Search Results (155)

High-precision monitoring of arid wetlands is vital for ecological conservation, yet traditional methods incur prohibitive labeling costs due to complex features. In this study, the wetland of Bosten Lake in Xinjiang is selected as a case area, where Pleiades and PlanetScope-3 multimodal remote sensing data are fused using the Gram–Schmidt method to generate imagery with high spatial and spectral resolution. Based on this dataset, we systematically compare the performance of fully supervised models (FCN, U-Net, DeepLabV3+, and SegFormer) with a weakly supervised learning model, One Model Is Enough (OME), for classifying 19 wetland–cropland mosaic types. Results demonstrate that: (1) SegFormer achieved the best overall performance (98.75% accuracy, 95.33% mIoU), leveraging its attention mechanism to enhance semantic understanding of complex scenes. (2) The weakly supervised OME, using only image-level labels, matched fully supervised performance (98.76% accuracy, 92.82% F1-score) while drastically reducing labeling effort. (3) Multimodal fusion boosted all models’ accuracy, most notably increasing U-Net’s mIoU by 63.39%. (4) Models exhibited complementary strengths: U-Net excelled in wetland vegetation segmentation, DeepLabV3+ in crop classification, and OME in preserving spatial details. This study validates a pathway integrating multimodal fusion with WSL to balance high accuracy and low labeling costs for arid wetland mapping. Full article

Rice lodging is a major agricultural disaster that reduces yield and quality. Accurate lodging detection and causal analysis are essential for disaster mitigation and precision management. To overcome the limited coverage and low automation of conventional approaches, we propose MSR-LodfNet, an enhanced semantic-segmentation model driven by multi-scale remote-sensing imagery, enabling high-precision lodging mapping from regional to field scales. The study selected 13 state-owned farms in Jiansanjiang, Heilongjiang Province, and jointly used PlanetScope satellite images (3 m) and UAV images (0.2 m) to build an integrated workflow of “satellite macro-monitoring, UAV fine verification, and agronomic factor coupling analysis.” The model synergistically optimizes WFNet, DenseASPP multi-scale context enhancement, and Condensed Attention, markedly improving feature extraction and boundary recognition under multi-source imagery. Experimental results show that the model achieves mIoU 84.34% and mPA 93.31% on UAV images and mIoU 81.96% and mPA 90.63% on PlanetScope images, demonstrating excellent cross-scale adaptability and stability. Causal analysis shows that the high-EVI range is significantly positively correlated with lodging probability; its risk is about 6 times that of the low-EVI range, and the lodging probability of direct-seeded rice is about 2.56 times that of transplanted rice, indicating that it may be associated with a higher lodging risk. The results demonstrate that multi-scale remote sensing combined with agronomic parameters can effectively support the mechanism analysis of lodging disasters, providing a quantitative basis and technical reference for precision rice management and lodging-resistant breeding. Full article

An optical remote sensing approach was developed to identify areas with high and low salinity within the mangrove swamp rice system in West Africa. Conducted between 2019 and 2024 in Guinea-Bissau, this study examined two contrasting rice-growing environments, tidal mangrove (TM) and associated mangrove (AM), to assess changes in vegetation dynamics, soil salinity concentration, and soil chemical properties. Field sampling was conducted during the dry season to avoid waterlogging, and soil analyses included texture, cation exchange capacity, micronutrients, and electrical conductivity (EC_e). Meteorological stations recorded rainfall and environmental conditions over the period. Moreover, orthorectified and atmospherically corrected surface reflectance satellite imagery from PlanetScope and Sentinel-2 was selected due to their high spatial resolution and revisit frequency. From this data, vegetation dynamics were monitored using the Normalized Difference Vegetation Index (NDVI), with change detection calculated as the difference in NDVI between sequential images (ΔNDVI). Thresholds of 0.15 ≤ NDVI ≤ 0.5 and ΔNDVI > 0.1 were tested to identify significant vegetation growth, with smaller polygons (<1000 m²) removed to reduce noise. In this process, at least three temporal images per season were analyzed, and multi-year intersections were done to enhance accuracy. Our parameter optimization tests found that a locally calibrated NDVI threshold of 0.26 improved site classification. Thus, this integrated field–remote sensing approach proved to be a reproducible and cost-effective tool for detecting AM and TM environments and assessing vegetation responses to seasonal changes, contributing to improved land and water management in the salinity-affected mangrove swamp rice system. Full article

Coral reef ecosystems are biodiversity hotspots that provide essential ecological and environmental services but are increasingly threatened by anthropogenic pressure and climate change. Effective conservation of reef systems within Marine Protected Areas (MPAs) can be enhanced using spatially explicit approaches that integrate habitat mapping and ecological metrics at seascape scales. In this study, we characterized the benthic seascape of Cayo Arenas and identified optimal priority conservation zones in one of the core zones of the recently established Southern Gulf of Mexico Reefs National Park (SGMRNP). In July 2023, ground-truthing was performed to quantify the cover of sand, calcareous matrix, macroalgae, hard corals and octocorals. Cluster analysis of quantitative data and ecological similarity between classes was used to identify the main benthic habitat classes. Object-based and supervised classification algorithms on a PlanetScope image were used to construct a thematic map of the benthic reef system. Based on the thematic map, habitat connectivity, β-diversity, patch compactness, and availability for commercial species were estimated. In addition, a benthic change analysis (2017–2013), based on the spectral characteristics of PlanetScope images, was performed. The layers obtained were then used to perform an iterative weighted overlay analysis (WOA) using 126 combinations. Six main habitat classes, with different coverages of hard corals, calcareous matrix, macroalgae, and sand, were identified. Habitats with calcareous matrix and sandy substrates dominated the seascape. High habitat compactness, connectivity, and β-diversity values were observed, suggesting habitat stability and ecologically dynamic areas. Based on the WOA, eight optimal priority areas for conservation were recognized. These areas are characterized by heterogeneous habitats, moderate coral cover, and high connectivity. We provide a spatially explicit approach that can strengthen conservation planning within the SGMRNP and other MPAs, particularly by assisting zonation and sub-zonation processes. Full article

The desert riparian forest oasis, dominated by Populus euphratica and Tamarix chinensis, is an important barrier to protect the economic production and habitat of the Tarim River Basin. However, there is still a lack of high-precision spatial distribution data of desert ri-parian forest species below 10 m. The recently launched PlanetScope CubeSat constella-tion, which provides daily earth observation imagery with a resolution of 3 m, offers a highly favorable dataset for mapping the high-resolution distribution of P. euphratica and T. chinensis and an unprecedented opportunity to explore the optimal phenology window to distinguish between them. In this study, time-series PlanetScope images were first used to extract phenological metrics of P. euphratica, dividing the annual life cycle into four phenology windows: duration of leaf expansion (DLE), duration of leaf maturity (DLM), duration of leaf fall (DLF), and duration of the dormancy period (DDP). The random forest model was used to obtain the classification accuracy of 16 phenological window combinations. Results indicate that after gap filling of vegetation index time series, the identification accuracy for P. euphratica and T. chinensis exceeded 0.90. Among individual phenology windows, the DLE window exhibited the highest classification accuracy (average F1-score 0.87). Among the two phenology window combinations, the DLE-DLF and DLE-DLM windows have the highest classification accuracy (average F1-score 0.90). Among the three phenology window combinations, DLE-DLM-DLF displayed the highest classification accuracy (average F1-score 0.91). Nevertheless, the inclusion of features within the DDP window led to a decrease in accuracy by 1–2% points, which was unfavorable for discriminating tree species. Additionally, features observed during the phenology asynchrony period were found to be more valuable for distinguishing between tree species. Our findings highlight the potential of PlanetScope constellation imagery in tree species classification, offering guidance for selecting optimal image acquisition timing and identifying the most valuable images within time series data for future large-scale tree mapping. Full article

Drought-induced forest mortality threatens biodiversity globally, particularly in arid, and semi-arid woodlands. The continual development of remote sensing approaches enables enhanced monitoring of forest health. Herein, we investigate the ability of a limited ground-truthed canopy dieback dataset and satellite image derived Normalised Difference Vegetation Index (NDVI) to make inferences about forest health as temporal and spatial extent from its collection increases. We used ground-truthed observations of relative canopy mortality from the Pinus edulis-Juniperus osteosperma woodlands of southeastern Utah, United States of America, collected after the 2017–2018 drought, and PlanetScope satellite imagery. Through assessing different modelling approaches, we found that NDVI is significantly associated with sitewide mean canopy dieback, with beta regression being the most optimal modelling framework due to the bounded nature of the variable relative canopy dieback. Model performance was further improved by incorporating the proportion of J. osteosperma as an interaction term, matching the reports of species-specific differential dieback. A time-series analysis revealed that NDVI retained its predictive power for our whole testing period; four years after the initial ground-truthing, thus enabling retrospective inference of defoliation and regreening. A spatial random forest model trained on our ground-truthed observations accurately predicted dieback across the broader landscape. These findings demonstrate that modest field campaigns combined with high-resolution satellite data can generate reliable, scalable insights into forest health, offering a cost-effective method for monitoring drought-impacted ecosystems under climate change. Full article

Accurate delineation of fields is difficult in fragmented landscapes where single-date images provide no seasonal cues and supervised models require labels. We propose a method that explicitly represents phenology to improve zero-shot delineation. Using 22 cloud-free PlanetScope scenes over a 5 × 5 km area, a single harmonic model is fitted to the NDVI per pixel to obtain the phase, amplitude and mean. These values are then mapped into cylindrical colour spaces (Hue–Saturation–Value, Hue–Whiteness–Blackness, Luminance-Chroma-Hue). The resulting recoloured composites are segmented using the Segment Anything Model (SAM), without fine-tuning. The results are evaluated object-wise, object-wise grouped by area size, and pixel-wise. Pixel-wise evaluation achieved up to F1 = 0.898, and a mean Intersection-over-Union (mIoU) of 0.815, while object-wise performance reached F1 = 0.610. HSV achieved the strongest area match, while HWB produced the fewest fragments. The ordinal time-of-day basis provided better parcel separability than the annual radian adjustment. The main errors were over-segmentation and fragmentation. As the parcel size increased, the IoU increased, but the precision decreased. It is concluded that recolouring using harmonic NDVI time series is a simple, scalable, and interpretable basis for field delineation that can be easily improved. Full article

The estimation of accurate and precise Satellite-Derived Bathymetries (SDBs) is important in marine and coastal applications for a better understanding of the ecosystems and science-based decision-making. Despite the advancements in related Machine Learning (ML) studies, quantifying the anticipated bias per pixel in the SDBs remains a significant challenge. This study aims to address this knowledge gap by developing a spatially explicit uncertainty index of a ML-derived SDB, capable of providing a quantifiable anticipation for biases of 0.5, 1, and 2 m. In addition, we explore the usage of this index for model optimization via the exclusion of training points of high or moderate uncertainty via a six-fold iteration loop. The developed methodology is applied across the national coastal extent of Belize in Central America (~7017 km²) and utilizes remote sensing data from the European Space Agency’s twin satellite system Sentinel-2 and Planet’s NICFI PlanetScope. In total, 876 Sentinel-2 images, nine NICFI six-month basemaps and 28 monthly PlanetScope mosaics are processed in this study. The training dataset is based on NASA’s system Ice, Cloud and Elevation Satellite (ICESat-2), while the validation data are in situ measurements collected with scientific equipment (e.g., multibeam sonar) and were provided by the National Oceanography Centre, UK. According to our results, the presented approach is able to provide a pixel-based (i.e., spatially explicit) uncertainty index for a specific prediction bias and integrate it to refine the SDB. It should be noted that the efficiency of the optimization of the SDBs as well as the correlations of the proposed uncertainty index with the absolute prediction error and the true depth are low. Nevertheless, spatially explicit uncertainty information produced by a ML-related SDB provides substantial insight to advance coastal ecosystem monitoring thanks to its capability to showcase the difficulty of the model to provide a prediction. Such spatially explicit uncertainty products can also aid the communication of coastal aquatic products with decision makers and provide potential improvements in SDB modeling. Full article

Various fusion methods of optical satellite images have been proposed for monitoring heterogeneous farmlands requiring high spatial and temporal resolution. In this study, a three-meter normalized difference vegetation index (NDVI) was generated by applying the spatiotemporal fusion (STF) method to simultaneously generate a full-length normalized difference vegetation index time series (SSFIT) and enhanced spatial and temporal adaptive reflectance fusion method (ESTARFM) to the NDVI of Sentinel-2 (S2) and PlanetScope (PS), using images from 2019 to 2021 of rice paddy and heterogeneous cabbage fields in Korea. Before fusion, S2 was processed with the maximum NDVI composite (MNC) and the spatiotemporal gap-filling technique to minimize cloud effects. The fused NDVI image had a spatial resolution similar to PS, enabling more accurate monitoring of small and heterogeneous fields. In particular, the SSFIT technique showed higher accuracy than ESTARFM, with a root mean square error of less than 0.16 and correlation of more than 0.8 compared to the PS NDVI. Additionally, SSFIT takes four seconds to process data in the field area, while ESTARFM requires a relatively long processing time of five minutes. In some images where ESTARFM was applied, outliers originating from S2 were still present, and heterogeneous NDVI distributions were also observed. This spatiotemporal fusion (STF) technique can be used to produce high-resolution NDVI images for any date during the rainy season required for time-series analysis. Full article

Small water bodies are widely spread and play crucial roles in supporting regional agricultural and aquaculture activities. PlanetScope imagery has a high resolution (3 m) with daily global coverage and has obviously enhanced small water body mapping. Recent studies have demonstrated the effectiveness of deep learning for mapping small water bodies using PlanetScope; however, a persistent challenge remains in the scarcity of high-quality, manually annotated water masks used for model training, which limits the generalization capability of data-driven deep learning models. In this study, we propose a transfer learning framework that leverages Sentinel-2 data to improve PlanetScope-based small water body mapping, capitalizing on the spectral interoperability between PlanetScope and Sentinel-2 bands and the abundance of open-source Sentinel-2 water masks. Eight state-of-the-art segmentation models have been explored. Additionally, this paper presents the first assessment of the VMamba model for small water body mapping, building on its demonstrated success in segmentation tasks. The models were pre-trained using Sentinel-2-derived water masks and subsequently fine-tuned with a limited set (1292 image patches, 256 × 256 pixels in each patch) of manually annotated PlanetScope labels. Experiments were conducted using 5648 image patches and two areas of 9636 km² and 2745 km², respectively. Among the evaluated methods, VMamba achieved higher accuracy compared with both CNN- and Transformer-based models. This study highlights the efficacy of combining global Sentinel-2 datasets for pre-training with localized fine-tuning, which not only enhances mapping accuracy but also reduces reliance on labor-intensive manual annotation in regional small water body mapping. Full article

The city of Acapulco was impacted by two near-consecutive hurricanes. On 25 October 2023, Hurricane Otis made landfall, reaching the highest Category 5 storm on the Saffir–Simpson scale, causing extensive coastal destruction due to extreme winds and waves. Nearly one year later (23 September 2024), Hurricane John—a Category 2 storm—caused severe flooding despite its lower intensity, primarily due to its unusual trajectory and prolonged rainfall. Digital shoreline analysis of PlanetScope images (captured one month before and after Hurricane Otis) revealed that the southern coast of Acapulco, specifically Zona Diamante—where the major seafront hotels are located—experienced substantial shoreline erosion (94 ha) and damage. In the northwestern section of the study area, the Coyuca Bar experienced the most dramatic geomorphological change in surface area. This was primarily due to the complete disappearance of the bar on October 26, which resulted in a shoreline retreat of 85 m immediately after the passage of Hurricane Otis. Sentinel-1 Synthetic Aperture Radar (SAR) showed that Hurricane John inundated 2385 ha, four times greater than Hurricane Otis’s flooding (567 ha). The retrofitted QGIS methodology demonstrated high reliability when compared to limited in situ local reports. Given the increased frequency of intense hurricanes, these methods and findings will be relevant in other coastal areas for monitoring and managing local communities affected by severe climate events. Full article

(1) Background: Wetlands are ecologically significant ecosystems that support biodiversity and contribute to essential environmental functions such as water purification, carbon storage and flood regulation. However, these ecosystems face increasing pressures from land-use change and degradation, prompting the need for scalable and accurate classification methods to support conservation and policy efforts. In this research, our motivation was to test whether high-spatial-resolution PlanetScope imagery can be used with pixel-based machine learning to support the mapping and monitoring of wetlands at a national scale. (2) Methods: This study compared four machine learning classification models—Random Forest (RF), XGBoost (XGB), Histogram-Based Gradient Boosting (HGB) and a Multi-Layer Perceptron Classifier (MLPC)—to detect and map wetland areas across New Zealand. All models were trained using eight-band SuperDove satellite imagery from PlanetScope, with a spatial resolution of ~3 m, and ancillary geospatial datasets representing topography and soil drainage characteristics, each of which is available globally. (3) Results: All four machine learning models performed well in detecting wetlands from SuperDove imagery and environmental covariates, with varying strengths. The highest accuracy was achieved using all eight image bands alongside features created from supporting geospatial data. For binary wetland classification, the highest F1 scores were recorded by XGB (0.73) and RF/HGB (both 0.72) when including all covariates. MLPC also showed competitive performance (wetland F1 score of 0.71), despite its relatively lower spatial consistency. However, each model over-predicts total wetland area at a national level, an issue which was able to be reduced by increasing the classification probability threshold and spatial filtering. (4) Conclusions: The comparative analysis highlights the strengths and trade-offs of RF, XGB, HGB and MLPC models for wetland classification. While all four methods are viable, RF offers some key advantages, including ease of deployment and transferability, positioning it as a promising candidate for scalable, high-resolution wetland monitoring across diverse ecological settings. Further work is required for verification of small-scale wetlands (<~0.5 ha) and the addition of fine-spatial-scale covariates. Full article

The 1992 reform of the European Common Agricultural Policy (CAP) introduced the Land Parcel Identification System (LPIS), a geodatabase of land parcels used to monitor and regulate agricultural subsidies. Traditionally, the LPIS has relied on high-resolution aerial orthophotos; however, recent advancements in very-high-resolution (VHR) satellite imagery present new opportunities to enhance its effectiveness. This study explores the feasibility of utilizing PlanetScope, a commercial VHR optical satellite constellation, to map agricultural parcels within the LPIS. A test was conducted in Umbria, Italy, integrating existing datasets with a series of PlanetScope images from 2023. A segmentation workflow was designed, employing the Normalized difference Vegetation Index (NDVI) alongside the Edge segmentation method with varying sensitivity thresholds. An accuracy evaluation based on geometric metrics, comparing detected parcels with cadastral references, revealed that a 30% scale threshold yielded the most reliable results, achieving an accuracy rate of 83.3%. The results indicate that the short revisit time of PlanetScope compensates for its lower spatial resolution compared to traditional orthophotos, allowing accurate delineation of parcels. However, challenges remain in automating parcel matching and integrating alternative methods for accuracy assessment. Further research should focus on refining segmentation parameters and optimizing PlanetScope’s temporal and spectral resolution to strengthen LPIS performance, ultimately fostering more sustainable and data-driven agricultural management. Full article

Recent advances in very high resolution PlanetScope imagery and deep-learning techniques have enabled effective mapping of small water bodies (SWBs), including ponds and ditches. SWBs typically occupy a minor proportion of remote-sensing imagery. This creates significant class imbalance that introduces bias in trained models. Most existing deep-learning approaches fail to adequately address this imbalance. Such an imbalance introduces bias in trained models. Most existing deep-learning approaches fail to adequately address the inter-class (water vs. non-water) and intra-class (SWBs vs. large water bodies) simultaneously. Consequently, they show poor detection of SWBs. To address these challenges, we propose an area-based weighted binary cross-entropy (AWBCE) loss function. AWBCE dynamically weights water bodies according to their size during model training. We evaluated our approach through large-scale SWB mapping in the middle and east of Hubei Province, China. The models were trained on 14,509 manually annotated PlanetScope image patches (512 × 512 pixels each). We implemented the AWBCE loss function in State-of-the-Art segmentation models (UNet, DeepLabV3+, HRNet, LANet, UNetFormer, and LETNet) and evaluated them using overall accuracy, F1-score, intersection over union, and Matthews correlation coefficient as accuracy metrics. The AWBCE loss function consistently improved performance, achieving better boundary accuracy and higher scores across all metrics. Quantitative and visual comparisons demonstrated AWBCE’s superiority over other imbalance-focused loss functions (weighted BCE, Dice, and Focal losses). These findings emphasize the importance of specialized approaches for comprehensive SWB mapping using high-resolution PlanetScope imagery in low-latitude regions. Full article

The precise identification and classification of tree species in young forests during their early development stages are vital for forest management and silvicultural efforts that support their growth and renewal. However, achieving accurate geolocation and species classification through field-based surveys is often a labor-intensive and complicated task. Remote sensing technologies combined with machine learning techniques present an encouraging solution, offering a more efficient alternative to conventional field-based methods. This study aimed to detect and classify young forest tree species using remote sensing imagery and machine learning techniques. The study mainly involved two different objectives: first, tree species detection using the latest version of You Only Look Once (YOLOv12), and second, semantic segmentation (classification) using random forest, Categorical Boosting (CatBoost), and a Convolutional Neural Network (CNN). To the best of our knowledge, this marks the first exploration utilizing YOLOv12 for tree species identification, along with the study that integrates digital aerial photogrammetry with Planet imagery to achieve semantic segmentation in young forests. The study used two remote sensing datasets: RGB imagery from unmanned aerial vehicle (UAV) ortho photography and RGB-NIR from PlanetScope. For YOLOv12-based tree species detection, only RGB from ortho photography was used, while semantic segmentation was performed with three sets of data: (1) Ortho RGB (3 bands), (2) Ortho RGB + canopy height model (CHM) + Planet RGB-NIR (8 bands), and (3) ortho RGB + CHM + Planet RGB-NIR + 12 vegetation indices (20 bands). With three models applied to these datasets, nine machine learning models were trained and tested using 57 images (1024 × 1024 pixels) and their corresponding mask tiles. The YOLOv12 model achieved 79% overall accuracy, with Scots pine performing best (precision: 97%, recall: 92%, mAP50: 97%, mAP75: 80%) and Norway spruce showing slightly lower accuracy (precision: 94%, recall: 82%, mAP50: 90%, mAP75: 71%). For semantic segmentation, the CatBoost model with 20 bands outperformed other models, achieving 85% accuracy, 80% Kappa, and 81% MCC, with CHM, EVI, NIRPlanet, GreenPlanet, NDGI, GNDVI, and NDVI being the most influential variables. These results indicate that a simple boosting model like CatBoost can outperform more complex CNNs for semantic segmentation in young forests. Full article