-validation, the Pearson correlation coefficient (R), root mean square error (RMSE), mean absolute error (MAE), and median absolute percentage deviation (MAPD) of the XGB model remained stable at approximately 0.65, 0.3, 0.25, and 16%, respectively, demonstrating high robustness. Via independent sample verification and comparison with major inversion models from the literature, the XGB model exhibited a high R value of 0.66 and the lowest RMSE, MAPD, and MAE, with values of 0.33, 16.81%, and 0.25, respectively, indicating its superior inversion capability. High η values were primarily found in oligotrophic regions and the iron-limited Southern Ocean, whereas low values were observed in coastal waters, exhibiting seasonal variation. The distribution of η in other ocean areas remained relatively stable, with nonsignificant seasonal changes. This study demonstrates the significant contributions of BGC-Argo observations in enhancing remote sensing estimation of global ocean η values, thereby providing valuable support for biogeochemical cycling research.

Achieving high-precision extraction of sea islands from high-resolution satellite remote sensing images is crucial for effective resource development and sustainable management. Unfortunately, achieving such accuracy for sea island extraction presents significant challenges due to the presence of extensive background interference. A more widely applicable noise-tolerant matched filter (NTMF) scheme is proposed for sea island extraction based on the MF scheme. The NTMF scheme effectively suppresses the background interference, leading to more accurate and robust sea island extraction. To further enhance the accuracy and robustness of the NTMF scheme, a neural dynamics algorithm is supplemented that adds an error integration feedback term to counter noise interference during internal computer operations in practical applications. Several comparative experiments were conducted on various remote sensing images of sea islands under different noisy working conditions to demonstrate the superiority of the proposed neural dynamics algorithm-assisted NTMF scheme. These experiments confirm the advantages of using the NTMF scheme for sea island extraction with the assistance of neural dynamics algorithm.

Intelligent detection and recognition of ship targets is the basis of naval battlefield situational assessment and threat estimation. Its core task is to determine whether there is a ship target in the image and to detect, identify, and locate the ship target, which also has broad application prospects in fisheries management, maritime rescue, maritime traffic management, and marine environment monitoring. To this end, a multi-scale ship target detection improved algorithm for remote sensing images is proposed based on the YOLOv5 model, called CPE-YOLO. Firstly, a multi-scale attention fusion module based on coordinate position is proposed to solve the problem of missing spatial perception ability for ship detection in optical remote sensing images, to process the multi-scale coordinate information of feature maps, and to effectively establish long-range channel dependency among multi-scale channel attention. Secondly, a more refined feature pyramid network is constructed to effectively mitigate the impact of target scale variation due to remote sensing images on model performance, and to provide richer feature information for feature fusion layer regression localization. Ultimately, the decoupled head is integrated into the YOLO Head to disentangle the classifier and regressor, facilitating accelerated network convergence and concurrent enhancement of network detection accuracy. Finally, thorough tests on the HRSC2016, ASDC and SIGE datasets are presented to support our methodology. The results show that the performance of our proposed methods is better than other existing methods, with an mAP of 94.0%, 76.5%, and 98.7% on the HRSC2016, ASDC, and SIGE datasets. Moreover, comparative assessments against alternative deep learning methodologies affirm the sustained superiority of the enhanced method across overall performance metrics.

The adjacency effect (AE) caused by the surrounding land cannot be ignored for satellite remote sensing of coastal and inland waters, especially for rivers with a narrow width. However, there is currently a lack of comprehensive understanding of AE for rivers, much less the precise correction methods for these effects. Here, a 3-D vector radiative transfer model for a 3-D radiative transfer model for a nonuniform underlying surface (NUS-MC) was developed. The ability of the NUS-MC to accurately simulate AEs in the case of Rayleigh scattering was validated by comparing its results with those from existing models for both uniform and nonuniform underlying surface cases. The mean absolute percentage deviations (MAPDs) for the simulated radiance are within 0.14% for a uniform water underlying and within 0.34% for a uniform land underlying surface. Based on the NUS-MC, AEs on Rayleigh scattering radiance for river waters under various conditions were systematically quantified. The results show that rho(AE) decreases with increasing solar zenith angle (SZA) but increases with increasing view zenith angle (VZA). Even for a large river, 10 km across, at a wavelength of 443 nm, the mean rho(AE) at the river center is 3.97% at an SZA of 0 degrees when the land albedo is 0.1, increasing to 18.19% and 33.37% at land albedos of 0.3 and 0.5, respectively. Overall, AEs significantly affect Rayleigh scattering even in rivers with widths of up to 10 km. In addition, the angle between the river channel and the observation plane as well as the distance from the riverbank to the point of view in the river can also affect the AEs. These findings suggest that the impact of land AEs on Rayleigh scattering must be thoroughly considered to retrieve water-leaving radiance in rivers accurately, and it is essential to develop atmospheric correction methods capable of removing AEs.

Large rivers without hydrological data from remote sensing observations have recently become a hot research topic. The Irrawaddy River is among the major tropical rivers worldwide; however, published hydrological data on this river have rarely been obtained in recent years. In this paper, based on the existing measured the total suspended matter flux (FTSM) and discharge data for the Irrawaddy River, an inversion model of the total suspended matter concentration (CTSM) is constructed for the Irrawaddy River, and the CTSM and FTSM from 1990 to 2020 are estimated using the L1 products of Landsat-8 OLI/TIRS and Landsat-5 TM. The results show that over the last 30 years, the FTSM of the Irrawaddy River decreased at a rate of 3.9 Mt/yr, which is significant at the 99% confidence interval. An increase in the vegetation density of the Irrawaddy Delta has increased the land conservation capacity of the region and reduced the inflow of land-based total suspended matter (TSM). The FTSM of the Irrawaddy River was estimated by fusing satellite data and data measured at hydrological stations. The research method employed in this paper provides a new supplement to the existing hydrological data for large rivers.

The traditional atmospheric correction models employed with the near-infrared iterative schemes inaccurately estimate aerosol radiance at high solar zenith angles (SZAs), leading to a substantial loss of valid products for dawn or dusk observations by the geostationary satellite ocean color sensor. To overcome this issue, we previously developed an atmospheric correction model suitable for open ocean waters observed by the first geostationary satellite ocean color imager (GOCI) under high SZAs. This model was constructed based on a dataset from stable open ocean waters, which makes it less suitable for coastal waters. In this study, we developed a specialized atmospheric correction model (GOCI-II-NN) capable of accurately retrieving the water-leaving radiance from GOCI-II observations in coastal oceans under high SZAs. We utilized multiple observations from GOCI-II throughout the day to develop the selection criteria for extracting the stable coastal water pixels and created a new training dataset for the proposed model. The performance of the GOCI-II-NN model was validated by in-situ data collected from coastal/shelf waters. The results showed an Average Percentage Difference (APD) of less than 23% across the entire visible spectrum. In terms of the valid data and retrieval accuracy, the GOCI-IINN model was superior to the traditional near-infrared and ultraviolet atmospheric correction models in terms of accurately retrieving the ocean color products for various applications, such as tracking/monitoring of algal blooms, sediment dynamics, and water quality among other applications.

Shallow water benthic habitats have been significantly degraded and seriously threatened by intensifying climate changes and anthropogenic stressors. Benthic reflectance (Rbλ) of optically shallow waters (OSWs) is a key parameter for remote sensing of benthic habitats’ composition and health status. The remote sensing reflectance (Rrsλ) just above the water surface contains the coupled spectral information of the water column and seabed in OSWs, which makes our effort challenging to separate the signals between seabed and water column properties. Due to too many unknowns of the water column and seabed optical properties, the current approaches are inaccurate or inadequate to retrieve benthic reflectance spectra over a large area from remotely sensed data. In this study, we proposed an enhanced large-scale benthic reflectance (LSBR) retrieval model from satellite data in OSWs with only input of Rrsλ. A new spectral library of benthic reflectance was built upon multi-source reflectance datasets obtained from several open-source field measurements. Then, the specific relationship between Rb443nm and Rb490nm was developed. By combining a semi-analytical benthic reflectance (SABR) model based on radiative transfer simulations and this relationship, water-column chlorophyll concentration (chl) was determined. Moreover, large-scale water depth was estimated by Rrsλ through satellite-derived bathymetry (SDB) technologies. Finally, benthic reflectance was retrieved with the inputs of the satellite-derived Rrsλ, chl, and water depth. In this way, the LSBR model fulfills the Rbλ retrieval with only input of satellite-derived Rrsλ. The new model was successfully applied to Sentinel-2/MSI data in the seagrass region of Xincun Bay and the coral region of Huaguang Reef and further verified by using field measurements and results of the along-track benthic reflectance (ATBR) retrieval model, which combined active lidar and passive high-resolution satellite images. Additionally, to gain insights into the long-term evolution of benthic communities, a typical application of the LSBR model for detecting benthic changes was conducted using eight years of Sentinel-2/MSI data in Heron Reef, depicting a similar consistency between the probable deterioration and restoration region and the sea surface temperature (SST) anomalies. The results demonstrated the robustness and accuracy of the LSBR model in retrieving benthic reflectance at large scales. The LSBR model should be helpful for further investigating benthic changes which can enhance the ability to monitor and assess the health status of submerged ecosystems in OSWs.