Optimizing satellite-based cropland area estimation through integrated map accuracy assessment and stratified sampling design across six African countries

Discrete Global Grid Systems (DGGS), as a next-generation framework for the digital Earth, inevitably suffer from geometric non-uniformity, which impacts the accuracy of data representation and analysis. Existing quality assessments, predominantly based on Goodchild's criteria, are inadequate for diamond-based grids, particularly in evaluating angular and distance uniformity. This paper addresses this gap by proposing a comprehensive evaluation framework for spherical diamond grids. We extend the Goodchild criteria by incorporating metrics for angular and distance uniformity, creating an integrated five-dimensional system (shape, topology, size, distance, and angle). Using this framework, we systematically compare three typical diamond DGGS derived from the cube, octahedron, and icosahedron. Our results demonstrate that the icosahedron-based grid exhibits optimal uniformity across all five dimensions. Critically, we reveal that the octahedron-based grid, despite having more initial faces, suffers from severe angular distortion in the across-face boundary regions, rendering its uniformity inferior to that of the cube-based grid. We further validate our framework by constructing a Spherical Residual Network for Diamond Grids (SResNet-DG) for a classification task. Our experimental results demonstrate a strong positive correlation between grid uniformity and the SResNet-DG's performance, substantiating the effectiveness and practical relevance of our proposed geometric evaluation system.

From satellite data to soil security: Closing the science–policy gap in soil erosion monitoring in West Africa

This study evaluates the suitability of Fractional Cover products for monitoring land productivity trajectory in support of Land Degradation Neutrality (LDN) reporting in Africa. Fractional Cover quantifies the proportional presence of Photosynthetic Vegetation (PV), Non-Photosynthetic Vegetation (NPV), and Bare Soil (BS), providing an alternative to commonly used vegetation indices such as the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Modified Soil Adjusted Vegetation Index (MSAVI). A case study was conducted in the Mion District of northern Ghana using statistical, spatial, and network-based analyses to compare Fractional Cover outputs with conventional indices. Pearson correlation analysis showed strong agreement between PV and NDVI (r = 0.99), EVI (r = 0.94), and MSAVI (r = 0.97). Spatial correlation mapping and network analysis further demonstrated distinct clustering of vegetation- and soil-related metrics, with PV effectively capturing gradients in vegetation productivity and BS identifying areas of exposed soil. NPV exhibited intermediate behavior associated with transitional land-cover conditions. Overall, the results indicate that the DE Africa Fractional Cover product provides a robust and complementary approach for assessing land productivity trajectory and surface condition, supporting operational monitoring and national LDN reporting.

Accurate delineation of agricultural parcels is essential for precision agriculture, global food security and sustainable land management within the Digital Earth framework. However, existing methods often exhibit limited global contextual awareness, inadequate boundary refinement, and weak adaptability to diverse landscapes. To address these challenges, we propose SLFNet, a boundary-sensitive multitask deep network for agricultural parcel delineation using high-resolution remote sensing imagery. SLFNet integrates three key innovations: (1) a Simplified Transformer-based Module (STM) for capturing long-range dependencies; (2) a Level-Aware Dilated Convolution and Shuffle (LDCS) module for adaptive multi-scale feature extraction; and (3) a Figure-of-Merit contour Loss (FOM Loss) to enhance boundary localization. SLFNet was evaluated on three geographically distinct regions (Denmark, Xinjiang, Sanyuan) with diverse heterogeneity using multiple high-resolution datasets. Experimental results indicate that SLFNet demonstrates competitive performance compared with several state-of-the-art baselines (e.g., HBGNet and REAUNet). Transferability experiments under limited target-domain supervision show that SLFNet can effectively adapt to fragmented agricultural landscapes, even with a small number of labeled samples. SLFNet provides a robust boundary-aware framework for agricultural parcel delineation and facilitates the extraction of structured agricultural information from high-resolution remote sensing imagery within the Digital Earth context.

The identification of risk regions is important for prioritising their risk mitigation and response efforts. In this paper, we simulate flooding and identify flood-prone as well as inundation areas between 1986 and 2023 in Lagos, Nigeria, using geospatial tools available in the Digital Earth Africa (DEA) Sandbox. By leveraging satellite imagery from Landsat and Sentinel-2 through the Water_extent_WOFS and Water_extent_sentinel notebooks, we evaluated longterm and seasonal water extent using the Modified Normalised Difference Water Index (MNDWI). A 3D terrain model was developed to assess water flow, accumulation, and coastal inundation risk. Also, through document analysis, we identified key policy options that can be strengthened within the study area. Overall, we found that some areas that were inundated with water before 2023 were revealed to be land in 2023, suggesting that water has either receded from these regions or they are being sandfilled. Meanwhile, some areas that were inundated before 2023 have remained flooded to date. These areas might have certain characteristics, such as natural depressions or poor drainage, that make them more prone to prolonged water presence. Also, we found new areas that have only become flooded in recent times, possibly due to changes in land use, climate patterns, or other factors. The implementation of the DEA sandbox geospatial resources shows the utility of the toolbox for environmental resource monitoring and city planning. With respect to the developing Lagos Megacity, our findings highlight the importance of paying attention to the flood-prone areas of the city to ensure sustainable megacity development. The paper concludes with recommendations for policy, urban planning and climate adaptation.

Abstract Floods are the second‐most deadly natural hazard in Australia, following heatwaves. Monitoring flood extent and depth in near real‐time (NRT) is crucial to minimize loss of life and socio‐economic impacts. This study leverages advanced computing, data management systems, and high‐quality data, including river gauge data APIs and Australian Water Outlook, Digital Earth Australia, Google Earth Engine and Amazon Web Service, to develop a flood monitoring workflow in Australia. Our framework provides NRT 5‐m spatial resolution flood extent and depth maps using airborne LiDAR observations through three approaches: (a) gauge data, (b) coupled hydrological and hydrodynamics model, and (c) satellite observations (i.e., Sentinel‐1, Sentinel‐2, Landsat‐7/8/9). We evaluated this flood monitoring framework in seven river catchments across Australia, using both deterministic and ensemble modes. This study highlights the importance of low‐latency gauge data for flood monitoring, as well as the necessity of high‐resolution airborne LiDAR DEMs for accurate flood mapping. In ungauged areas, the ensemble modeling approach enhances the model's ability to capture flood inundation dynamics. In cases where this remains challenging, multi‐source remote sensing can help mitigate the limitations of the modeling approach. We also demonstrated the potential for transferring this flood monitoring framework to other regions around the world. Overall, this study advances the operationalization of high‐resolution flood analytics, offering a replicable blueprint to strengthen community resilience against escalating flood risks under climate change.

Globally, grasslands, savannas, and wetlands are degrading rapidly and increasingly being replaced by woody vegetation. Woody Plant Encroachment (WPE) disrupts natural landscapes and has significant consequences for biodiversity, ecosystem functioning, and key ecosystem services. This review synthesizes findings from 159 peer-reviewed studies identified through a PRISMA-guided systematic literature review to evaluate the drivers of WPE, its ecological impacts, and the remote sensing (RS) approaches used to monitor it. The drivers of WPE are multifaceted, involving interactions among climate variability, topographic and edaphic conditions, hydrological change, land use transitions, and altered fire and grazing regimes, while its impacts are similarly diverse, influencing land cover structure, water and nutrient cycles, carbon and nitrogen dynamics, and broader implications for ecosystem resilience. Over the past two decades, RS has become central to WPE monitoring, with studies employing classification techniques, spectral mixture analysis, object-based image analysis, change detection, thresholding, landscape pattern and fragmentation metrics, and increasingly, machine learning and deep learning methods. Looking forward, emerging advances such as multi-sensor fusion (optical– synthetic aperture radar (SAR), Light Detection and Ranging (LiDAR)–hyperspectral), cloud-based platforms including Google Earth Engine, Microsoft Planetary Computer, and Digital Earth, and geospatial foundation models offer new opportunities for scalable, automated, and long-term monitoring. Despite these innovations, challenges remain in detecting early-stage encroachment, subcanopy woody growth, and species-specific patterns across heterogeneous landscapes. Key knowledge gaps highlighted in this review include the need for long-term monitoring frameworks, improved socio-ecological integration, species- and ecosystem-specific RS approaches, better utilization of SAR, and broader adoption of analysis-ready data and open-source platforms. Addressing these gaps will enable more effective, context-specific strategies to monitor, manage, and mitigate WPE in rapidly changing environments.

Cloud-based Earth observation platforms, such as data cubes, enable reproducible analyses of long-term satellite time series for climate and urban studies. In parallel, Essential Climate Variables (ECVs) provide a standardised framework for monitoring climate dynamics, with urban land cover and temperature being particularly relevant in historic urban contexts. This study analyses long-term trends and statistical associations between satellite-based ECVs and urbanisation indicators within the Historic Urban Landscape (HUL) of Sfax (Tunisia) from 1985 to 2021. Using the Digital Earth Africa (DEA) data cube, we derived six urban spectral indices (USIs), land surface temperature, air temperature at 2 m, wind characteristics, and precipitation from Landsat and ERA5 reanalysis data. An automated and reproducible Python-based workflow was implemented to assess USI behaviour, evaluate their performance against the Global Human Settlement Layer (GHSL), and explore spatio-temporal co-variations between urbanisation and climate variables. Results reveal a consistent increase in air and surface temperatures alongside a decreasing precipitation trend over the study period. The USIs demonstrate comparable accuracy levels (≈88–90%) in delineating urban areas, with indices based on SWIR and NIR bands (NDBI, BUI, NBI) showing the strongest statistical associations with temperature variables. Correlation and multivariate regression analyses indicate that temporal variations in USIs are more strongly associated with air temperature than with land surface temperature; however, these relationships reflect statistical co-variation rather than causality. By integrating satellite-based ECVs within a data cube framework, this study provides an operational methodology for long-term monitoring of urban-climate interactions in historic Mediterranean cities, supporting both climate adaptation strategies and the objectives of the UNESCO HUL approach.

We develop a spatially explicit methane inventory for Alberta’s upstream oil and gas sector using the rHEALPix Discrete Global Grid System. The objective is to demonstrate an equal-area, hierarchy-aware framework that assigns facility-reported emissions to native locations and supports multi-scale analysis and reporting. We compile monthly facility activity from Petrinex for 2020 to 2023, geolocate facilities using the Oil and Gas Infrastructure Mapping database, calculate methane emissions from venting, fuel use, and flaring using province-standard factors, and bin results to rHEALPix cells before exact aggregation to coarser levels. Our analysis revealed persistent high-emission hotspots, with 5% of grid cells accounting for 34% of total annual methane emissions. The equal-area lattice enables fair intensity comparisons across latitude, stable hotspot tracking over time, and mass-conserving aggregation that maintains consistent totals across resolutions. Practical implications include a standard spatial fabric that integrates facility reports, satellites, and ground sensors, provides persistent cell buckets for facility and asset management, enables accurate intensity comparisons across space and time with quantitative spatial resolution, preserves spatial integrity in visualization, supports consistent mass conserving aggregation at any scale with multiple granularities for analysis and reporting, allows precise hotspot tracking and trend monitoring, and informs targeted monitoring and survey design.

<p>Digital Twin Earth (DTE) stands at a transformative crossroads: evolve beyond sophisticated “digital mimicry” into a revolutionary planetary governance engine or remain confined to passive observation. We propose that DTE's transformative potential lies in integrating Human-AI synergy to enable proactive Earth governance through continuous intervention-feedback loops connecting real Earth systems, digital simulations, and policy implementation. This paradigm transforms passive environmental monitoring into active navigation of Earth's trajectory within planetary boundaries. While deliberate large-scale intervention carries inherent risks, the potential benefits—building systemic resilience against cascading environmental crises—substantially outweigh the drawbacks of continued passive evolutionary approach. We posit DTE as the high-fidelity simulation environment and AI as the core analytical intelligence within this dynamic governance loop, enabling human stakeholders to continuously design, test, implement, and refine governance strategies. We propose Digital Cousins as a pragmatic implementation pathway that balances scientific robustness with computational feasibility, offering targeted regional governance solutions. By operationalizing AI-powered DTE as an adaptive decision-making system, humanity can consciously navigate Earth’s futures through integrated sensing, simulation, and policy implementation.</p>

The convergence of Digital Earth technologies and the field of marine environmental support represent a significant trend in global scientific and technological advancement. In response to the characteristics of marine environmental data such as sparsity, dynamic complexity, and multi-scale interdependence, and by integrating the evolution of international Digital Earth technology with China’s independent innovation practices, this study constructs three core methodological paradigms. The first is the element-integrated ontology, which proposes a hybrid representation paradigm of “spatiotemporal embedded field + ontology” to achieve integrated and dynamically evolving modeling of marine physical environments alongside geological, biological, chemical, and other multidimensional elements. The second is the cyclically nested spatiotemporal principle, introducing a spatiotemporal framework of “multi-period temporal decomposition and cross-scale structural transformation” to provide a physically interpretable and uncertainty-quantifiable predictive framework for complex marine phenomena. The third is the information transition and value-adding method, which establishes an information transition mechanism of “structure-semantics-utility” to facilitate the transformation and enhancement of massive raw data into high-value decision-making knowledge. Building on the above research, targeted policy recommendations are proposed across three dimensions: national-level infrastructure planning, data fusion development models, and value-adding channels for industrial applications. These recommendations aim to support the development and industrial implementation of an autonomous and controllable marine Digital Earth technology system in China.

List of reviewers for International Journal of Digital Earth

Multiscale 3D geological characterization and joint analysis are increasingly important topics in spatial information science. However, the non-uniform spatial distribution of objects and scale heterogeneity in geological surveys lead to dispersed storage, long access paths, and limited query performance in managing multiscale 3D geological model data. This study presents a W-Hilbert curve-based multiscale data model (WH-MSDM) that improves data indexing and management through a unified data structure (UDS) for multi-scale blocks and a bidirectional mapping model (BMM) linking spatial coordinates to memory locations. It supports spatial, attribute, hybrid, and cross-scale queries for diverse retrieval tasks. By exploiting the space-filling properties of the W-Hilbert curve to linearize multidimensional geological data into a one-dimensional index, it preserves locality and increases query efficiency across scales. Experimental results on a real 3D geological model demonstrate that WH-MSDM outperforms three mainstream baselines in both unified data organization and diverse query workloads. It thus provides a data-model foundation for Digital Earth-oriented multiscale geological analysis.

Deep learning reduced macadamia planting year error at a national scale, achieving a pixel-level Mean Absolute Error (MAE) of 1.2 years and outperforming a vegetation index threshold baseline (MAE 1.6 years) and tree-based models—Random Forest (RF; MAE 3.02 years) and Gradient Boosted Trees (GBT; MAE 2.9 years). Using Digital Earth Australia Landsat annual geomedians (1988–2023) and block-level, industry-supplied planting year data, models were trained and evaluated at the pixel level under a strict Leave-One-Region-Out cross-validation (LOROCV) protocol; a secondary block-level random split (80/10/10) is reported only to illustrate the more optimistic setting, where shared regional conditions yield lower errors (0.6–0.7 years). Predictions reconstruct planting year retrospectively from the full historical record rather than providing real-time forecasts. The final model was then applied to all Australian Tree Crop Map (ATCM) macadamia orchard polygons to produce wall-to-wall planting year estimates. The approach enables fine-grained mapping of planting patterns to support yield forecasting, resource allocation, and industry planning. Results indicate that sequence-based deep models capture informative temporal dynamics beyond thresholding and conventional machine learning baselines, while remaining constrained by regional and temporal data sparsity. The framework is scalable and transferable, offering a pathway to planting year mapping for other perennial crops and to more resilient, data-driven agricultural decision-making.

Operationalizing Earth Observation (EO)-based Machine Learning (ML) algorithms (or GeoML) for ingestion in environmental Digital Twins remains a challenging task due to the complexities associated with balancing real-time inference with cost, data, and infrastructure requirements. In the field of GHG monitoring, most GeoML models of land use CO2 fluxes remain at the proof-of-concept stage, limiting their use in policy and land management for net-zero goals. In this study, we develop and demonstrate a Digital Twin-ready framework to operationalize a pre-trained Random Forest model that estimates the Net Ecosystem Exchange of CO2 (NEE) from drained peatlands into a biweekly, field-scale CO2 flux monitoring system using EO and weather data. The system achieves an average response time of 6.12 s, retains 98% accuracy of the underlying model, and predicts the NEE of CO2 with an R2 of 0.76 and NRMSE of 8%. It is characterized by hybrid data ingestion (combining non-time-critical and real-time retrieval), automated biweekly data updates, efficient storage, and a user-friendly front-end. The underlying framework, which is part of an operational Digital Twin under the UK Research & Innovation AI for Net Zero project consortium, is built using open source tools for data access and processing (including the Copernicus Data Space Ecosystem OpenEO API and Open-Meteo API), automation (Jenkins), and GUI development (Leaflet, NiceGIU, etc.). The applicability of the system is demonstrated through running real-world use-cases relevant to farmers and policymakers concerned with the management of arable peatlands in England. Overall, the lightweight, modular framework presented here integrates seamlessly into Digital Twins and is easily adaptable to other GeoMLs, providing a practical foundation for operational use in environmental monitoring and decision-making.

ABSTRACT This paper presents a method to obtain Digital Height Models from massive airborne lidar data processing (billions of points) over tens of thousands of km2, along with a new Geospatial User Feedback (GUF) geoservice. The method avoids large errors common in previous procedures and achieves high accuracy thanks to the synergy with other Remote Sensing (RS) data and vector/raster heuristics. Additionally, it provides per-cell metadata about algorithm decisions and reliability (10 categories each), as well as the day-month-year of the lidar pulse used (key for forest growth studies). Tests carried out on 5355 + 4163 points on buildings + forests have shown median errors of 19 and 43 cm, MAE of 64 and 98 cm, and RMSE of 157 and 164 cm. The time-series dataset is available to everyone both for download and through a geoservice following FAIR principles. Moreover, the geoservice includes GUF functionalities allowing users to comment and report issues at both the per-cell level and through areas defined by coordinates (not only at dataset level). The result of this research, LidarTeam, is a teamwork involving RS sources, producers and end users to create a piece of a futuristic library of Digital Earth products, specifically a multitemporal 3D representation of the World.

This study explores the potential of Digital Earth Africa (DE Africa) coastlines products for assessing the Coastal Vulnerability Index (CVI) along the Casablanca coastline, Morocco. The analysis integrates remotely sensed shoreline data with elevation, slope, and geomorphological information from ASTER GDEM and geological maps within a GIS environment. Shoreline change metrics, including Shoreline Change Envelope (SCE), Net Shoreline Movement (NSM), Linear Regression Rate (LRR), and End Point Rate (EPR), were used to evaluate erosion trends from 2000 to 2023. Results show that sandy beach areas, particularly those below 12 m elevation, are highly exposed to erosion (up to 1.5 m/yr) and vulnerable to coastal hazards. Approximately 44% and 23% of the study area were classified as having very high and high vulnerability, respectively. The results indicate that remotely sensed data and GIS techniques are valuable and cost-effective tools for multi-scale geo-hazard coastal assessment studies. The study demonstrates that DE Africa products, combined with local landscape data, provide a valuable tool for coastal vulnerability assessment and monitoring in Africa.

Abstract. Understanding land surface morphology and its relief components, which record the dynamics of the planet's evolution and interaction of multiple environmental factors, constitutes a critical aspect of Earth system science. Advances in Earth observation technologies have enabled access to higher-resolution data, e.g. remote sensing imagery and digital elevation models (DEMs). However, classified relief and landform data with a resolution of approximately 1 arcsec (approximately 30 m) are lacking at the global scale, which limits the progress of related studies at finer scales. Here, we propose a novel framework for global relief classification and release a unique dataset called global relief classification (GRC), which incorporates a comprehensive set of objects that constitute the range of terrains and landforms on Earth. Constructed from multiple 1 arcsec DEMs, GRC covers the global land and ranks among the highest-resolution global geomorphic datasets to date. Its development integrates land surface ontologies, with cores, transitions and boundaries, and key derivatives to strike a balance between mitigating local noise and preserving valuable landform details. GRC categorizes Earth's land relief into two levels, yielding raster files and discrete vector units that record relief type and distribution. Comparative analyses with previous datasets reveal that GRC better captures details of surface morphology, enabling a more precise depiction of geomorphological boundaries. This refinement facilitates the identification of finer and more precise spatial disparities in landform patterns than before, exemplified by marked contrasts between Asia and other continents, and highlights the distinct prominence of Peru and China in terms of relief diversity. Given that the data resolution of GRC accords well with accessible remote sensing imagery and other Earth science datasets, it is readily incorporated into analytical workflows, exploring the relationship between land morphology, surface runoff, climate, and land cover. The full dataset is available on the Deep-time Digital Earth Geomorphology platform and from Zenodo (https://doi.org/10.5281/zenodo.15641257, Yang et al., 2024).

Implementing a digital Earth requires processing an enormous amount of data. Compared with classical algorithms, quantum algorithms can take exponential or quadratic acceleration and are more suitable for processing big data. However, existing quantum models require massive resources to process data on the Earth due to their low compatibility with spheres. So, we propose a quantum representation of the Earth based on the Bloch sphere (QREB), which is tailor-made for the Earth. Based on this model, the scale transformation and specific rotation transformation only need to change one variable. And the time complexity of the quantum point of interest extraction algorithm based on this model is the square root of the classical algorithm. Moreover, in the example of extracting thin black lines on the basketball, the accuracy is improved from 50% to 87.5%. The QREB model can reduce the resources required to process spherical data and accelerate the realization of a digital Earth.