Applied Geomatics最新文献

{"title":"Characterization and retrieval of snow grain size in the Bhilangana region of the Upper Himalayas using hyperspectral PRISMA data","authors":"Manish Rawat,&nbsp;Ashish Pandey,&nbsp;Dhananjay Paswan Das,&nbsp;Praveen Kumar Gupta","doi":"10.1007/s12518-025-00627-5","DOIUrl":"10.1007/s12518-025-00627-5","url":null,"abstract":"<div><p>Rapid urbanization processes have significantly increased freshwater consumption, prompting the need for precise predictions of snowmelt-derived streamflow in glacierized Himalayan basins, which are highly susceptible to climate change. However, understanding snow characteristics, such as snow cover and snow grain size, remains a challenge due to inaccessibility of these terrains and the limitations of in-situ data collection. Hyperspectral remote sensing datasets offer a promising solution for monitoring and retrieving snow properties at both micro and macro levels. In this study, the PRISMA hyperspectral dataset was employed to estimate snow grain sizes in the Bhilangana basin of the Upper Himalayan region through the Spectral Angle Mapper (SAM) and Snow Grain Size Index (SGSI) methods. The SGSI approach uses visible and near-infrared wavelengths to classify snow grains, while the SAM method applies endmember spectral signatures validated against the USGS spectral library. The results shows that both SGSI and SAM effectively classified snow grains into fine (&lt; 0.5 mm), medium (0.5–1.0 mm), and coarse (1.0–2.0 mm) categories, achieving a classification accuracy of approximately 88%. The SGSI method utilized the bi-spectral reflectance ratio of PRISMA bands 6 (441.63 nm) and 69 (1028.79 nm) to classify snow grains with spatial variability. The outcomes of the study disclose the competency of PRISMA data for spatial representation of snow grain size variability. The spatial analysis shows that fine and medium grain sizes dominate the snowpack, particularly during the seasonal accumulation observed in February. The findings indicate that fine-grained snow distribution at higher altitudes is crucial for assessing avalanche risks assessment and predicting snowmelt timing. This research demonstrates PRISMA data’s effectiveness in detailed snow grain size mapping, offering valuable insights for applications in climatology, hydrology, and mountain hazard management. Enhanced snow grain size mapping contributes to improved avalanche forecasting and resource planning, ultimately supporting the safety and resilience of the Himalayan mountain regions.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 3","pages":"431 - 447"},"PeriodicalIF":2.3,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144909693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Extracting urban patterns in undulating landscapes from SAR data with thresholding approach","authors":"Noyingbeni Kikon,&nbsp; Deepak Kumar,&nbsp; Syed Ashfaq Ahmed","doi":"10.1007/s12518-025-00626-6","DOIUrl":"10.1007/s12518-025-00626-6","url":null,"abstract":"<div><p>Urban footprint extraction is used for the extraction or classification of various land use classes like water bodies, urban areas, vegetation, and others over any region. But this is quite difficult to perform in the hilly terrains. The work recognises the optimal threshold value for the extraction of urban features is based on the coherence properties of the processed SAR dataset. The work utilises two Sentinel-1 A satellite images acquired on 7th January 2020 and 31st January 2020 respectively. The work of urban footprint is accomplished with (a) the creation of a coherence image with a pair of SAR imageries; (b) further pre-processing of the coherence image to apply multi-looking and terrain correction; (c) the derived coherence image is stacked to create a false colour composite image to provide an input for feature extraction; (d) feature extraction is performed by masking out the urban areas at different thresholds levels. The results of the extracted urban footprint are authenticated with a comparison to the optical dataset. Some sample locations are selected for validating the results from Google Earth historical imagery. Results indicate that the urban features extracted at a threshold value of 0.5 provide improved results in comparison to the threshold values of 0.4, 0.6, and 0.7. The pixels of urban features at a coherence threshold of 0.5 are lying at the same position where urban areas are present. The work can be further propagated for the identification and monitoring of other urban features regardless of any weather conditions for several other applications.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 3","pages":"411 - 429"},"PeriodicalIF":2.3,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00626-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144909794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ensembling satellite monitoring and numerical cartography towards the safety assessment of infrastructures","authors":"Francesca Grassi,&nbsp;Paolo Rossi,&nbsp;Benedetta Brunelli,&nbsp;Francesco Mancini,&nbsp;Cristina Castagnetti,&nbsp;Loris Vincenzi,&nbsp;Elisa Bassoli,&nbsp;Alessandro Capra","doi":"10.1007/s12518-025-00625-7","DOIUrl":"10.1007/s12518-025-00625-7","url":null,"abstract":"<div><p>This paper explores new technologies that can advance the state-of-the-practice in safety assessment and health monitoring of existing infrastructures. In this context, multi-temporal interferometric Synthetic Aperture Radar techniques combined with the use of digital models of infrastructures represent a powerful integration to conventional approaches in the monitoring and assessment of structural safety of infrastructures. Although the interferometric method is widely used for ground deformation investigations, using displacement data from satellite observation in structural monitoring is less investigated. The joint use of multi-frequency satellite radar data provided by the European Space Agency Copernicus project and Italian Space Agency will be explored. The paper introduces the workflow implemented for processing satellite radar data from the X-band COSMO-SkyMed constellation by the Italian Space Agency over the municipality of Modena (Italy). An open-source workflow based on Multi-Temporal Interferometric technique and Persistent Scatterers Interferometry is adopted, enabling the detection of displacements of stable targets and the generation of corresponding time series. Radar data products, derived from the processing of both COSMO-SkyMed and Sentinel-1 data, are analyzed in a Geographic Information System alongside the available geospatial dataset of infrastructures. This approach enables the extraction of displacement components related to the ground and infrastructures. The method’s potential for characterizing infrastructures behaviour is assessed through the analysis of selected case studies. The results aim to establish the foundations for a method capable of assessing infrastructure safety.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 2","pages":"401 - 410"},"PeriodicalIF":2.3,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00625-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145171097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated monitoring of dams and large ponds: the role of satellite radar interferometry and the European ground motion service","authors":"Antonio Miguel Ruiz-Armenteros,&nbsp;Miguel Marchamalo-Sacristán,&nbsp;Francisco Lamas-Fernández,&nbsp;Álvaro Hernández-Cabezudo,&nbsp;Alfredo Fernández-Landa,&nbsp;José Manuel Delgado-Blasco,&nbsp;Matus Bakon,&nbsp;Milan Lazecky,&nbsp;Daniele Perissin,&nbsp;Juraj Papco,&nbsp;Gonzalo Corral,&nbsp;José Luis García-Balboa,&nbsp;José Luis Mesa-Mingorance,&nbsp;Admilson da Penha Pacheco,&nbsp;Juan Manuel Jurado-Rodríguez,&nbsp;Joaquim J. Sousa","doi":"10.1007/s12518-025-00624-8","DOIUrl":"10.1007/s12518-025-00624-8","url":null,"abstract":"<div><p>Satellite radar interferometry (InSAR) has become an invaluable tool for monitoring dams and large ponds, providing significant advantages when complemented with geotechnical and geodetic monitoring. InSAR uses radar signals from satellites to detect ground movements with millimeter precision by comparing phase differences between images taken at different times. This technique enables large-scale, continuous monitoring, which is critical for identifying potential structural problems and preventing catastrophic failures. Unlike traditional geotechnical and geodetic monitoring, which require extensive equipment and manual data collection, InSAR provides a non-intrusive, efficient solution that covers vast areas with high temporal frequency. The European Ground Motion Service (EGMS) exemplifies these advantages by providing standardized ground motion data across Europe, derived from Sentinel-1 satellite radar data. EGMS enables routine and comprehensive monitoring of ground stability and infrastructure integrity, assisting in the early detection of deformation patterns and supporting proactive maintenance and risk management. For dam managers, the integration of InSAR with traditional methods enhances the reliability of structural health assessments. Geotechnical sensors offer localized information on soil and material properties, while geodetic methods provide precise positional data; InSAR complements these by delivering comprehensive, continuous deformation maps. This synergy ensures robust monitoring and enhances the ability to predict and mitigate potential problems, significantly improving the effectiveness and efficiency of monitoring dams and large ponds, and contributing to safer and more resilient infrastructure management. This work presents several case studies from the SIAGUA project as examples, highlighting the practical applications and benefits of combining InSAR with traditional monitoring techniques.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 2","pages":"379 - 392"},"PeriodicalIF":2.3,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00624-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145169562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ground-based InSAR and GNSS integration for enhanced dam monitoring","authors":"Matthieu Rebmeister,&nbsp;Andreas Schenk,&nbsp;Jakob Weisgerber,&nbsp;Malte Westerhaus,&nbsp;Stefan Hinz,&nbsp;Frédéric Andrian,&nbsp;Maxime Vonié","doi":"10.1007/s12518-025-00622-w","DOIUrl":"10.1007/s12518-025-00622-w","url":null,"abstract":"<div><p>The monitoring of dams is essential to ensure their safe operation for the production of renewable energy. Common tools to monitor dams are permanently installed plumblines and surveying by means of total station and leveling within a geodetic network. The main drawback of these methods is their limited spatial and temporal resolution. Recent studies have shown promising results using Ground-Based InSAR for geodetic dam monitoring. The fast acquisition speed combined with the surface monitoring capabilities enable to monitor several hundreds to thousands of points on the dam every day or several times a day. However, GB-SAR is a relative phase-measurement technique, and any interruption in the data acquisition leads to difficulties to unwrap differential phase observations and join the disjunct time series. The combination with other absolute measurement tools is promising to create an absolute deformation map of the dam. GNSS is a very efficient and reliable method providing point-wise absolute displacement time series and mm-accuracy. This paper proposes a combination of GNSS and GB-SAR observations to enhance the consistency of the surface-based dam displacement maps obtained by solely GB-SAR measurements. A method to detect unwrapping errors over long time series is proposed. The corrected GB-SAR time displacement maps are compared to a numerical model and confirm the correctness of the applied corrections.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 2","pages":"393 - 400"},"PeriodicalIF":2.3,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00622-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145170153","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detecting change points in time series of inSAR persistent scatterers using deep learning models","authors":"Kourosh Shahryarinia,&nbsp;Mohammad Omidalizarandi,&nbsp;Mohammadreza Heidarianbaei,&nbsp;Mohammad Ali Sharifi,&nbsp;Ingo Neumann","doi":"10.1007/s12518-025-00621-x","DOIUrl":"10.1007/s12518-025-00621-x","url":null,"abstract":"<div><p>Accurately detecting significant changes in the Earth’s surface is essential for timely intervention. As a key techniques in Interferometric Synthetic Aperture Radar (InSAR), Persistent Scatterer Interferometry (PSI) generates time series data of Persistent Scatterers (PS), which are stable points on the Earth’s surface that enable precise displacement measurements over time. While many studies have focused on statistical methods for identifying anomalies in PS time series, few have explored the potential of deep learning for change point (CP) detection. A major challenge with supervised deep learning is the need for large labeled datasets. To overcome this, we implemented a simulation algorithm to generate an extensive set of PS points with corresponding target CPs, reflecting the statistical characteristics of PS time series. To identify changes in slope and intercept, We used two deep learning models: Bidirectional Long Short-Term Memory (BiLSTM), designed for time series data, and U-Net, developed for image data. A spectral analysis technique is applied to remove seasonal components from the time series data before feeding into the networks. The models were evaluated using metrics such as F1-score, precision, and recall, and were compared to a Bayesian-based approach. Additionally, the methodology was applied to real PS time series from a study area in Germany. We analyzed the detected CPs along with the neighboring PS time series within a 15-meter radius. The results indicated that the deep learning models outperformed the Bayesian approach in terms of precision, recall, and F1-score with simulated PS time series, highlighting their potential for precise CP detection. Furthermore, the models demonstrated their effectiveness when applied to the real PS time series.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 2","pages":"357 - 366"},"PeriodicalIF":2.3,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00621-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145168336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated change detection in photogrammetric 4D point clouds – transferability and extension of 4D objects-by-change for monitoring riverbank dynamics using low-cost cameras","authors":"Maximilian Ulm,&nbsp;Melanie Elias,&nbsp;Anette Eltner,&nbsp;Eliisa Lotsari,&nbsp;Katharina Anders","doi":"10.1007/s12518-025-00623-9","DOIUrl":"10.1007/s12518-025-00623-9","url":null,"abstract":"<div><p>This paper is dedicated to an automated detection of geomorphological changes in photogrammetric 4D point clouds, which are acquired using low-cost wildlife cameras at a subarctic riverbank. In these regions, a better understanding of complex erosion processes is required for modelling sediment dynamics and to understand climate change effects. Therefore, a spatiotemporally detailed dataset was collected with two-hourly images from four cameras over six months (approx. 900 epochs). Changes are extracted as 4D objects-by-change (4D-OBCs), a method of spatiotemporal segmentation that considers time series information which was originally developed for permanent terrestrial laser scanning data. This contribution investigates the transferability of the 4D-OBC method to noisy photogrammetric point clouds in terms of detection reliability and quantification accuracy. Focus is on the detection methods for linear changes in time series. An extension of the method is developed for fusing 4D-OBCs in a second step, as the fully automatic extraction often leads to oversegmentation. This object fusion is based on spatial and temporal overlap of individual objects. For quantitative evaluation, reference objects are extracted manually. Further validation is performed visually using the original time-lapse photos. The analysis results in a total of 946 4D-OBCs extracted as erosion or accumulation events. The object fusion results in a significantly higher agreement with the reference objects (volume ratio between 4D-OBCs and references of 0.26 before and 0.85 after fusion). By this, our research increases the applicability of an automatic time series-based change analysis method to low-cost photogrammetric data and to new change types of riverbank erosion. The use case further contributes to the interpretation of riverbank processes in subarctic regions enabled by time-lapse photogrammetry.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 2","pages":"367 - 378"},"PeriodicalIF":2.3,"publicationDate":"2025-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00623-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145168335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Applying Multi-Criteria Analysis in GIS to predict suitability for recreational green space interventions in Kigali City, Rwanda","authors":"Laban Kayitete,&nbsp;Charles Bakolo,&nbsp;James Tomlinson,&nbsp;Jade Fawcett,&nbsp;Marie Fidele Tuyisenge,&nbsp;Jean de Dieu Tuyizere","doi":"10.1007/s12518-025-00609-7","DOIUrl":"10.1007/s12518-025-00609-7","url":null,"abstract":"<div><p>Green spaces improve societal well-being, foster connectivity to nature, and attenuate climate change. Despite Rwanda and other developing countries increasingly pursuing green economies, urban greening efforts still need multi-conceptual models that comprehensively address socio-economic and environmental requirements. This study employs a GIS-based Multi-Criteria Analysis (MCA) constructed on an Analytical Hierarchy Process (AHP) to predict green space intervention suitability across Kigali City, Rwanda. The study was based on nine factors namely: population density, slope, land cover types, proximity to roads, Normalised Difference Vegetation Index (NDVI), proximity to existing green spaces, proximity to water bodies, nitrogen dioxide concentrations, and elevation, to be used as criteria for the MCA. The findings indicate that 2.49% (1,816.19 ha) of Kigali City is highly suitable while 12% (8,744.68 ha) is unsuitable for green space interventions. Population density emerged as the most influential factor, with the city’s densely populated west-central areas exhibiting high suitability for green space initiatives. Strategically placing green spaces near population centres enhances their contribution to societal well-being, reduces transport costs, and encourages frequent use. By integrating GIS-based MCA with AHP, this study offers a robust framework for addressing green space accessibility challenges in Kigali, while simultaneously advancing climate-resilient urban development. We recommend planners prioritise Kigali City’s west-central areas for green space interventions, researchers leverage the GIS-MCA-AHP methodology for climate-resilient urban studies, and practitioners replicate this framework to advance socio-economically inclusive greening strategies.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 1","pages":"163 - 175"},"PeriodicalIF":2.3,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s12518-025-00609-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143594683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing Meghna Riverbank dynamics and morphological changes in Bangladesh using geospatial techniques","authors":"Masuda Sultana,&nbsp;Muhammad Al-Amin Hoque,&nbsp;Biswajeet Pradhan","doi":"10.1007/s12518-025-00620-y","DOIUrl":"10.1007/s12518-025-00620-y","url":null,"abstract":"<div><p>Riverbank erosion is one of the most frequent natural hazards worldwide. Bangladesh is highly affected by this natural hazard every year. The lower segment of the Meghna River is highly vulnerable to this phenomenon. While previous studies have primarily focused on socio-economic impacts in study area or erosion-accretion detection in other major rivers, this study aimed to investigate the spatiotemporal dynamics of riverbank erosion, bank line shifting, and morphological changes in the Meghna River at Haimchar Upazila, Chandpur. Additionally, the study explored the factors driving erosion and potential mitigation strategies. A combination of primary and secondary data was used, including field surveys and satellite image analysis. Normalized Difference Water Index (NDWI) and unsupervised classification techniques were employed to analyze Landsat images from 1980, 1988, 2000, 2010, and 2021. Morphometric parameters such as river width, sinuosity index, and braided index were quantified to assess morphological changes using cross-sections and equations. Results indicate that the highest erosion (4219 ha) occurred between 1988 and 2000, while the lowest (2218 ha) was recorded from 2010 to 2021. Accretion peaked (4215 ha) between 2000 and 2010 and declined thereafter. Over the 42-year study period, the average annual rates of erosion and accretion were 85 ha/yr and 87.8 ha/yr, respectively. Variations in morphological parameters reflect dynamic channel changes, including the formation of bars and islands. Field surveys identified key erosion drivers and highlighted mitigation strategies relevant to the region. The findings underscore the need for integrated river management and adaptive planning to mitigate the adverse effects of riverbank erosion on local livelihoods. Incorporating social factors into future erosion management frameworks could enhance the effectiveness of mitigation measures. This study provides a foundation for developing targeted interventions and sustainable river management practices.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 1","pages":"147 - 161"},"PeriodicalIF":2.3,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143594682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multimodal and meta-learning approach for improved estimation of 3D vegetation structure from satellite imagery","authors":"Ram C. Sharma","doi":"10.1007/s12518-025-00619-5","DOIUrl":"10.1007/s12518-025-00619-5","url":null,"abstract":"<div><p>This research presents a multimodal and meta-learning approach that integrates multi-source satellite sensor and field plot-level data for enhanced retrieval of 3D vegetation structure. Specifically, the combined effect of integrating multispectral data from Landsat 8 OLI and Sentinel-2 MSI with radar data from Sentinel-1 CSAR was examined. For the utilization of multi-source inputs, the synergistic integration was implemented using efficient machine learning regressors—Random Forest Regressor (RFR) and Extreme Gradient Boosting Regressor (GBR)—ensembled within a meta-learning framework. Three meta-model layers—Multiple Linear Regressor (MLR), K-Nearest Neighbors Regressor (KNR), and RFR—were employed and evaluated. As a subroutine of this integration, a model-specific and data-type-specific feature selection method was employed, which involved training each model on a unique subset of features identified through permutation importance. The idea of the multimodal and meta-learning approach was implemented using extensive plot-wise data from diverse forest types in the New England region utilizing a rich dataset comprising spectral, spectral indices, and backscattering characteristics to capture the variability of forest biomass. The efficacy of multiple ensembling strategies was evaluated, specifically ensembling across data types or regressors, as well as meta-learning across both data types and regressors. Ensembling across data types, which leverages the strengths of both spectral and backscattering information, demonstrated a higher predictive ability, achieving an R2 of 0.68 and an RMSE of 54.21 Mg/ha. This was higher than the ensembling strategy across regressors using the same data type, which yielded an R2 of 0.59 and an RMSE of 61.4 Mg/ha. Nevertheless, the multimodal and meta-learning approach, which collectively leverages both data types and machine learning regressors, achieved superior performance, with an R2 of 0.82 and an RMSE of 40.5 Mg/ha. This was significantly greater than a conventional ensemble method, which lacked the meta-layer integration. Additionally, the meta-model layer using RFR yielded better results compared to the KNR or MLR layers, demonstrating the capability of RFR in handling complex interactions across variables. These results highlight the superior accuracy and reliability of the multimodal and meta-learning approach, indicating its substantial potential to enhance precision in ecological monitoring and carbon management.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":"17 2","pages":"255 - 268"},"PeriodicalIF":2.3,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145163239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}