Global Digital Elevation Model Research Articles

Geology of Tolstoj quadrangle (H08), Mercury

ABSTRACT Tolstoj quadrangle (H08) is located in the equatorial area of Mercury, between 22.5°N and 22.5°S and 144° and 216°E. Using the NASA/MESSENGER data, we compiled a geological map of the quadrangle at a scale of 1:3.000.000. The main basemap we used was the MDIS 166 m/pixel BDR mosaic. In addition, mosaics with high-incidence illumination from west and east, MDIS global color mosaics, and the MDIS global DEM have been taken into account. In the map, we considered three feature classes: geological units, lineaments, and surface features. Geological units consist of crater material and plains. Lineaments include crater and pit rims, and structures. Finally, surface features are subdivided into crater chains or clusters, hollows, faculae, bright and dark material. The geological map will be integrated into the global 1:3M geological map of Mercury, which is being prepared in support to the ESA/JAXA BepiColombo mission.

Stereo bathymetry to monitor small seasonal agriculture water ponds in ungauged areas

Abstract. Small reservoirs represent a critical water supply to farmers across semi-arid regions, but their hydrological modelling suffers from data scarcity and highly variable and localised rainfall intensities. Over 200,000 ancient rainwater harvesting reservoirs (“tanks”) exist across South India, but with their complex history, considerable size variation, and widespread distribution, understanding the hydrological role of these tanks has been difficult. Fortunately, the last decade has seen improvements in sensors and technologies that enhance our ability to assess the hydrological role of these tanks. In particular, high-resolution Digital Elevation Model (DEMs), now much easier to produce, can be used to improve the characterization of tanks and their surrounding watersheds. Here, a high-resolution DEM is created during the lowest reservoir conditions using Pléiades stereoscopy, and along with two global DEMs, compared with volume estimates from a field-derived UAV reference DEM for a set of tanks in South India. This study demonstrates that a Pléiades-derived DEM can capture accurate reservoir geometry and, when simulating water volume, achieves volumetric differences of 2–8% compared to our UAV reference data. The Pléiades-derived-DEM produced an equivalent height bias less than the expected bias for the recently launched Surface Water and Ocean Topography (SWOT) mission. Deriving high-resolution tank bathymetry from space during low-water conditions provides an opportunity to systematically and repeatably measure tank volume. These results are encouraging in efforts to utilize very high resolution DEMs chosen at an appropriate hydrological time, particularly in regions where water management and security is paramount.

Terrain analysis in forested areas: consequences of using DSM instead of DTM

Abstract. Relief mapping through dense tropical forest is a challenge, which can be met by processing radar images. Various global digital elevation models (DEMs) are available, either free or paid, the most widely used methods for producing them at global scale being photogrammetry and short wavelength radar interferometry. However, the resulting DEMs do not directly represent the ground surface but rather the canopy or something in between. Since most users need terrain models for a variety of applications in geoscience, water management etc. but use such canopy models for lack of anything better, it was relevant to assess the consequences of this inappropriate but unavoidable choice. This paper presents a study carried out in the Brazilian Amazon, where an airborne P-band radar interferometric DEM was used as a reference based on its accuracy assessed in previous studies. A well-established DEM (SRTM) and newer ones (Copernicus and Fabdem), as well as a DEM obtained from aerial X-bands radar data, were compared to the reference according to different criteria related to elevation (accuracy and a typical application, i.e., dam filling simulation) and to slope at different scales (accuracy and a typical application, i.e., terrain classification). The results highlighted the risks of using short wavelengths to represent the terrain and emphasized the importance of slope in different resolution scales.

Assessment of digital elevation models accuracy for local geoid modeling

One of the critical factors influencing the accuracy of a local geoid model is the quality of the digital elevation model (DEM). A properly selected high-resolution DEM can significantly mitigate errors in geoid modeling, gravity anomaly processing, and topography and downward continuation correction. Purpose. Evaluating the accuracy of five global DEMs obtained from open sources to identify the most suitable model for creating a local geoid. Methodology. The vertical accuracy of the DEMs was assessed by comparing the heights between the DEM and control points across different types of terrain. The reference values are based on 344 ground benchmarks, where GNSS observations were performed with subsequent adjustment of coordinates and heights. The accuracy analysis involved calculating statistical indicators of the height differences between the GNSS data and the DEM data. Findings. The standard deviation assessment showed favorable values for the COPERNICUS and ALOS DEMs, followed by SRTM, ASTER, and ETOPO. In the mean absolute error calculations for mountainous areas, the ALOS model performed best, followed by COPERNICUS, SRTM, ASTER, and ETOPO. For other types of terrain, COPERNICUS demonstrated the best results in mean absolute error. Originality. This study distinguishes itself through the incorporation of advanced high-resolution DEMs, such as GLO30, providing a modern and thorough evaluation of DEM accuracy specifically for Kazakhstan. What is new is a detailed analysis of the impact of terrain features (plain, hilly, mountainous) on modeling accuracy. This approach advances beyond previous assessments, delivering new and significant insights into the performance of contemporary DEMs. Practice value. The practical value of the results obtained consists in issuing recommendations regarding the possibility of using the studied DEM for the regions of Kazakhstan which differ among themselves in terms of landscape characteristics. The findings indicate that COPERNICUS and ALOS DEMs are highly suitable for precise geoid modeling in southern Kazakhstan. These models can significantly improve the accuracy of local geoid models, benefiting applications in geospatial science and engineering.

Assessing the accuracy of global digital elevation models for hydrological analysis: the cases of Syria and Russia

Digital elevation models provide valuable information about the surface topography. The resolution and accuracy of the data used to create the models are critical for the results of hydrological analysis. Available global digital elevation models obtain high potential in terms of data sources; however, their performance requires meticulous evaluation for different terrain characteristics and applications. Aim. To determine the accuracy and resolution of data provided by current and old global models, namely SRTM 1, SRTM GL1, V3 ASTER, GMTED2010, PALSAR ALOS, and GTOPO 30, using the example of two terrains. The study considers the areas in Syria (Syrian coast) and Russia (Chernoyarsky and Akhtubinsky districts in the Astrakhan region) represented by different elevation models. A matching degree between model and real elevations measured by GPS was determined to carry out the hydrological analysis of the land surface. For this purpose, three statistical measures, including range, standard deviation and correlation were defined by means of BaseCamp, ARCGIS PRO & SAGA_GIS software. Data resolution indicates the degree of detail represented in the DEM dataset, which is established by the spatial sampling interval or the size of the grid cell used to represent the land surface. Data accuracy refers to the matching degree between the elevation values derived from the DEM and the actual data obtained on the surface. Results. In terms of correlation coefficient and standard deviation, the GMTED 2010 and ASTER V3 model for the Syrian region and the ALOS PALSAR and SRTM GL 1 model for the Russian region prove to be the most effective for hydrological analysis in the absence of accurate local models.

Elevation-Dependent Snow Cover Dynamics and Associated Topo-Climate Impacts in Upper Indus River Basin

In the present study, Improved Moderate Resolution Imaging Spectro-radiometer (MODIS) snow cover product (MOYDGL06*) has been used to evaluate the snow cover area (SCA) in Kabul, Jhelum, and Indus river basins for the time period of 2003 – 2020 with available MODIS land surface temperature (LST), and CHIRPS (precipitation) with objectives to evaluate the spatio-temporal SCA, and climate variables with respect to different elevations analyzed from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model Version 3 (GDEM v3) and also to correlate the climatic variables with SCA. The results presented average annual SCA is around 50.7% – 64.7% in sub-basins of UIB, further it has been observed that SCA is decreasing on annual and seasonal timescale in all three basins. Elevation-dependent SCA, temperature, and precipitation presented a mix of trend on annual, seasonal, and monthly timescale at lower and higher altitude in all selected basins. Moreover, it was noticed that topography (slope, & aspect) also influences the SCA in the region. Furthermore, it has been examined that temperature has significant inverse relationship with SCA at middle and higher altitude in Indus, while in Kabul, and Jhelum no significant relationship observed at extreme lower and higher altitudes. It is also evident from relationship between SCA and climate variable that temperature is significantly responsible for decreasing trend of SCA rather than intense precipitation in all three river basins. Thus, all these elevation-dependent changes can improve our hydrological understanding which can have a considerable implication for hydrology, climate science, water resource management and socio-economic activities.

FLOOD RISK MAPPING IN THE IRTYSH RIVER BASIN USING SATELLITE DATA

Floods are among the most frequent and devastating natural disasters, causing significant economic damage and loss of life worldwide. Effective flood risk management relies on accurate modeling techniques that can predict vulnerable areas and assess potential impacts. In this study, flood dynamics are simulated in the Irtysh River Basin near Ust-Kamenogorsk, a city in East Kazakhstan prone to seasonal flooding using high-resolution satellite imagery and digital elevation data. The primary objective is to visually model flood risks based on terrain characteristics. The study utilizes imagery sourced from the Mapbox platform, which combines data from MODIS, Landsat 7, Maxar, and the Google Earth Engine, providing access to Sentinel-2 surface reflectance imagery at 10-meter resolution. Elevation data from the Copernicus global digital elevation model, with a 30-meter resolution, is used to simulate flood progression. The flood simulation involves calculating flood depth relative to the terrain’s elevation, allowing for a pixel-by-pixel determination of submerged areas. Each simulation incrementally increases water levels to generate a sequence of images, showcasing the progression of flooding over time. The study describes hydraulic soil characteristics usage, and focuses on visualizing flood risk based on terrain data and water level changes. The simulation results indicate that flooding initially impacts riverbanks as water flow starts from the northwest of the city with critical infrastructure becoming vulnerable once water levels exceed 2 meters from the lowest elevation point. These findings highlight the potential of high-resolution satellite imagery and terrain data for flood risk assessment and improving urban flood preparedness. The results provide valuable insights into flood progression enabling more informed decision-making for disaster mitigation.

Transfer learning reconstructs submarine topography for global mid-ocean ridges

Mid-ocean ridges are unique, tectonically active geographical units on Earth that profoundly control the ocean environment and dynamics at the global scale. However, high-resolution topographic data from mid-ocean ridges are rarely available due to the difficulty in detecting ocean floors, which further limits ocean research at the global scale. Here, we divide the global mid-ocean ridge system into 2805 tiles and reconstruct their high-resolution topography by using a transfer learning approach with freely available low-resolution digital elevation models (DEMs) and limited high-resolution DEMs. A high-frequency terrain feature-based deep residual network is proposed to generate high-resolution global mid-ocean ridge DEMs. In this network, topographic knowledge related to mid-ocean ridges is integrated and quantified to improve the learning efficiency and reconstruction quality of the network. A series of verifications and evaluations demonstrate the reliability of reconstructed topographies for submarine topography research. We observe that reconstructed topography can achieve good environmental understanding and information acquisition in the global mid-ocean ridge range. We find that the complexity of the previous terrain environment is underestimated by 26.63% in terms of the slope gradient and by 14.95% in terms of terrain relief, while a 101.10% information improvement can be obtained for the reconstructed topography. The reconstructed topography indicates that diverse and intricate topographical environments of mid-ocean ridges exist among different ocean regions. The proposed transfer learning method for reconstructing high-resolution mid-ocean ridge topographies is valuable and can be utilized for reconstructing information in regions that are difficult to observe directly and lack sufficient data.

Spatial distribution and dynamics of thermocirques in a key area of Central Yamal based on remote sensing data

Т. In recent decades, the distribution and activation of thermodenudation, which leads to the formation of specific landforms — thermocirques (also referred to as retrogressive thaw slumps, RTS), have been intensively studied. In different regions of the cryolithozone and at different time intervals, both activation and stabilization of thermocirques are observed. As a rule, studies focus on the climatic controls of the phenomena observed, the environmental controls are discussed less often. This study presents an analysis of the dynamics of thermocirques in relation to the relief features in a specific key area of Central Yamal. To achieve this aim, the spatial distribution of thermocirques at different geomorphological levels is considered based on multi-temporal remote sensing data. Satellite images obtained in 2009, 2018, and 2023, as well as a global digital elevation model (ArcticDEM), were used. We outlined five geomorphic levels and determined their parameters: area, altitude, steepness, and the aspect of the slopes. Thermocirques were identified in the images and their parameters were measured. The dynamics of the thermocirques were analyzed by their number, area, length, width, slope aspect and angle for the periods 2009–2018 and 2018–2023, and for 14 years in total, separately for each geomorphic level. It was found that thermocirques predominate on the slopes of the III alluvial-marine plain, 5–12° steep. In 14 years, the total area of thermocirques increased by 296 %, and their number — by 61 %. A larger increase in the total area and number of thermocirques occurred during the period 2009–2018 in response to climate extremes in 2012 and 2016. Thermocirques that face west cover a higher total area, partly due to the predominance of such slopes over the area of the key site. In all the years of observation, the average areas and lengths of thermocirques are maximum on south-facing slopes. Some of the results are close to those obtained in other regions of Russia and in North America. In many of the areas studied, the increase in the total area of thermocirques exceeded the increase in their number, which means that the expansion of the existing forms prevails over the inception of the new ones. The discrepancies observed in different studies in the results of assessing the effect of relief on thermocirques are due to both the regional features and differences in satellite imagery and methods of its processing.

Spatial interpolation of global DEM using federated deep learning

The digital elevation model (DEM) provides important data support for 3D terrain modeling. However, due to the complex and changeable terrain in the real world and the high cost of field measurement, it is extremely difficult to obtain continuous and high-density elevation data directly. Therefore, it is necessary to rely on spatial interpolation technology to restore the DEM overall picture in the original sampling area. The traditional spatial interpolation method usually has the characteristics of low model complexity and high computational cost, which leads to low real-time performance and low precision of the interpolation process. The interpolation operation based on DEM data can be considered as a special image generation process where the input is a DEM image with missing values and the output is a complete DEM image. At present, a large number of studies have proved that deep learning methods are very effective in image generation tasks. However, the training of deep learning models requires the support of a large number of high-quality data sets. DEM data in various countries, especially in key regions, are usually restricted by privacy protection regulations and cannot be disclosed. The emergence of Federated Learning (FL) provides a new solution, which supports local training on multiple end nodes, without sending local data to a remote center server for centralized training, effectively protecting data privacy. In this study, we propose a DEM interpolation model based on FL and multiScale U-Net. The experimental results show that compared with the traditional method, this model has faster processing speed and lower interpolation precision. At the same time, this research result provides a new way for efficient and secure use of terrain information, especially in those application scenarios that have strict requirements for DEM data privacy and security.

Ranking of 10 Global One-Arc-Second DEMs Reveals Limitations in Terrain Morphology Representation

At least 10 global digital elevation models (DEMs) at one-arc-second resolution now cover Earth. Comparing derived grids, like slope or curvature, preserves surface spatial relationships, and can be more important than just elevation values. Such comparisons provide more nuanced DEM rankings than just elevation root mean square error (RMSE) for a small number of points. We present three new comparison categories: fraction of unexplained variance (FUV) for grids with continuous floating point values; accuracy metrics for integer code raster classifications; and comparison of stream channel vector networks. We compare six global DEMs that are digital surface models (DSMs), and four edited versions that use machine learning/artificial intelligence techniques to create a bare-earth digital terrain model (DTM) for different elevation ranges: full Earth elevations, under 120 m, under 80 m, and under 10 m. We find edited DTMs improve on elevation values, but because they do not incorporate other metrics in their training they do not improve overall on the source Copernicus DSM. We also rank 17 common geomorphic-derived grids for sensitivity to DEM quality, and document how landscape characteristics, especially slope, affect the results. None of the DEMs perform well in areas with low average slope compared to reference DTMs aggregated from 1 m airborne lidar data. This indicates that accurate work in low-relief areas grappling with global climate change should use airborne lidar or very high resolution image-derived DTMs.

A Robust Tie-Points Matching Method with Regional Feature Representation for Synthetic Aperture Radar Images

The precise tie-points (TPs) on synthetic aperture radar (SAR) images are a critical cornerstone in the global digital elevation model (DEM) and digital ortho map (DOM) production process. While there are abundant studies on SAR TPs matching, improvement opportunities persist in large areas. The correspondences have pixel-level errors during geocoding, which result in misalignment between global products. Consequently, this paper proposed a robust method for SAR images TPs matching, which consists of three key steps: (1) interest point extraction based on the dynamic Harris area entropy (DHAE) grid; (2) adaptive determination of template size; (3) normalized cross correlation (NCC) template matching. DHAE is a regional texture information grid based on the SAR-Harris map, and it is achieved through dynamic block division. Generating the DHAE grid over SAR images enables the extraction of interest points that have regional feature representation and distribution uniformity. A variable-size matching template is adaptively determined based on DHAE to enhance template quality while maintaining computational efficiency. Subsequently, the NCC algorithm is employed to find subpixel-precise correspondences. The proposed method is applied on TPs matching in 57 Terra-SAR images, which cover a large geographical area. Furthermore, the overlapping area is partitioned into five segments according to different coverage types. The experimental results demonstrate that the proposed method outperforms other template matching methods. For all coverage types, the proposed method exhibits high-precision sub-pixel results that reach up to 38.64% in terms of the relative positioning error (RPE), particularly in texture-weak and large areas.

Decadal-scale assessment of sediment denudation rates in the Zhoukou River Basin, Taiwan: insights from improved DEMs of differencing based on spectral analysis

This study assesses sediment erosion rates in the Zhoukou River Basin in southern Taiwan over the past three decades, focusing on the impact of extreme rainfall events. While various established methods determine erosion rates over different temporal scales, we employ an independent approach for decadal-scale erosion rate calculation by utilizing open-source global and regional digital elevation models (DEMs) data for two distinct periods. We introduce a new method that applies Fourier analysis to vertically register the DEMs, significantly minimizing their vertical bias. Through spectral analysis, we identify long wavelength topography crucial for correcting vertical offsets. Erosion rates, computed through DEMs of Difference (DoD), exhibit a similar trend of significant reduction in sediment export rates from 1990–2010 to 2011–2020 due to decreased extreme rainfall events, aligning with erosion rate estimates derived from mean suspended load data at gauge stations. Over the entire period from 1990 to 2020, the calculated denudation rate was 14.19 mm/yr, whereas in the recent decade (2011–2020), it decreased to 10.46 mm/yr. Our study suggests that the improved DoD method can effectively estimate sediment transport rates by leveraging underutilized DEMs captured at distinct points in time, especially when the erosional signal dominates data noise.

A dataset of the distribution and characteristics of debris flows along the National Highway 318 in the Sino-Nepalese transport corridor (2022)

Due to the unique geographical and geological conditions in the Sino-Nepal transportation corridor, there is a lack of reliable and comprehensive data on debris flow distribution data in the region. This dataset is compiled based on early identification of debris flows and field surveys, integrating Digital Elevation Models (DEM) and geological maps. It provides a catalog and detailed feature information of debris flows within the corridor along the National Highway 318 between China and Nepal from 1990 to 2022. Precipitation and temperature parameters are derived from 30 years (1980–2009) of interpolated data from national meteorological stations. Fault distribution information is sourced from the 1:50,000 digital geological maps published by the China Geological Survey. Terrain information is obtained from NASA's Shuttle Radar Topography Mission (SRTM) data, which provides a global DEM at a resolution of 12.5 m. Within the geographical coordinates of 28.0°–29.4°N and 85.1°–88.9°E, a total of 255 debris flow records have been extracted, covering an area of 1660.73 km². We primarily employed hydrological analysis methods in ArcGIS for the extraction of debris flow basins. Considering the limitations in DEM accuracy, we made necessary manual corrections with reference to Google Earth imagery. The dataset comprises three files: a vector file depicting the main distribution range of debris flows along the National Highway 318 in the Sino-Nepalese transport corridor; a vector file containing the catalog of debris flows; and a document detailing the parameters and characteristics of the debris flows.The dataset has undergone on-site field surveys and confirmation to ensure its reliability. Therefore, it can serve as a foundational dataset applicable to debris flow studies in the region.

Global Attention-Based DEM: A Planet Surface Digital Elevation Model-Generation Method Combined with a Global Attention Mechanism

Digital elevation models (DEMs), which can provide an accurate description of planetary surface elevation changes, play an important role in scientific tasks such as long-distance path planning, terrain analysis, and planetary surface reconstruction. However, generating high-precision planetary DEMs currently relies on expensive equipment together with complex remote sensing technology, thus increasing the cost and cycle of the task. Therefore, it is crucial to develop a cost-effective technology that can produce high-quality DEMs on the surfaces of planets. In this work, we propose a global attention-based DEM generation network (GADEM) to convert satellite imagery into DEMs. The network uses the global attention mechanism (GAM) together with a multi-order gradient loss function during training to recover precise terrain. The experimental analysis on lunar and Martian datasets not only demonstrated the effectiveness and accuracy of GADEM in bright regions, but also showed its promising reconstruction ability in shadowed regions.

ANADEM: A Digital Terrain Model for South America

Digital elevation models (DEMs) have a wide range of applications and play a crucial role in many studies. Numerous public DEMs, frequently acquired using radar and optical satellite imagery, are currently available; however, DEM datasets tend to exhibit elevation values influenced by vegetation height and coverage, compromising the accuracy of models in representing terrain elevation. In this study, we developed a digital terrain model for South America using a novel methodology to remove vegetation bias in the Copernicus DEM GLO-30 (COPDEM) model using machine learning, Global Ecosystem Dynamics Investigation (GEDI) elevation data, and multispectral remote sensing products. Our results indicate considerable improvements compared to COPDEM in representing terrain elevation, reducing average errors (BIAS) from 9.6 m to 1.5 m. Furthermore, we evaluated our product (ANADEM) by comparison with other global DEMs, obtaining more accurate results for different conditions of vegetation fraction cover and land use. As a publicly available and open-source dataset, ANADEM will play a crucial role in advancing studies that demand accurate terrain elevation representations at large scales.

Interferometric Calibration Model for the LuTan-1 Mission: Enhancing Digital Elevation Model Accuracy

The LuTan-1 (LT-1) mission, China’s first civilian bistatic spaceborne Synthetic Aperture Radar (SAR) mission, comprises two L-band SAR satellites. These satellites operate in bistatic InSAR strip map mode, maintaining a formation flight with an adjustable baseline to generate global digital elevation models (DEMs) with high accuracy and spatial resolution. This research introduces a dedicated interferometric calibration model for LT-1, tackling the unique challenges of the bistatic system, such as interferometric parameter coupling and the π-ambiguity problem caused by synchronization phase errors. This study validates the model using SAR images from LT-1 and Xinjiang corner reflector data, achieving interferometric phase accuracy better than 0.1 rad and baseline accuracy better than 2 mm, thereby producing high-precision DEMs with a height accuracy meeting the 5 m requirement.

Comparison of machine learning and statistical approaches for Digital Elevation Model (DEM) correction: interim results

Abstract. The correction of digital elevation models (DEMs) can be achieved using a variety of techniques. Machine learning and statistical methods are broadly applicable to a variety of DEM correction case studies in different landscapes. However, a literature survey did not reveal any research that compared the effectiveness or performance of both methods. In this study, we comparatively evaluate three gradient boosted decision trees (XGBoost, LightGBM and CatBoost) and multiple linear regression for the correction of two publicly available global DEMs: Copernicus GLO-30 and ALOS World 3D (AW3D) in Cape Town, South Africa. The training datasets are comprised of eleven predictor variables including elevation, slope, aspect, surface roughness, topographic position index, terrain ruggedness index, terrain surface texture, vector ruggedness measure, percentage bare ground, urban footprints and percentage forest cover as an indicator of the overland forest distribution. The target variable (elevation error) was derived with respect to highly accurate airborne LiDAR. The results presented in this study represent urban/industrial and grassland/shrubland/dense bush landscapes. Although the accuracy of the original DEMs had been degraded by several anomalies, the corrections improved the vertical accuracy across vast areas of the landscape. In the urban/industrial and grassland/shrubland landscapes, the reduction in the root mean square error (RMSE) of the original AW3D DEM was greater than 70%, after correction. The corrections improved the accuracy of Copernicus DEM, e.g., > 44% RMSE reduction in the urban area and >32% RMSE reduction in the grassland/shrubland landscape. Generally, the gradient boosted decision trees outperformed multiple linear regression in most of the tests.

Characteristics and effects of global sloping land urbanization from 2000 to 2020

Cities usually expand on flat land. However, in recent decades, the increasing scarcity of available flat land has compelled many cities to expand to sloping land (sloping land urbanization, SLU), and the understanding for global SLU is still unclear. This study, based on the currently available high-precision global Digital Elevation Model (FABDEM) and global land cover dataset (GlobeLand30), investigated the characteristics and impacts of SLU in 26,402 urban residential areas worldwide from 2000 to 2020. Results show that the total area of SLU globally is 16,383 km2, accounting for 9.54 % of the overall urban expansion. This phenomenon is widespread globally and relatively concentrated in a few countries, with 42.78 %, 24.35 %, and 21.83 % of the area coming from cultivated land, forest, and grassland respectively. Global SLU has accommodated 34.78 million urban population, and indirectly protected 8922 km2 of flat cultivated land, while causing a net loss of 4373 km2 of green ecological land. Deliberately balancing the dual effects of SLU is crucial for advancing sustainable global urbanization.

Detection of karst depression in Brazil comparing different semantic and instance segmentations and global digital elevation models

This research assessed the efficacy of deep segmentation in identifying and measuring natural karst depressions in the Bambuí Group's carbonate rocks in Western Bahia, Brazil. The investigation used five Global Digital Elevation Models (DEMs) with 30-m resolution: Advanced Spaceborne Thermal Emission and Reflection Radiometer - Global Digital Elevation Model version 3 (ASTER-GDEM v3), Advanced Land Observing Satellite World 3D – 30 m version 3.2 (AW3D30 v3.2), Copernicus 30 m global DEM (GLO-30), NASA Digital Elevation Model version 1 (NASADEM v1), and Shuttle Radar Topography Mission version 3 (SRTM v3). The karst feature detection analysis compared five semantic segmentation architectures: Feature Pyramid Networks (FPN), LinkNet, UNet, UNet++, DVL3+, using an EfficientNet-B7 backbone, one instance segmentation model (Mask-RCNN), and a semantic-to-instance conversion method. This comparison considers different datasets, including one, two, or eleven variables (DEM, DEM-based sink depth, and nine terrain attributes). Besides, we performed a combinatorial analysis of the DEMs to evaluate the improvement in karst feature detection. The methodology involved the generation of geomorphometric attributes, sample labeling from Sentinel-2 and Operational Land Imager-Landsat 8 datasets, training-validation-testing steps with 128 × 128 samples, reconstruction of large images for semantic and instance segmentation, and accuracy analysis. The findings revealed that GLO-30 and AW3D30 data were the most accurate, while ASTER GDEM performed poorly in both segmentation forms. Among the semantic models, FPN showed the highest accuracy. The 11-variable models preferentially outperformed those with fewer in both types of segmentation. The approach of semantic-to-instance conversion with Geographic Information System tools favored individualizing karst depressions and obtaining efficient quantification, consisting of an easy alternative to achieve instance segmentation. Reconstruction of large remote sensing images through sliding windows considering the specifics of instance and semantic segmentation demonstrated that smaller stride enhances object coverage and reduces the risk of losing crucial information, effectively improving the predictions. The combinatorial analysis for semantic and instance segmentation indicates that models incorporating more DEMs (4- and 5-DEM models) and variables achieve generally higher accuracy. The best result in semantic segmentation was combining the five DEM datasets using 11 variables, reaching an F-score of 85.06 and IoU of 74.00. In instance segmentation, the best result for the bounding box was the model that integrates AW3D30, GLO-30, NASADEM, and SRTM with eleven variables reaching Average Precision (AP) of 45.64, AP50 of 88.40 and AP75 of 39.38, while the best result for the segmentation mask was the model that integrates the five models with eleven variables with AP of 42.61, AP50 of 86.79 and AP75 of 35.43. These results demonstrate that integrating a wide range of DEM data, even the lowest performing ones, improves model generalization and accuracy in segmentation, leveraging strengths and mitigating individual weaknesses. Future work could explore high-resolution DEMs and the integration of various deep-learning methods.