Multispectral Instrument Research Articles

Integrating Multisource Data and Machine Learning for Supraglacial Lake Detection: Implications for Environmental Management and Sustainable Development Goals in High Mountainous Regions

The accurate detection and monitoring of supraglacial lakes in high mountainous regions are crucial for understanding their dynamic nature and implications for environmental management and sustainable development goals. In this study, we propose a novel approach that integrates multisource data and machine learning techniques for supra-glacial lake detection in the Passu Batura glacier of the Hunza Basin, Pakistan. We extract pertinent features or parameters by leveraging multisource datasets such as radar backscatter intensity VH and VV parameters from Sentinel-1 Ground Range Detected (GRD) data, near-infrared (NIR), NDWI_green, NDWI_blue parameters from Sentinel-2 Multi-spectral Instrument(MSI) data, and surface slope, aspect, and elevation parameters from topographic data. The entire dataset is partitioned into training and testing sets, with machine learning models including the artificial neural network (ANN), the support vector machine (SVM), logistic regression (LR), random forest (RF), and K-nearest neighbour (KNN) trained on the training data (70%). Accuracy assessment employs testing data and involves the evaluation of metrics such as ROC curves and confusion matrices. The best-performing model, ANN, is validated against manually digitized lake polygons derived from Sentinel-2 and Google Earth Pro imagery. Furthermore, the digitized lake polygons are used to analyze glacial lake dynamics from 2016 to 2022. Key findings of this research presented that the NDWI_green, Sigma0_VH, and elevation are the most significant predictors in detecting supra-glacial lakes. Among the various trained and evaluated models, the Artificial Neural Network (ANN) achieved the highest performance (accuracy: 95%, AUC: 0.99) and accurately mapped supra-glacial lakes regardless of their small size. The findings have significant implications for understanding glacial lake behavior in the context of climate change and informing future research and monitoring efforts.

A Copernicus-based evapotranspiration dataset at 100 m spatial resolution over four Mediterranean basins

Abstract. Evapotranspiration (ET) is responsible for regulating the hydrological cycle, with a relevant impact on air humidity and precipitation that is particularly important in the context of acute drought events in recent years. With the intensification of rainfall deficits and extreme heat events, the Mediterranean region requires regular monitoring to enhance water resource management. Even though remote sensing provides spatially continuous information for estimating ET on large scales, existing global products with spatial resolutions ≥ 0.5 km are insufficient for capturing spatial detail at a local level. In the framework of ESA's 4DMED-Hydrology project, we generate an ET dataset at both high spatial and high temporal resolutions using the Priestley–Taylor Two-Source Energy Balance (TSEB-PT) model driven by Copernicus satellite data. We build an automatic workflow to generate a 100 m ET product by combining data from Sentinel-2 (S2) MultiSpectral Instrument (MSI) and Sentinel-3 (S3) land surface temperature (LST) with ERA5 climate reanalysis derived within the period 2017–2021 over four Mediterranean basins in Italy, Spain, France, and Tunisia (Po, Ebro, Hérault, and Medjerda). First, original S2 data are pre-processed before deriving 100 m inputs for the ET estimation. Next, biophysical variables, like leaf area index and fractional vegetation cover, are generated, and then they are temporally composited within a 10 d window according to S3 acquisitions. Consequently, decadal S2 mosaics are used to derive the remaining TSEB-PT inputs. In parallel, we sharpen 1 km S3 by exploiting the dependency between coarse-resolution LST and 100 m S2 reflectances using a decision tree algorithm. Afterwards, climate forcings are utilized to model energy fluxes and then for daily ET retrieval. The daily ET composites demonstrate reasonable TSEB-PT estimates. Based on the validation results against eight eddy covariance (EC) towers between 2017 and 2021, the model predicts 100 m ET with an average RMSE of 1.38 mm d−1 and a Pearson coefficient equal to 0.60. Regardless of some constraints mostly related to the high complexity of EC sites, TSEB-PT can effectively estimate 100 m ET, which opens up new opportunities for monitoring the hydrological cycle on a regional scale. The full dataset is freely available at https://doi.org/10.48784/b90a02d6-5d13-4acd-b11c-99a0d381ca9a, https://doi.org/10.48784/fb631817-189f-4b57-af6a-38cef217bad3, https://doi.org/10.48784/70cd192c-0d46-4811-ad1d-51a09734a2e9, and https://doi.org/10.48784/7abdbd94-ddfe-48df-ab09-341ad2f52e47 for the Ebro, Hérault, Medjerda, and Po catchments, respectively (Bartkowiak et al., 2023a–d).

Monitoring Salinity in Inner Mongolian Lakes Based on Sentinel-2 Images and Machine Learning

Salinity is an essential parameter for evaluating water quality and plays a crucial role in maintaining the stability of lake ecosystems, particularly in arid and semi-arid climates. Salinity responds to changes in climate and human activity, with significant impacts on water quality and ecosystem services. In this study, Sentinel-2A/B Multi-Spectral Instrument (MSI) images and quasi-synchronous field data were utilized to estimate lake salinity using machine learning approaches (i.e., XGB, CNN, DNN, and RFR). Atmospheric correction for MSI images was tested using six processors (ACOLITE, C2RCC, POLYMER, MUMM, iCOR, and Sen2Cor). The most accurate model and atmospheric correction method were found to be the extreme gradient boosting tree combined with the ACOLITE correction algorithm. These were used to develop a salinity model (N = 70, mean absolute percentage error = 9.95%) and applied to eight lakes in Inner Mongolia from 2016 to 2024. Seasonal and interannual variations were explored, along with an examination of potential drivers of salinity changes over time. Average salinities in the autumn and spring were higher than in the summer. The highest salinities were observed in the lake centers and tended to be consistent and homogeneous. Interannual trends in salinity were evident in several lakes, influenced by evaporation and precipitation. Climate factors were the primary drivers of interannual salinity trends in most lakes.

Tracking the impact of typhoons on maize growth and recovery using Sentinel-1 and Sentinel-2 data: A case study of Northeast China

The increasing frequency of typhoon events, attributed to global climate change, has significantly affected agricultural production, predominantly resulting in substantial negative consequences. Accurate and timely assessment of crop damage is crucial for understanding economic implications, devising effective agricultural strategies, and enhancing resilience amid mounting climate uncertainties. This study investigates the utility of Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 MultiSpectral Instrument (MSI) data in tracking maize damage severity following typhoon events. By employing continuous field sampling techniques and conducting visual interpretation of high-resolution remote sensing imagery, sample sets representing a spectrum of maize damage severity were systematically established. The application of time series analysis on Sentinel data enables a comprehensive exploration of spectral and polarization responses, providing insights into the correlation with maize damage severity. Segmentation of the maize damage timeline into pre-disaster, disaster, and recovery periods, coupled with optimization of relevant feature parameters, was undertaken to bolster monitoring precision. Leveraging the Google Earth Engine (GEE) cloud platform, a Random Forest algorithm was used to develop a model for monitoring maize damage severity across different post-typhoon periods, yielding maps delineating the distribution and magnitude of maize damage in Northeast China. Results indicate that integrating spectral indices from the pre-disaster phase with backscatter variations of polarization bands during various post-typhoon periods enhances maize damage assessment. Maize damage severity is notably elevated during the disaster period, achieving an overall accuracy of 87.22 %. While mitigated during the recovery phase, localized exacerbation occurs in severely affected regions, yielding an overall accuracy of 88.54 %. Analysis incorporating terrain and meteorological data reveals that post-typhoon maize disasters predominantly occur in low-lying and flat areas, with meteorological factors, particularly maximum wind speed and daily cumulative precipitation, exerting significant influence on damage severity. This study underscores the critical role of SAR and optical data fusion in elucidating typhoon-induced crop damage dynamics, thereby providing essential insights for proactive mitigation strategies against future agricultural losses.

The first global catalogue of gas flaring sources derived from a multi-temporal time series of OLI and MSI daytime data: the DAFI v2 algorithm

Abstract Gas flaring (GF) is an industrial phenomenon requiring a special attention for the serious impacts on environment, climate and human health. To analyses and map GF sites (GFs), the Daytime Approach for GF Investigation (DAFI), which is based on daytime data from the Operational Land Imager (OLI) aboard the Landsat-8 satellite, has recently been developed. The GFs catalogue from the DAFI system, spanning over the years 2013–2021, represents the first worldwide GF product generated from mid-high spatial resolution data (30 m). The DAFI version 2 (v2), also including information from the Multispectral Instrument (MSI) aboard Sentinel-2 satellites, already showed an improved capacity in the mapping of GFs over Iran and Iraq. In this work, the new catalogue arising from the porting of DAFI v2 at global scale is presented and discussed. By using a parallel workflow, based on a loop tiling scheme running in Google Earth Engine, DAFI v2 allowed us to extend the GF inventory globally up to 2023, by analyzing ∼4 million of OLI/OLI-2 (Collection 2) and MSI imagery. Results of this study show: (i) an increased DAFI v2 capacity, of about 41%, in detecting GFs compared to the previous DAFI version; (ii) a downward trend (∼4.6%) in terms of GF sites detected from 2021 to 2023. These findings demonstrated that DAFI v2, which will also include the estimates of radiative power at different spatial scales, may contribute in assessing and monitoring the GFs activities and in evaluating the effectiveness of the greenhouse gas-emission reduction strategies at global, continental, national and local scales.

A First Extension of the Robust Satellite Technique RST-FLOOD to Sentinel-2 Data for the Mapping of Flooded Areas: The Case of the Emilia Romagna (Italy) 2023 Event

Extreme meteorological events hit our planet with increasing frequency, resulting in an ever-increasing number of natural disasters. Flash floods generated by intense and violent rains are among the most dangerous natural disasters that compromise crops and cause serious damage to infrastructure and human lives. In the case of such a kind of disastrous events, timely and accurate information about the location and extent of the affected areas can be crucial to better plan and implement recovery and containment interventions. Satellite systems may efficiently provide such information at different spatial/temporal resolutions. Several authors have developed satellite techniques to detect and map inundated areas using both Synthetic Aperture Radar (SAR) and a new generation of high-resolution optical data but with some accuracy limits, mostly due to the use of fixed thresholds to discriminate between the inundated and unaffected areas. In this paper, the RST-FLOOD fully automatic technique, which does not suffer from the aforementioned limitation, has been exported for the first time to the mid–high-spatial resolution (20 m) optical data provided by the Copernicus Sentinel-2 Multi-Spectral Instrument (MSI). The technique was originally designed for and successfully applied to Advanced Very High Resolution Radiometer (AVHRR), Moderate Resolution Imaging Spectroradiometer (MODIS), and Visible Infrared Imaging Radiometer Suite (VIIRS) satellite data at a mid–low spatial resolution (from 1000 to 375 m). The processing chain was implemented in a completely automatic mode within the Google Earth Engine (GEE) platform to study the recent strong flood event that occurred in May 2023 in Emilia Romagna (Italy). The outgoing results were compared with those obtained through the implementation of an existing independent optical-based technique and the products provided by the official Copernicus Emergency Management Service (CEMS), which is responsible for releasing information during crisis events. The comparisons carried out show that RST-FLOOD is a simple implementation technique able to retrieve more sensitive and effective information than the other optical-based methodology analyzed here and with an accuracy better than the one offered by the CEMS products with a significantly reduced delivery time.

Enhancing Surface Water Monitoring through Multi-Satellite Data-Fusion of Landsat-8/9, Sentinel-2, and Sentinel-1 SAR

Long revisit intervals and cloud susceptibility have restricted the applicability of earth observation satellites in surface water studies. Integrating multiple satellites offers potential for more frequent observations, yet combining different satellite sources, particularly optical and SAR satellites, presents complexities. This research explores the data-fusion potential and limitations of Landsat-8/9 Operational Land Imager (OLI), Sentinel-2 Multispectral Instrument (MSI), and Sentinel-1 Synthetic Aperture (SAR) satellites to enhance surface water monitoring. By focusing on segmented surface water images, we demonstrate that combining optical and SAR data is generally effective and straightforward using a simple statistical thresholding algorithm. Kappa coefficients(κ) ranging from 0.80 to 0.95 indicate very strong harmony for integration across reservoirs, lakes, and river environments. In vegetative environments, integration with S1SAR shows weak harmony, with κ values ranging from 0.27 to 0.45, indicating the need for further studies. Global revisit interval maps reveal significant improvement in median revisit intervals from 15.87 to 22.81 days using L8/9 alone, to 4.51 to 7.77 days after incorporating S2, and further to 3.48 to 4.62 days after adding S1SAR. Even during wet season months, multi-satellite fusion maintained the median revisit intervals to less than a week. Maximizing all available open-source earth observation satellites is integral for advancing studies requiring more frequent surface water observations, such as flood, inundation, and hydrological modeling.

Water quality hotspot identification using a remote sensing and machine learning approach: A case study of the River Ganga near Varanasi

Turbidity (Turb) and Chlorophyll-a (Chl-a) are crucial indicators of water quality because they can reveal the presence of suspended particles and algae, respectively. Understanding the health of rivers and spotting long-term water quality changes can both benefit from monitoring these measures. Traditional methods of monitoring these parameters, like in-situ measurements, is time-consuming, expensive, and inconvenient in some places. Sentinel-2, a multispectral satellite, might offer a more workable and economical option for monitoring water quality, though. This study used 100 in-situ data collected from the Ganga River near Varanasi in the pre-monsoon season (pre-MS) and post-monsoon season (post-MS) in order to create a model for the prediction of optically active water quality parameters by combining Multispectral Instrument (MSI) data and machine learning method (Random Forest). To create spatial distribution maps for Chl-a and Turb, 14 spectral indices and band ratios were employed as independent variables. The results showed that the prediction accuracy for Turb (R2 = 0.91, MAE = 1.13 and MAPE=7.76 % during pre-MS and R2 = 0.93, MAE = 0.88 and MAPE=2.29 % during post-MS) and for Chl-a (R2 = 0.97, MAE = 0.59, and MAPE=2.07 % during pre-MS and R2 = 0.95, MAE = 0.61, and MAPE = 2.71 % during post-MS). The Ganga near Varanasi abruptly turned green due to an increase in algal bloom in May and June 2021. This study not only revealed the reasons behind the green appearance but also identified potential areas of concern or hotspots. In order to identify hotspot locations, drainage networks, point source discharge locations and LU-LC were used.

Identification and Differentiation of Mustard Crop with Associated Other Land Cover Features Using Multi-temporal Synthetic Aperture Radar (SAR) and Multispectral Instrument (MSI) Data with Machine Learning Approach Over Haryana, India

Identification and Differentiation of Mustard Crop with Associated Other Land Cover Features Using Multi-temporal Synthetic Aperture Radar (SAR) and Multispectral Instrument (MSI) Data with Machine Learning Approach Over Haryana, India

Assessment of Atmospheric Correction Algorithms for Correcting Sunglint Effects in Sentinel-2 MSI Imagery: A Case Study in Clean Lakes

The Sentinel-2 Multi-Spectral Instrument (MSI) is characterized by short revisit times (5 days), red-edge spectral bands (665 nm and 705 nm), and a high spatial resolution (10 m), making it highly suitable for monitoring water quality in both inland and coastal waters. Unlike SeaWiFS, which can adjust its viewing angles to minimize sunglint, the Sentinel-2 MSI operates with fixed near-nadir angles, which makes it more susceptible to sunglint. Additionally, the complex optical properties of water pose challenges in accurately determining its water-leaving reflectance. Therefore, we compared the effectiveness of six atmospheric correction (AC) algorithms (POLYMER, MUMM, DSF, C2RCC, BP, and GRS) in correcting sunglint using two typical lakes in Xinjiang, China, as examples. The results indicated that POLYMER achieved the highest overall evaluation score (1.61), followed by MUMM (1.21), while BP exhibited the lowest performance (0.62). Specifically, POLYMER showed robust performance at the 665 nm band with RMSE = 0.0012 sr−1, R2 = 0.74, and MAPE = 30.68%, as well as at the 705 nm band with RMSE = 0.0014 sr−1, R2 = 0.42, and MAPE = 38.44%. At the 443, 490, and 560 nm bands, MUMM showed better performance (RMSE ≤ 0.0026 sr−1, R2 ≥ 0.86, MAPE ≤ 28.20%). In terms of band ratios, POLYMER exhibited the highest accuracy (RMSE ≤ 0.093 and MAPE ≤ 22.2%), particularly for the ratio Rrs(490)/Rrs(560) (R2 = 0.71). In general, POLYMER is the best choice for the sunglint correction of Xinjiang’s clean lakes. This study assessed the capability of different AC algorithms for sunglint correction and enhanced the monitoring capability of MSI data in clean waters.

Russian olive distribution and invasion dynamics along the Powder River, Montana and Wyoming, USA

The invasive shrub, Russian olive (Elaeagnus augustifolia), is widely established within riparian areas across North America and eastern Europe. Limited information on its distribution and invasion dynamics in northern regions has hampered understanding and management efforts. Given this lack of spatial and ecological information we worked with local stakeholders and developed two main objectives: (1) map the distribution of Russian olive along the Powder River (Montana and Wyoming, United States) as of 2020 with field data and remote sensing; and (2) relate that distribution to environmental variables to understand its habitat suitability and community/invasion dynamics. Field data showed Russian olive has reached near equal canopy cover (18.3%) to native Plains cottonwood (Populus deltoides; 19.1%) and has a broader distribution. At the watershed scale, we modeled Russian olive distribution using field surveys, ocular sampling of aerial imagery, and spectral variables from Sentinel-2 MultiSpectral Instrument using a random forest model (RMSE = 15.42, R2 = 0.64). A statistical model linking the resulting Russian olive percent cover detection map to environmental variables for the entire watershed indicated Russian olive cover increased with flow accumulation and decreased with elevation, and was associated with poorer soil types. We attribute the success of Russian olive to its broad habitat suitability combined with changing hydrologic conditions favoring it over natives. The maps of Russian olive cover along the Powder River and its main tributaries in northern Wyoming and southern Montana revealed regions of the watershed with high and low cover, which can guide landscape-scale management prioritization. This study provides a repeatable Russian olive detection method due to the use of Sentinel-2 imagery that is available worldwide and provides insight into Russian olive’s ecological relationships and success with relevance for management across areas with similar environmental conditions.

Suitability of different in-water algorithms for eutrophic and absorbing waters applied to Sentinel-2 MSI and Sentinel-3 OLCI data

Optically complex waters present significant challenges for remote sensing due to high concentrations of optically active substances (OASs) and their inherent optical properties (IOPs), as well as the adjacency effect. OASs and IOPs can be derived from atmospheric correction processors’ in-water algorithms applied to data from Sentinel-2 MultiSpectral Instrument (S2 MSI) and Sentinel-3 Ocean and Land Color Instrument (S3 OLCI). This study compared S3 OLCI Level-2 in-water products for Case-2 waters with alternative in-water algorithms derived from ACOLITE, POLYMER, C2RCC, and A4O. Fifty in-water algorithms were evaluated using an extensive match-up dataset from lakes and coastal areas, focusing particularly on small lakes with high colored dissolved organic matter absorption at 442 nm (up to 48 m-1). The Chl a band ratio introduced by Gons et al. (2022) applied to data processed by ACOLITE performed best for S3 OLCI Chl a retrieval (dispersion = 23%, bias = 10%). Gons et al. (2022) band ratio also showed consistent agreement between S3 OLCI and S2 MSI resampled data (intercept of 6.27 and slope of 0.83, close to the 1:1 line); however, lower Chl a values (<20 mg/m3) were overestimated by S2 MSI. When estimating errors associated with proximity to land, S2 MSI Chl a in-water algorithms had higher errors close to the shore (on average 315%) compared to S3 OLCI (on average 150%). Chl a retrieved with POLYMER had the lowest errors close to the shore for both S2 MSI and S3 OLCI data (on average 70%). Total suspended matter (TSM) retrieval with C2RCC performed well for S2 MSI (dispersion 24% and bias −12%). Total absorption was most accurately derived from C2RCC applied to S3 OLCI L1 data (dispersion < 43% and bias < −39%), and it was better estimated than its individual components: phytoplankton, mineral particles, and colored dissolved organic matter absorption. However, none of the colored dissolved organic matter absorption in-water algorithms performed well (dispersion > 59% and bias < −29%).

Monitoring and assessing the effectiveness of the biological control implemented to address the invasion of water hyacinth (Eichhornia crassipes) in Hartbeespoort Dam, South Africa

Water hyacinth is one of the most aggressive alien invasive plants, which invades freshwater resources and destroys native biodiversity. The plant proliferates rapidly over a short space of time, forming thick dense layer on the surface of freshwater bodies. Monitoring and management of water hyacinth is essential to protect water resources affected by the presence of this plant. The study assessed the effectiveness of biological agent (Megamelus scutellaris) applied in the Hartbeespoort Dam from pre (2016–2017) and post (2018–2023) biological control to manage water hyacinth spread and proliferation. In achieving this main goal, the study used advanced cloud-computing machine learning techniques and multi date Sentinel-2 Multispectral Instrument (MSI) data to monitor the effectiveness of such biological control. During this analysis, remote sensing data was acquired for two time periods namely: pre-intervention (2016–2017) and post intervention (2018–2023) to establish variation in the spatio-temporal distribution of water hyacinth in the Hartbeespoort Dam using various machine learning techniques (Support Vector Machine (SVM), Classification and Regression Tree (CART), Random Forest (RF) and Naïve Bayes (NB)) in Google Earth Engine cloud computing platform, and assessed the spectral separability of water hyacinth from numerous land cover types, within and around the Hartbeespoort Dam using the Sentinel-2 derived spectral reflectance curves. The results indicated that the extent of water hyacinth area coverage decreased from 15% to below 6% between the period of 2018 and 2021, however, a significant increase was noted between November 2022 and April 2023, after the biological control was introduced. The significant increase observed during the time period of November 2022 and April 2023 can be attributed to nutrient rich water discharging into the dam from the Crocodile River during the time of flooding reported in November 2022. The result further indicate that RF produced the highest overall accuracies ranging between 93.42% and 98.70%. While NB produced the lowest accuracies ranging between 87.76% and 92.08%. These findings underscore the relevance of new generation satellite dataset and machine learning algorithms in monitoring the effectiveness of the biological controls of alien invasive spread provide information regarding alien plant invasion. Therefore, aligning with Sustainable Development Goals (SDG 6) emphasizing on the importance of implementing effective control measures to control invasive species and their impact on water resources thus ensuring the sustainability of freshwater ecosystems and the availability of clean water resources.

Methods for detecting green tide in the Yellow Sea using Google Earth Engine platform

The green tide was a blooming phenomenon consisting of outbreaks and aggregations of marine macroalgae, causing severe damage to the ecosystem of the Yellow Sea in China. Remote sensing technology was considered an effective means of green tide detection. Currently, commonly used low spatial resolution satellites cannot satisfy the green tide detection in the fine-scale range. To better improve the accuracy of green tide identification, based on the Commission Internationale del'éclairage (CIE) system, two new green tide detection algorithms (CIE xy, hue angle α) were established using Sentinel-2 MultiSpectral Instrument (MSI) and Landsat-8 Operational Land Imager (OLI) data on the Google Earth Engine (GEE) platform. These algorithms were validated by introducing the quantitative index Dgw from "Interclass Distance". The CIE xy algorithm and the hue angle α algorithm based on Sentinel-2 MSI exhibited higher Dgw values of 10.29 and 5.87, respectively, surpassing the performance of the two algorithms based on the Landsat-8 OLI in green tide identification (Dgw values of 3.36 and 5.17, respectively). Meanwhile, the proposed algorithms were compared with normalized difference vegetation index (NDVI), difference vegetation index (DVI), ratio vegetation index (RVI), floating algae index (FAI), enhanced vegetation index (EVI), and algal bloom detection index (ABDI) algorithms using Sentinel-2 MSI images. The results showed that the CIE xy algorithm and the NDVI algorithm outperformed the other algorithms. Especially in turbid water, the CIE xy algorithm exhibited significant advantages. However, the hue angle α algorithm still showed some potential in identifying green tide, particularly in the presence of thin cloud interference. These findings can provide effective information support for the monitoring of green tide in the Yellow Sea.

Discrimination of Degraded Pastures in the Brazilian Cerrado Using the PlanetScope SuperDove Satellite Constellation

Pasture degradation poses significant economic, social, and environmental impacts in the Brazilian savanna ecosystem. Despite these impacts, effectively detecting varying intensities of agronomic and biological degradation through remote sensing remains challenging. This study explores the potential of the eight-band PlanetScope SuperDove satellite constellation to discriminate between five classes of pasture degradation: non-degraded pasture (NDP); pastures with low- (LID) and moderate-intensity degradation (MID); severe agronomic degradation (SAD); and severe biological degradation (SBD). Using a set of 259 cloud-free images acquired in 2022 across five sites located in central Brazil, the study aims to: (i) identify the most suitable period for discriminating between various degradation classes; (ii) evaluate the Random Forest (RF) classification performance of different SuperDove attributes; and (iii) compare metrics of accuracy derived from two predicted scenarios of pasture degradation: a more challenging one involving five classes (NDP, LID, MID, SAD, and SBD), and another considering only non-degraded and severely degraded pastures (NDP, SAD, and SBD). The study assessed individual and combined sets of SuperDove attributes, including band reflectance, vegetation indices, endmember fractions from spectral mixture analysis (SMA), and image texture variables from Gray-level Co-occurrence Matrix (GLCM). The results highlighted the effectiveness of the transition from the rainy to the dry season and the period towards the beginning of a new seasonal rainy cycle in October for discriminating pasture degradation. In comparison to the dry season, more favorable discrimination scenarios were observed during the rainy season. In the dry season, increased amounts of non-photosynthetic vegetation (NPV) complicate the differentiation between NDP and SBD, which is characterized by high soil exposure. Pastures exhibiting severe biological degradation showed greater sensitivity to water stress, manifesting earlier reflectance changes in the visible and near-infrared bands of SuperDove compared to other classes. Reflectance-based classification yielded higher overall accuracy (OA) than the approaches using endmember fractions, vegetation indices, or texture metrics. Classifications using combined attributes achieved an OA of 0.69 and 0.88 for the five-class and three-class scenarios, respectively. In the five-class scenario, the highest F1-scores were observed for NDP (0.61) and classes of agronomic (0.71) and biological (0.88) degradation, indicating the challenges in separating low and moderate stages of pasture degradation. An initial comparison of RF classification results for the five categories of degraded pastures, utilizing reflectance data from MultiSpectral Instrument (MSI)/Sentinel-2 (400–2500 nm) and SuperDove (400–900 nm), demonstrated an enhanced OA (0.79 versus 0.66) with Sentinel-2 data. This enhancement is likely to be attributed to the inclusion of shortwave infrared (SWIR) spectral bands in the data analysis. Our findings highlight the potential of satellite constellation data, acquired at high spatial resolution, for remote identification of pasture degradation.

A national-scale assessment on the spatial and temporal variations in water color for urban lakes in China

Monitoring the variations of lake water quality is essential for urban water security and sustainable eco-environment health. However, it is challenging to investigate the water quality of urban lakes at large scales due to the need for large-amount in situ data with diverse optical properties for developing the remote sensing inversion algorithms. Forel-Ule Index (FUI), a proxy of quantifying water color, whose calculation does not require in situ data of specific properties, can comprehensively reflect water quality conditions. However, the spatial and temporal distribution of water color in Chinese urban lakes is still poorly understood. To fill this research gap, this study investigated the spatial distribution of water color in 523 urban lakes (area > 0.5 km2) in China using the FUI derived from the high-quality Multi-Spectral Instrument (MSI) data onboard Sentinel-2 during the ice-free period (April–October) from 2019 to 2022. The monthly and seasonal variation patterns of water color in urban lakes were also analyzed. Our results show that green domain is the most common color of urban lakes, with about 86 % of urban lakes in China being green, and non-green lakes accounting for only 14 % of the total number of lakes. The monthly variation of FUI in urban lakes across the country and multiple geographic regions is basically the same. The monthly average FUI first increases, then decreases, and then rebounds. We also found that the seasonal variation of water color in most urban lakes in southern and northern China is opposite. This study helps to comprehensively understand the spatial and temporal variation of water color and quality of urban lakes in China, providing key basic information for the protection and governance of urban lakes.

Mapping the recovery of Mountain Ash (Eucalyptus regnans) and Alpine Ash (E. delegatensis) using satellite remote sensing and a machine learning classifier

This research presents a random forest classification approach to map the response of the obligate-seeder Eucalyptus species, Mountain Ash (Eucalyptus regnans) and Alpine Ash (E. delegatensis), to disturbance from timber harvesting in the Victorian Central Highlands in south-eastern Australia. A Sentinel-2 MultiSpectral Instrument (MSI) composite image was classified and analysed using a random forest algorithm trained using field data collected within fifty-three sites. Training and validation datasets were produced by randomly sub setting using a 70:30 split. Validation was performed by producing a confusion matrix using the points which were excluded from model training. The random forest model demonstrated strong performance at distinguishing Eucalyptus regrowth from the dominant understory species, Silver Wattle (Acacia dealbata), achieving an F1-score of 97.3% and true skill statistic of 96.4%.This study showcases the operational insights that satellite remote sensing data and machine learning can provide for regional-scale monitoring and management of E. regnans and E. delegatensis dominant ecosystems following disturbance. Due to the high conservation value of these communities, and their sensitivity to frequent high intensity disturbance and low precipitation during regeneration, this research seeks to provide a means to assess the condition of regenerating forest and in doing so enhance our understanding of these ecologically significant ecosystems in response to changing environmental conditions.

Vision Transformer for Flood Detection Using Satellite Images from Sentinel-1 and Sentinel-2

Floods are devastating phenomena that occur almost all around the world and are responsible for significant losses, in terms of both human lives and economic damages. When floods occur, one of the challenges that emergency response agencies face is the identification of the flooded area so that access points and safe routes can be determined quickly. This study presents a flood detection methodology that combines transfer learning with vision transformers and satellite images from open datasets. Transformers are powerful models that have been successfully applied in Natural Language Processing (NLP). A variation of this model is the vision transformer (ViT), which can be applied to image classification tasks. The methodology is applied and evaluated for two types of satellite images: Synthetic Aperture Radar (SAR) images from Sentinel-1 and Multispectral Instrument (MSI) images from Sentinel-2. By using a pre-trained vision transformer and transfer learning, the model is fine-tuned on these two datasets to train the models to determine whether the images contain floods. It is found that the proposed methodology achieves an accuracy of 84.84% on the Sentinel-1 dataset and 83.14% on the Sentinel-2 dataset, revealing its insensitivity to the image type and applicability to a wide range of available visual data for flood detection. Moreover, this study shows that the proposed approach outperforms state-of-the-art CNN models by up to 15% on the SAR images and 9% on the MSI images. Overall, it is shown that the combination of transfer learning, vision transformers, and satellite images is a promising tool for flood risk management experts and emergency response agencies.

Expanding the Application of Sentinel-2 Chlorophyll Monitoring across United States Lakes.

Eutrophication of inland lakes poses various societal and ecological threats, making water quality monitoring crucial. Satellites provide a comprehensive and cost-effective supplement to traditional in situ sampling. The Sentinel-2 MultiSpectral Instrument (S2 MSI) offers unique spectral bands positioned to quantify chlorophyll a, a water-quality and trophic-state indicator, along with fine spatial resolution, enabling the monitoring of small waterbodies. In this study, two algorithms-the Maximum Chlorophyll Index (MCI) and the Normalized Difference Chlorophyll Index (NDCI)-were applied to S2 MSI data. They were calibrated and validated using in situ chlorophyll a measurements for 103 lakes across the contiguous U.S. Both algorithms were tested using top-of-atmosphere reflectances (ρ t), Rayleigh-corrected reflectances (ρ s), and remote sensing reflectances (R rs ). MCI slightly outperformed NDCI across all reflectance products. MCI using ρ t showed the best overall performance, with a mean absolute error factor of 2.08 and a mean bias factor of 1.15. Conversion of derived chlorophyll a to trophic state improved the potential for management applications, with 82% accuracy using a binary classification. We report algorithm-to-chlorophyll-a conversions that show potential for application across the U.S., demonstrating that S2 can serve as a monitoring tool for inland lakes across broad spatial scales.

Harmonized NDVI time-series from Landsat and Sentinel-2 reveal phenological patterns of diverse, small-scale cropping systems in East Africa

Capturing the phenological responses of crops in diverse, small-scale agricultural systems is one of the most notable remote sensing research gaps in Eastern Africa. Because of the heterogeneous and complex agricultural environments, remote sensing methodologies successfully implemented elsewhere yield poor results in Eastern Africa. The reliance on local rainfall and dry spells can lead to alterations in planting patterns, impacting growing season timing and yield, underscoring the importance of high temporal resolution for capturing crop phenology. We generated a high-resolution land use/land cover (LULC) classification map of 10 m using Sentinel-1 and Sentinel-2 imagery together with a large set of reference data acquired in situ in 2022. The LULC classification yielded an overall accuracy of 87.9% and Cohen's kappa value of 0.83. Based on the LULC map, we investigated the phenological metrics of the most important crops in the Kamuli district of Eastern Uganda by combining a multi-sensor harmonization approach, thresholding method, and cloud computing. Previously unavailable Landsat 9 band adjustment coefficients were derived. By harmonizing Landsat (LS) 7 Enhanced Thematic Mapper Plus (ETM+), Landsat 8 and 9 Operational Land Imager (OLI), with Sentinel-2 Multispectral Instrument (MSI), the phenological patterns during the March to May growing season were captured using a dense temporal Normalized Difference Vegetation Index (NDVI) time series. The seasonal onset for most crop types was in the end of March to middle of April while the end of the growing season occurred from the end of June to middle of July. Investigation of the 13 crop types using a high-resolution LULC map and dense, harmonized NDVI time series signifies the first field-level phenological metrics estimation of smallholder farming systems in Eastern Africa. Future research should focus on expanding the methodology in space and time to provide insights into past and current growing season development.