MétaCan
Menu
Back to cohort
Record W4408439946 · doi:10.5194/egusphere-egu25-4345

Investigating the Landslide Susceptibility Assessment Methods for Multi-Scale Slope Units Based on SDGSAT-1 and Graph Neural Networks.

2025· preprint· en· W4408439946 on OpenAlexaff
Xiangqi Lei, Hanhu Liu, Zhe Chen, Shaoda Li, Hang Chen, Shuai Zeng, Xiao Wang, Wenqian Bai, Wei Li, Lorenzo Picco

Bibliographic record

Venuenot available
Typepreprint
Languageen
FieldEngineering
TopicSimulation and Modeling Applications
Canadian institutionsInternational Development Research Centre
Fundersnot available
KeywordsLandslideArtificial neural networkScale (ratio)GraphComputer scienceGeologyData miningArtificial intelligenceGeographyGeotechnical engineeringCartographyTheoretical computer science

Abstract

fetched live from OpenAlex

Landslide susceptibility assessment is crucial for preventing landslide risks. However, existing methods only consider local environmental features related to landslides, neglecting remote yet interconnected geographical features, leading to unreliable landslide susceptibility maps. This study fully considers the complex terrain and landform features of mountainous areas where landslides occur. From the perspectives of mapping units and susceptibility assessment models, it introduces geographical environmental correlations to achieve a comprehensive association between landslides and affected environments, thereby improving the accuracy of landslide susceptibility assessments. At the same time, since the world's first scientific satellite dedicated to serving the United Nations 2030 Agenda for Sustainable Development, the Sustainable Development Goals Scientific Satellite 1 (SDGSAT-1), was launched in 2021, its potential in monitoring and assessing landslide disasters remains to be developed. Therefore, this study innovatively applies SDGSAT-1 data in the field of landslide research and conducts landslide susceptibility assessment in Jiulong County, Ganzi, based on the optimal scale slope units and Graph Neural Networks (GNN).We propose the following method: First, establish appropriately sized slope units using R.Slopeunits to simulate complex mountainous terrain. Second, extract various landslide influencing factors using SDGSAT-1 satellite imagery data. Then, select the most representative graph nodes by constraining environmental similarity and influencing factor feature similarity, constructing a graph structure. Finally, perform landslide susceptibility assessment in the study area using the GraphSage model, which includes environmental information aggregation.This study's distinctive feature lies in fully considering the complex terrain and landform characteristics of mountainous areas where landslides occur. From the perspectives of mapping units and evaluation models, it introduces geographical environmental correlations to achieve a comprehensive association between landslides and affected environments. Furthermore, to validate the effectiveness of the proposed method, we selected raster units and the classic Artificial Neural Network (ANN) model as control experiments. Simultaneously, we conducted comparative experiments using Landsat and SDGSAT-1 satellite imagery, analyzing differences from two aspects: landslide influencing factors and landslide susceptibility evaluation results.The results indicate that: (1) Compared to the commonly used Landsat series satellite data in previous studies, SDGSAT-1 satellite imagery offers higher spatial resolution, capturing more spectral information with richer hue and detail. Additionally, it can generate more angles of landslide influencing factors compared to Landsat satellite data. (2) Employing global heterogeneity evaluation metrics allows for reasonable determination of slope unit scales, thereby maximizing internal consistency and external heterogeneity control within slope units. (3) By utilizing the Graph Neural Network (GNN) model that incorporates environmental information aggregation for landslide susceptibility assessment in the study area, it can, to some extent, overcome spatial limitations and integrate complex mountainous environmental information, facilitating the induction of reliable landslide characteristics.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.002
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.019
Threshold uncertainty score0.038

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.002
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0020.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.097
GPT teacher head0.396
Teacher spread0.299 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicSimulation and Modeling ApplicationsFrench-language works237,207