Objectives

The LEAD-Landslide project aims to transform landslide detection, mapping, and monitoring by combining Earth Observation (EO) data with state-of-the-art Artificial Intelligence (AI). The project defines four major scientific and operational objectives, each associated with distinct innovation elements and work packages.

O1 — Advance Landslide Detection through AI and Multi-Sensor Data Integration

This objective focuses on developing cutting-edge AI models that integrate multispectral, hyperspectral, SAR, UAV, and LiDAR datasets to improve landslide detection across varied environments. The project advances beyond current approaches by using CNNs, transformers, hybrid AI models, and super-resolution algorithms to capture detailed landslide signatures.
Goal: Significantly increase detection accuracy for subtle, diffuse, or vegetation-covered landslides.
Supported by: WP1 & WP5.

O2 — Build a Time-Enabled Inventory of Landslide Events Using Long-Term Satellite Data

This objective develops a temporal landslide inventory by analysing historical and contemporary satellite imagery (Landsat, Sentinel-1/2, Planet) and UAV data. The aim is to detect changes, trace the evolution of landslides through time, and identify reactivation patterns.
Goal: Provide long-term insights into landslide dynamics and triggering conditions.
Supported by: WP2 & WP5.

O3 — Create a Multi-Scale, Multi-Temporal Landslide Mapping and Communication Framework

This objective delivers a flexible and scalable framework that transforms complex landslide information into accessible, decision-support tools. Outputs include WebGIS-based visualizations, exposure models, and risk-relevant spatial products tailored for policy-makers, local authorities, and planners.
Goal: Ensure high-resolution scientific outputs are usable in real-world risk management and planning.
Supported by: WP3 & WP5.

O4 — Strengthen Cross-Border Collaboration Between Romania and the Republic of Moldova

This objective enhances scientific and operational cooperation through shared fieldwork, harmonized data protocols, joint analysis, and workshops. It builds a sustainable regional partnership in geohazard monitoring.
Goal: Improve landslide understanding and mitigation capacity on both sides of the border.
Supported by: WP4 & WP5.

Together, these objectives establish an integrated, AI-enabled, multi-sensor, cross-border system for advanced landslide detection, temporal monitoring, and actionable risk communication.