Landform Classification Research Articles

Constructing Soil–Landscape Units Based on Slope Position and Land Use to Improve Soil Prediction Accuracy

Topography is one of the dominant factors in regional soil formation and development. Soil distribution has a certain pattern from high to low in space, and this pattern has a high degree of consistency with slope position. Most of the current research on soil mapping uses landscape types generated by existing methods directly as environmental covariates, and there are few landscape classification methods specifically oriented toward soil surveys. There is rarely any research on landform classification using relative slope position (RSP) and elevation. Therefore, we designed a landform classification method based on RSP and elevation, Terrainforms (TF), and combined the landform type with land use type to construct soil–landscape units for soil type and attribute spatial prediction. In this study, two commonly used landform classification methods, Geomorphons and Landforms, were also used to compare with this design method. It was found that the constructed soil–landscape units had a high consistency with the soil spatial distribution. The landform types based on RSP and elevation obtained the second-highest prediction accuracy in both soil type and soil organic carbon (SOC), and the constructed soil–landscape types obtained the highest prediction accuracy. The results show that the landform classification method based on RSP and elevation is not easily limited by the analysis scale, and is an efficient and accurate landform classification method. The TF landform type and its constructed soil–landscape types can be used as an important environmental variable in soil prediction and sampling, which can provide some guidance and reference for landform classification and digital soil mapping.

Design and Use of a Stratum-Based Yield Predictions to Address Challenges Associated with Spatial Heterogeneity and Sample Clustering in Agricultural Fields Using Remote Sensing Data

Machine learning (ML) models trained with remote sensing data have the potential to improve cereal yield estimation across various geographic scales. However, the complexity and heterogeneity of agricultural landscapes present significant challenges to the robustness of ML-based field-level yield estimation over large areas. In our study, we propose decomposing the landscape complexity into homogeneous zones using existing landform, agroecological, and climate classification datasets, and subsequently applying stratum-based ML to estimate cereal yield. This approach was tested in a heterogeneous region in northern Morocco, where wheat is the dominant crop. We compared the results of the stratum-based ML with those applied to the entire study area. Sentinel-1 and Sentinel-2 satellite imagery were used as input variables to train three ML models: Random Forest, Extreme Gradient Boosting (XGBoost), and Multiple Linear Regression. The results showed that the XGBoost model outperformed the other assessed models. Furthermore, the stratum-based ML approach significantly improved the yield estimation accuracy, particularly when using landform classifications as homogeneous strata. For example, the accuracy of XGBoost model improved from R2 = 0.58 and RMSE = 840 kg ha−1 when the ML models were trained on data from the entire study area to R2 = 0.72 and RMSE = 809 kg ha−1 when trained in the plain area. These findings highlight that developing stratum-based ML models using landform classification as strata leads to more accurate predictions by allowing the models to better capture local environmental conditions and agricultural practices that affect crop growth.

Mapping landforms of a hilly landscape using machine learning and high-resolution LiDAR topographic data

Landform maps are important tools in assessment of soil- and eco-hydrogeomorphic processes and hazards, hydrological modeling, and natural resources and land management. Traditional techniques of mapping landforms based on field surveys or from aerial photographs can be time and labor intensive, highlighting the importance of remote sensing products based automatic or semi-automatic approaches. In addition, the time-intensive manual labeling can also be subjective rather than an objective identification of the landform. Here we implemented such an objective approach applying a random forest machine learning algorithm to a set of observed landform data and 1m horizontal resolution bare-earth digital elevation model (DEM) developed from airborne light detection and ranging (LiDAR) data to rapidly map various landforms of a hilly landscape. The landform classification includes upland plateaus, ridges, convex slopes, planar slopes, concave slopes, stream channels, and valley bottoms, across a 400 km2 hilly landscape of the Ozark Mountains in northeastern Oklahoma. We used 4200 landform observations (600 per landform) and eight topographic indices derived from 2 m, 5 m and 10 m resolution LiDAR DEM in random forest algorithm to develop 2 m, 5 m and 10 m resolution landform models. We test the effectiveness of DEM resolution in mapping landforms via comparison of observed landforms with modeled landforms. Results showed that the approach mapped ∼84% of observed landforms when covariates were at 2 m resolution to ∼89% when they were at 10 m resolution. However, predicted maps showed that the 2 m resolution covariates performed better at capturing accurate landform boundaries and details of small-sized landforms such as stream channels and ridges. The approach presented here significantly reduces the time required for mapping landforms compared to traditional techniques using aerial imagery and field observations and allows incorporation of a wide variety of covariates. The landform map developed using this approach has several potential applications. It could be utilized in hydrological modeling, natural resource management, and characterizing soil-geomorphic processes and hazards in a hilly landscape.

GeomorPM: a geomorphic pretrained model integrating convolution and Transformer architectures based on DEM data

As the domain of artificial intelligence has advanced, the integration of deep learning techniques into terrain and landform analysis has become more prevalent. Nevertheless, many existing methods are fully supervised and designed for specific tasks; thus, their transferability is limited and massive annotated samples are required. This study introduces a geomorphic pretrained model (GeomorPM) capable of performing multiple tasks. First, an architecture was designed that combined a convolution-based Vector Quantised-Variational Autoencoder (VQVAE) with a Transformer-based masked autoencoder (MAE) framework, allowing it to autonomously learn local details and global patterns from large-scale digital elevation model (DEM) data. Subsequently, GeomorPM, based on the VQMAE architecture, was pretrained on massive DEM data and fine-tuned for three specific tasks: DEM void filling, DEM superresolution, and landform classification. GeomorPM outperformed the traditional and other deep learning methods in all three tasks, demonstrating the superior learning ability and transferability of the model. This study provides a practical framework for developing pretrained models based on DEMs that can be expanded to other continuous geoscientific data.

Morphological and Structural Characterization of Shortening Landforms on Mars

AbstractThe lithosphere of Mars accommodates horizontal shortening through folding and faulting, producing landforms described as wrinkle ridges or lobate scarps. Despite this nomenclature, we lack a deep understanding of the drivers of morphological differences observed between landform types. This study aims to develop a quantitative model for shortening landform classification based on surface morphology, subsurface architecture, and strain accommodation, facilitating interpretations of where and how lithospheric stresses are recorded. We developed this model by mapping 100 shortening landforms in a Geographic Information System, recording 12 unique geomorphic parameters such as length and asymmetry, and estimating the strain of each landform. We conducted a Discriminant Function Analysis (DFA) using surface morphometrics. This DFA produced a predictive linear function for categorizing wrinkle ridges and lobate scarps and for quantifying which landforms were exemplars within those categories. The three most influential variables on the surface morphometry DFA were the maximum width, forelimb slope, and back limb length. We then modeled the subsurface structural geology of 50 landforms using MOVE Structural Geology Modeling Software and conducted a second DFA based on subsurface metrics. DFA was most influenced by the dip and depth of the lower ramp base. When both surface morphology and subsurface geometry are input into single DFA, wrinkle ridges and lobate scarps can be distinguished quantitatively 96% of the time. Our results also show that lobate scarps accommodate more strain and imply that studies should consider landform type when interpreting local, regional, and global geological stress histories.

Estimating sediment delivery ratio using the RUSLE-IC-SDR approach at a complex landscape: A case study at the Lower Snowy River area, Australia

Understanding the dynamics of sediment transport and deposition in natural landscapes is critical to developing cost-effective mitigation measures to control soil erosion and protect ecosystems. However, none of a single existing model can quantify sediment delivery ratio (SDR) and the impact factors such as vegetation and geomorphology, especially in a complex landscape. In this case study, we applied an integrated approach including the revised universal soil loss equation (RUSLE) and the index of connectivity (IC) to assess hillslope erosion and SDR, namely RUSLE-IC-SDR, across a complex landscape in the Lower Snowy River area, Australia. The RUSLE factors were derived from a high-resolution (2 m) digital elevation model (DEM), digital soil maps, high-resolution rainfall data and remotely sensed fractional vegetation cover. A seven-class landform classification was delineated from the high-resolution DEM using a fuzzy logic landform model (FLAG). We further examined the impacts of rainfall, vegetation cover and geomorphology on sediment dynamics and distribution across the study area. Field and laboratory data from 10 plot sites across the study area were collected and used for model validation. This case study showed that the RUSLE-IC-SDR approach can assess the overall sediment budget and the impacts of rainfall, vegetation cover and geomorphology across a complex landscape. Findings from this study can identify and track the areas likely to generate high sediment yield for developing ecological restoration, feral animal management and other catchment management measures.

Accuracy assessment of the geomorphon approach to detect ecological sites in the Dry Chaco region of Argentina

Ecological sites are a classification of local-land types based on differences in important environmental factors including soil properties, slope and landscape position within a geomorphologic and climatic zone. The concept is pivotal to conduct adaptative management in arid rangelands because it defines homogeneous areas over which specific soil-vegetation-disturbance dynamic occurs. Geomorphon is a computationally efficient method that uses Digital Elevation Models (DEMs) for the classification of landforms at multiple scales. The objective of this study was to evaluate the accuracy of the geomorphon approach to map ecological sites in a 11,259-ha study area within the Dry Chaco region of Argentina, utilizing local topographic positions along soil catena as a predictor of broader ecological sites classes. We used two DEM pixel resolutions (12.5 m and 30 m) and optimized the flatness threshold (t) and search radius (L) geomorphon parameters. As “ground truth” we used legacy data of 62 soil profiles descriptions. The geomorphon elements detected were reclassified into the three main ecological sites of the study area: highlands, midlands and lowlands. We used the overall disagreement (D) as the main metric for evaluating the accuracy of ecological site classifications. We found that: i) the lowest t value (0.05°) captured subtle topographical changes and thus more effectively reflected the soil-landscape relationship, and ii) larger L values paired with the lower pixel resolution (30 m) diminished the impact of minor landforms, improving the accuracy of ecological site detections. We determined that the geomorphon model with the pixel resolution of 30 m, t-value of 0.05°, and L-value of 12 produced the ecological site classification map with the highest accuracy, achieving a moderate-high accuracy (D of 33.9 %). Our study suggests that the geomorphon approach shows potential for consistent, reproducible and easily updatable ecological site mapping over larger areas.

Will cinereous vulture (Aegypius monachus L.) become extinct in the forests of Türkiye in the future?

Changing climate conditions and different climate scenarios on a global scale are associated with the shrinkage, fragmentation and even extinction of habitats of important bird species. Based on this, the aim is to conduct habitat suitability modelling and mapping of the cinereous vulture (Aegypius monachus), which has the largest body among the four different vulture species in the world, under the circumstances of climate change in Türkiye. MaxEnt method was performed to reveal the current habitat suitability model of the cinereous vulture, which is an indicator of old and high-quality black pine forests in terms of biological diversity in Türkiye. It was determined that the variables contributing to the current habitat suitability model of the cinereous vulture were bedrock, isothermality, landform classification and seasonal precipitation. Chelsa climate scenarios (SSP126-SSP370-SSP585) for the year 2100 were used to reveal the effects of changing climate conditions on the cinereous vulture. Mapping results according to different scenarios were classified as 0.5 – unsuitable, 0.51-0.8 – suitable and 0.81-1.0 – the most suitable habitat. According to the mapping results based on different year and scenarios, cinereous vulture has suitable habitat in a minimum of 16.13% of the study area in the present state, 13.95% in the year 2100 in the SPP126 climate scenario, 10.11% in the SPP370 climate scenario and 7.36% in the SPP585climate scenario. As a result, when the 2100 SSP585 climate scenario mapping was compared to the current habitat suitability mapping, it was determined that habitat suitability for the cinereous vulture decreased by approximately 55%. Therefore, these results will be a source of information to prevent the extinction of the cinereous vulture, in order to protect its current and potential distributions in advance, and to reduce the impact of changing climate conditions.

Remnant of the Late Holocene sand beach reveals ancient settlement-related sea level change from western Thailand

Archaeological findings and analysis of the sand beach at the Ban Khu Bua site in western Thailand indicate substantial sea-level shifts, which relate to sea-level regression during the late Holocene. This research focused on analyzing the paleo-shoreline and determining the age of ancient sandy beach deposits. Satellite imagery (Sentinel-2A) taken in 2021 facilitated classification of the paleo-coastal landforms into four main categories: colluvium, recent floodplain, old tidal flat, and old sandy beach. The latter, running predominantly north to south, is composed of fine to very fine sand characterized by sub-angular to round shapes, high sphericity, and well sorted. The sediment’s mineral composition is primarily quartz, accompanied by smaller rock fragments, organic matter, heavy minerals, and feldspars. Optically Stimulated Luminescence analysis of quartz-rich samples from the inner and outer areas suggested deposition of the old sandy beach sediments between 1,500 and 3,000 years ago in the inner part, pre-dating the establishment of the Khu Bua community in the Dvaravati period. Continuous beach sand deposition along Thao U-Thong Road aligns with the community settlement approximately 200 to 1,000 years ago.

Modeling of Explosive Pingo-like Structures and Fluid-Dynamic Processes in the Arctic Permafrost: Workflow Based on Integrated Geophysical, Geocryological, and Analytical Data

Understanding the mechanisms responsible for the origin, evolution, and failure of pingos with explosive gas emissions and the formation of craters in the Arctic permafrost requires comprehensive studies in the context of fluid dynamic processes. Properly choosing modeling methods for the joint interpretation of geophysical results and analytical data on core samples from suitable sites are prerequisites for predicting pending pingo failure hazards. We suggest an optimal theoretically grounded workflow for such studies, in a site where pingo collapse induced gas blowout and crater formation in the Yamal Peninsula. The site was chosen with reference to the classification of periglacial landforms and their relation to the local deformation pattern, according to deciphered satellite images and reconnaissance geophysical surveys. The deciphered satellite images and combined geophysical data from the site reveal a pattern of periglacial landforms matching the structural framework with uplifted stable permafrost blocks (polygons) bounded by eroded fractured zones (lineaments). Greater percentages of landforms associated with permafrost degradation fall within the lineaments. Resistivity anomalies beneath pingo-like mounds presumably trace deeply rooted fluid conduits. This distribution can be explained in terms of fluid dynamics. N–E and W–E faults, and especially their junctions with N–W structures, are potentially the most widely open conduits for gas and water which migrate into shallow sediments in the modern stress field of N–S (or rather NEN) extension and cause a warming effect on permafrost. The results obtained with a new workflow and joint interpretation of remote sensing, geophysical, and analytical data from the site of explosive gas emission in the Yamal Peninsula confirm the advantages of the suggested approach and its applicability for future integrated fluid dynamics research.

Investigating the spatial distribution of flood inundation and landforms using topographic position index (TPI) and geomorphon-based automated landform classification methods

Investigating the spatial distribution of flood inundation and landforms using topographic position index (TPI) and geomorphon-based automated landform classification methods

The geomorphometric atlas of Romania: an open-access database on landform classifications and morphometric variables

ABSTRACT This paper presents the first digital geomorphometric atlas of Romania. It primarily includes existing international classifications of landforms, as well as land-surface variables (LSVs) derived from Forest and Buildings removed Copernicus Digital Elevation Model (FABDEM) at a resolution of 30 m. Besides the standard moving window of 3 × 3 (90 × 90 m), most LSVs were also derived in moving windows of 11 × 11 (330 × 330 m), and 27 × 27 (810 × 810 m), which represent the topographic grain values specific to the Romanian topography. Up to this point, the atlas contains 10 sets of landform classification data including the geomorphons and 24 sets of LSVs as follows: Slope, Aspect, Sin (Eastness), Cos (Northness), Profile Convexity, Plan Convexity, Longitudinal Curvature, Cross-Sectional Curvature, Minimum Profile Convexity and Multidirectional Hillshade. The data are available open-source for both interactive visualizations on the ArcGIS Online platform and for download on the Google Drive platform.

Discriminating viscous-creep features (rock glaciers) in mountain permafrost from debris-covered glaciers – a commented test at the Gruben and Yerba Loca sites, Swiss Alps and Chilean Andes

Abstract. Viscous-flow features in perennially frozen talus/debris called rock glaciers are being systematically inventoried as part of the global climate-related monitoring of mountain permafrost. In order to avoid duplication and confusion, guidelines were developed by the International Permafrost Association to discriminate between the permafrost-related landform “rock glacier” and the glacier-related landform “debris-covered glacier”. In two regions covered by detailed field measurements, the corresponding data- and physics-based concepts are tested and shown to be adequate. Key physical aspects which cause the striking morphological and dynamic differences between the two phenomena/landforms concern the following: tight mechanical coupling of the surface material to the frozen rock–ice mixture in the case of rock glaciers, contrasting with essential non-coupling of debris to the glaciers they cover; talus-type advancing fronts of rock glaciers exposing fresh debris material from inside the moving frozen bodies, as opposed to massive surface ice exposed by increasingly rare advancing fronts of debris-covered glaciers; and increasing creep rates and continued advance of rock glaciers as convex landforms with structured surfaces versus predominant slowing down and disintegration of debris-covered glaciers as often concave landforms with primarily chaotic surface structure. Where debris-covered surface ice is or has recently been in contact with thermally controlled subsurface ice in permafrost, complex conditions and interactions can develop morphologies beyond simple either–or-type landform classification. In such cases, the remains of buried surface ice mostly tend to be smaller than the lower size limit of “glaciers” as the term is applied in glacier inventories and to be far thinner than the permafrost in which they are embedded.

Coupling morphometric analysis and soil erosion modeling for the characterization of the geomorphological setting in the surrounding of the archaeological site of Chimtou (Central Medjerda Valley, Tunisia)

ABSTRACT This study focuses on the characterization of the geomorphological setting in the hinterland of the archaeological sites of Chimtou and Bordj Hellal located in the central Medjerda Valley, North Tunisia. Our approach integrates the algorithm Geomorphons for semi-automatic landform classification with the soil erosion model Unit Stream Power-based Erosion Deposition (USPED), providing information on the intensity and regional distribution of erosional and depositional processes. Data from geomorphological field mapping provide a database for a ground-truth of the semi-automatic landform classification derived by the algorithm Geomorphons and complemented the database for the creation of a detailed map of the geomorphology in the hinterland of Chimtou. In line with the delineation of the spatial distribution of erosional and depositional processes, the results deepen the understanding of the geomorphology and the sediment routing of this region.

Explanation of the influence of geomorphometric variables on the landform classification based on selected areas in Poland

In recent years, automatic image classification methods have significantly progressed, notably black box algorithms such as machine learning and deep learning. Unfortunately, such efforts only focused on improving performance, rather than attempting to explain and interpret how classification models actually operate. This article compares three state-of-the-art algorithms incorporating random forests, gradient boosting and convolutional neural networks for geomorphological mapping. It also attempts to explain how the most effective classifier makes decisions by evaluating which of the geomorphometric variables are most important for automatic mapping and how they affect the classification results using one of the explainable artificial intelligence techniques, namely accumulated local effects (ALE). This method allows us to understand the relationship between predictors and the model’s outcome. For these purposes, eight sheets of the digital geomorphological map of Poland on the scale of 1:100,000 were used as the reference material. The classification results were validated using the holdout method and cross-validation for individual sheets representing different morphogenetic zones. The terrain elevation entropy, absolute elevation, aggregated median elevation and standard deviation of elevation had the greatest impact on the classification results among the 15 geomorphometric variables considered. The ALE analysis was conducted for the XGBoost classifier, which achieved the highest accuracy of 92.8%, ahead of Random Forests at 84% and LightGBM at 73.7% and U-Net at 59.8%. We conclude that automatic classification can support geomorphological mapping only if the geomorphological characteristics in the predicted area are similar to those in the training dataset. The ALE plots allow us to analyze the relationship between geomorphometric variables and landform membership, which helps clarify their role in the classification process.

Exploring the geomorphological adequacy of the landslide susceptibility maps: A test for different types of landslides in the Bidente river basin (northern Italy)

Landslide susceptibility modelling is a crucial tool for implementing effective strategies in landslide risk mitigation. A plethora of statistical methods is available for generating accurate prediction images; however, the reliability of these models in terms of geomorphological adequacy is often overlooked by scholars. This critical flaw may result in concealed prediction errors, undermining the trustworthiness of the obtained maps. A key aspect of evaluating the geomorphological soundness of these models lies in factor analysis, specifically considering the correlation of explanatory variables with the final susceptibility score rather than solely focusing on their impact on model accuracy.This study delves into research conducted in the Bidente river basin (Italy) that analyes results obtained from slide, flow, and complex susceptibility models using Weight of Evidence (WoE) and Multivariate Adaptive Regression Splines (MARS) statistical methods. The research critically examines each factor class's role in defining susceptibility scores for different landslide typologies. The comparison between susceptibility maps generated by WoE and MARS for each typology (slide = 0.78; flow = 0.85; complex: 0.79) (slide = 0.78; flow = 0.85; complex: 0.79)reveals good to excellent prediction skill, with MARS demonstrating a 5 % higher performance index.The study emphasises the importance of spatial relationships between variables and landslide occurrences, highlighting that individual classes of variables influence the final susceptibility score based on their combined role with other predictor classes. In particular, in this study, results highlight that lithotecnical and landform classification classes delimit the landslide domain, while topographic attributes (steepness, curvatures, SPI and TWI) modulate the score inside. The proposed approach offers insights into investigating the geomorphological adequacy of landslide prediction images, emphasising the significance of factor analysis in evaluating model reliability and uncovering potential errors in susceptibility maps.

Spatio-temporal analysis of geomorphic recovery along an altered ephemeral stream using automated image processing

Ephemeral rivers in the Mediterranean region have been exposed to significant human disturbance over the last century. Recently, there has been a growing interest in restoring their morpho-sedimentary condition, backed by the European Water Framework Directive. Previous research has highlighted the severe geomorphic impacts of sediment extraction, such as reduced channel width and riverbed incision, but the recovery of these channels during annual flows remains poorly understood. This study presents a methodological framework to investigate geomorphologic recovery (signs of landform change) and constraints in the Rambla de Cervera (eastern Spain) from 2018 to 2021. Aerial imagery, LiDAR topography, and field surveys were used for stream segmentation, fluvial landform classification, temporal landform change, and comparison with flow characteristics from a 2D hydraulic model. The study identified three groups of segments with different channel types and recovery mechanisms. Upstream (G1), unconfined valley settings showed minor changes, indicating morphological stability. Middle segments in confined valleys (G2) showed greater alluvial channel recovery and vegetation encroachment in response to flood/no-flood periods. Downstream segments (G3), incised on cemented gravel, were highly responsive to floods but lacked landform stability due to high energy conditions. The findings suggest that channel recovery during annual floods is slow and occurs mainly downstream in degraded areas (G3). Increased sediment flux from upstream and additional sediment storage in new forms are required to recover these channels. This analysis underscores the challenge of interpreting short-term morpho-dynamic changes for long-term river recovery, but provides insight into the morpho-sedimentary conditions required to accelerate recovery of ephemeral streams.

Mapping and Geomorphic Characterization of the Vast Cold-Water Coral Mounds of the Blake Plateau

A coordinated multi-year ocean exploration campaign on the Blake Plateau offshore of the southeastern U.S. has mapped what appears to be the most expansive cold-water coral (CWC) mound province thus far discovered. Nearly continuous CWC mound features span an area up to 500 km long and 110 km wide, with a core area of high-density mounds up to 254 km long by 42 km wide. This study synthesized bathymetric data from 31 multibeam sonar mapping surveys and generated a standardized geomorphic classification of the region in order to delineate and quantify CWC mound habitats and compare mound morphologies among subregions of the coral province. Based on the multibeam bathymetry, a total of 83,908 individual peak features were delineated, providing the first estimate of the overall number of potential CWC mounds mapped in the region to date. Five geomorphic landform classes were mapped and quantified: peaks (411 km2), valleys (3598 km2), ridges (3642 km2), slopes (23,082 km2), and flats (102,848 km2). The complex geomorphology of eight subregions was described qualitatively with geomorphic “fingerprints” (spatial patterns) and quantitatively by measurements of mound density and vertical relief. This study demonstrated the value of applying an objective automated terrain segmentation and classification approach to geomorphic characterization of a highly complex CWC mound province. Manual delineation of these features in a consistent repeatable way with a comparable level of detail would not have been possible.

Classification of Glacial and Fluvioglacial Landforms by Convolutional Neural Networks Using a Digital Elevation Model

Classification of Glacial and Fluvioglacial Landforms by Convolutional Neural Networks Using a Digital Elevation Model

Slope unit-based genetic landform mapping on Tibetan plateau- a terrain unit-based framework for large spatial scale landform classification

Genetic landform classification and mapping provide crucial information concerning the formation and evolution of the earth surface, aiding in understanding geological processes, climate variation, and driving mechanisms of earth dynamics. Conventional pix and object unit-based research paradigms poss challenges in classifying and mapping the genetic landform on a large spatial scale, while the terrain unit-based method offers an inspiring opportunity for this typical landform task. In this context, we herein developed a slope unit (SU)-based landform classification framework, involving improving the mean curvature method for SU segmentation, developing the spectra method to determine the minimal area size of SU, introducing a specific quantitative indicator system for describing SU, and a universal classification model based on XGBoost. Using the entire Tibetan Plateau as the study area, we first implemented the genetic landform map and divided it into nine genetic landform types including aeolian, arid, loess, karst, periglacial, fluvial, glacial, volcanic-lava, and lacustrine. Classification results achieved an encouraging performance with overall accuracy, kappa coefficient, and mean accuracy of up to 90.2%, 84.71%, and 88.65%, respectively. Experimental results suggested that delineated genetic landform regions and their boundaries are generally closely correlated with regional climatic conditions, geological factors, or surface processes. Compared to pix or object-unit-based methods, the SU delineation results possess internal homogeneity and clear terrain boundaries with geo-meaning. Compared to the watershed-unit-based method, our method is better suited for large-scale landform classification. In the end, we confirmed that the proposed framework can be extended to large-scale planetary genetic landform mapping via two case studies on Mars and Moon, which is transferable yet straightforward. In brief, we argued that the proposed framework offered a powerful and effective method in large-scale genetic landform classification, somewhat effectively circumventing the limitations of existing landform classification methods.