Recent Progress in and Future Perspectives on the Monitoring, Assessment, and Mitigation of Geological Disasters

1. Introduction

Among all the geological disasters (GDs), collapses, landslides, and debris flows are the most severe. Predicting geological disasters is one of the most effective means of reducing casualties and property losses. However, technologies for the early identification and warning of collapses have been difficult to implement effectively. Therefore, the scope of this research topic includes the quantitative identification of dangerous rocks, risk assessment of landslide instability precursors, quantitative modeling of movable solid sources in debris flow source areas, application studies on new monitoring and early warning technology systems, and the corresponding control and reinforcement measures.

This Special Issue brings together 25 cutting-edge research articles addressing critical challenges in geotechnical engineering, landslide dynamics, rock mechanics, and disaster resilience. As global climate change and urbanization intensify geohazard risks, from landslides and slope failures to seismic disasters, this collection offers innovative methodologies and practical insights to enhance prediction, monitoring, and mitigation strategies. The contributions span theoretical advancements, experimental validations, and real-world applications, reflecting a multidisciplinary effort to safeguard infrastructure, ecosystems, and communities. Taken together, the 25 articles in this Special Issue address these critical challenges, offering innovative methodologies and practical insights to advance prediction accuracy, monitoring capabilities, and mitigation effectiveness. Key themes and breakthroughs are discussed in the subsequent sections to provide a comprehensive understanding of the topic.

2. Slope Stability and Landslide Mechanisms

Several studies focus on novel approaches for evaluating slope stability. Jia et al. [1] propose using natural vibration frequency to rapidly assess perilous rock stability on soil slopes, linking buried depth to safety factors. Cao et al. [2] integrate UAV mapping, blasting tests, and numerical simulations to analyze multi-factor influences on open-pit mine slopes, revealing instability risks in excavation designs. Moresi et al. [3] compare GIS-based models (4SLIDE vs. SHALSTAB) for shallow landslide prediction, demonstrating 4SLIDE’s superior accuracy in Mediterranean terrains. Komu et al. [4] model landslide runout distances in Türkiye’s flysch formations, incorporating climate scenarios (RCP 4.5/8.5) to forecast future hazard zones.

Collectively, these advances signify a transformative shift in slope stability research, moving from reactive post-failure analysis to proactive approaches leveraging real-time monitoring (e.g., vibration sensors and UAVs) and predictive modeling (e.g., climate-integrated runout simulations) [1]. This shift is characterized by the integration of multi-factor data (geotechnical, geophysical, and operational), which is essential for capturing complex instability mechanisms in engineered slopes [2], and crucially, by embedding climate uncertainty (e.g., RCP scenarios) into assessments, a necessity for infrastructure resilience under intensifying extreme weather [4].

Coupling AI (e.g., real-time sensor analytics), cross-scale modeling (micro-fracture to regional slope systems), and social vulnerability metrics will transform slope stability analysis from a technical exercise into a pillar of sustainable development.

3. Material Behavior Under Environmental Stress

Understanding soil and rock responses to environmental triggers is critical. Chen et al. [5] employ high-precision CT scanning to quantify pore evolution in tuff under frost salt weathering, revealing magnesium sulfate’s severe erosive impact. Zhu et al. [6] develop a coupled damage model for deep expansive soil under freeze–thaw cycles, validated via triaxial tests. Guo et al. [7] review the geomechanical complexities of “block-in-matrix” rocks (Bimrocks), emphasizing the block proportion and welding degree as key failure controls. Valentino’s [8] microstructural analysis of lateritic soils in Rwanda explains landslide triggers through mineralogical alterations unseen in conventional tests.

These studies illuminate three transformative shifts in geomechanics research:

1.

From Macro to Micro/Nano Scales:

Advanced tools like high-resolution CT [5] and mineralogical mapping [8] expose hidden mechanisms, such as magnesium sulfate exhibiting a pore expansion rate approximately two times greater than that of sodium sulfate in tuff that dictates the macroscopic failure. On a different scale, in granular systems, arching structures that develop during excavation constitute a critical failure mode under stress redistribution. These have been effectively characterized through the optical and mechanical modeling of force chains [9]. This scale-bridging is redefining failure prediction;

2.

From Single-Stress to Multi-Field Coupling:

Zhu et al.’s [6] freeze–thaw-loading damage model exemplifies a critical advance, quantifying how concurrent environmental and mechanical stresses (e.g., cyclic freezing and structural loads) accelerate material degradation, essential for infrastructure in climate-volatile zones [10];

3.

From Homogeneous Assumptions to Heterogeneity Embracing:

Guo et al.’s [7] Bimrock analysis proves that block–matrix interfaces control stability. Similarly, Valentino’s [8] discovery of 7.71% kaolinite porosity gaps in lateritic soils reveals why “identical” strata fail differentially, demanding stochastic modeling over simplistic averages.

The frontier demands:

• AI-accelerated multi-scale modeling merging quantum-scale chemistry with continuum mechanics;
• Field-deployable micro-sensors to capture real-time mineralogical changes;
• Global degradation atlases mapping material vulnerabilities under projected climate scenarios [11].

Ultimately, decoding environmental stress responses transforms geomaterials from passive hazards into intelligently engineered assets—where every mineral alteration becomes a calculable risk variable.

4. Advanced Monitoring and Early Warning Systems

Innovative technologies enable real-time hazard detection. Chen et al. [12] introduce MEMS sensors with dual indicators (natural frequency and RMS velocity) to warn of rock collapses 3× faster than tilt-based systems. Chen et al. [13] utilize time-series InSAR (PS/SBAS) to monitor subsidence along China’s Yangtze River, identifying urbanization and underground engineering as key drivers. Damiano et al. [14] characterize pyroclastic soils for landslide early warning, emphasizing liquefaction susceptibility’s role in post-failure evolution. He et al. [15] validate an aerial remote sensing platform with real-time satellite transmission for disaster response, achieving <1 cm resolution imagery with <2 s latency.

Contemporary advanced geohazard monitoring is undergoing a paradigm shift, transitioning from post-event diagnosis to real-time early warning systems, as demonstrated by MEMS sensors enabling second-level rock collapse alerts [12] and satellite-linked drones transmitting centimeter-resolution imagery within 2 s [15]. This evolution integrates multi-scale observation networks, synergizing InSAR-based subsidence tracking [13], UAV remote sensing, and soil behavior characterization [14] to bridge microscale mechanisms with macroscale hazard evolution. Intelligent algorithms now supersede empirical thresholds, reducing false alarms by approximately 50% through dynamic indicators like coupled frequency–velocity indices, significantly enhancing warning precision [12]. These advancements deliver critical societal benefits: namely, ≤15 min warnings could reduce landslide mortality by 90% (UNDRR) and InSAR-based urban monitoring plays a critical role in mitigating considerable economic and infrastructural losses, while low-latency satellite transmission enables broader access for high-risk regions [16]. Future progress necessitates miniaturized sensors, AI-driven big data analytics, and standardized protocols to realize a planetary-scale hazard immune system—transforming disaster resilience into an ethical imperative through ubiquitous, intelligent safeguarding.

5. Earthquake and Flood Impact Assessment

Rapid post-disaster modeling saves lives. Miura et al. [17] create a near-real-time model for estimating seismic intensities and building losses in Bogotá, generating impact maps within 5 min. Annunziato et al. [18] reconstruct Libya’s catastrophic Derna dam break using satellite/social media data, highlighting the critical 15–30 min evacuation window. Zhang et al. [19] apply Neural Radiance Fields (NeRFs) to rapidly generate 3D landslide models, reducing reconstruction time by 30% compared to photogrammetry for emergency response.

Advanced near-real-time modeling, demonstrated by seismic loss mapping within 5 min [17] and dam break flood reconstruction leveraging satellite/social media fusion [18], has significantly shortened critical assessment timelines, enabling proactive evacuation and resource allocation before secondary disasters escalate. This acceleration is underpinned by multi-source data integration (e.g., InSAR, crowdsourcing, and NeRF-generated 3D terrain), which deciphers cascading events like earthquake-triggered landslides [19], climate-induced mega-slides [20], or subsequent flooding through unified impact narratives. Crucially, physics-based scenario engineering now supports the preemptive stress testing of infrastructure resilience against compound hazards, such as post-quake dam breaches or sediment-driven floods, validated by mortality reduction rates where sub-30 min warnings reduce fatalities by 8% per minute in multi-hazard scenarios (World Bank). To fully transform impact assessment into proactive risk interception, future work must prioritize cross-hazard algorithms that unify seismic, hydrologic, and geotechnical predictors to anticipate and mitigate domino effect disasters.

6. Risk Mapping and Infrastructure Resilience

Spatial and statistical tools enhance risk management. Lu et al. [21] combine the Analytic Hierarchy Process–Information Content Model (AHP-ICM) and entropy methods for collapse risk zoning in Huinan County, achieving 87.4% prediction accuracy (AUC). Nie et al. [22] couple information quantity with random forest to assess highway collapse risks, showing a reduction in high-risk areas when including emergency response capacity. Rouhana et al. [23] use network theory to evaluate Beirut’s transportation fragility, revealing low redundancy and high disruption sensitivity.

Advanced geological disaster mapping has evolved from static zoning to dynamic intelligence, exemplified by machine learning-enhanced models achieving 87.4% prediction accuracy [21], field-tested DMT-based methods for sliding surface localization [24], and random forest-driven assessments that reduce high-risk areas by 4.66% through integrated emergency capacity metrics [22]. This paradigm shift quantifies systemic vulnerabilities, such as infrastructure redundancy deficits that paralyze 40% of urban emergency routes [23], highlighting critical interdependencies between human systems and infrastructure, particularly in regions where high landslide hazard aligns with limited resilience (UNDRR, 2023). The integration of social vulnerability indices ensures climate justice, directing resilience investments toward marginalized communities, while high-precision zoning helps to optimize global mitigation funding by significantly reducing false alarms (World Bank). Future frontiers demand digital twin cities for adaptive evacuation planning, AI-powered regulatory frameworks to enforce building compliance in high-risk zones, and Global South-centered protocols addressing informal settlement realities.

7. Synthesis and Future Directions

Collectively, the contributions in this Special Issue highlight a transformative shift toward integrated, data-driven approaches in geohazard mitigation. From the fusion of field data with AI models (e.g., Mercurio’s [25] MARS model for earthquake-induced landslides) to the scaling of micro-scale structural insights to macro-scale slope stability assessments [8], and the integration of emerging technologies such as Neural Radiance Fields (NeRFs) with conventional monitoring frameworks [19], these studies present actionable and forward-looking frameworks. Yet, challenges remain in translating laboratory findings to field environments, capturing nonlinear site responses [17], and adapting predictive models to evolving climatic and anthropogenic pressures. Addressing these translational challenges remains critical. Recent experimental research has proposed high vibration amplitude and uncoordinated dynamic responses as key indicators for early warning of long-runout landslides [26]. Nonetheless, solutions such as MEMS-based early warning systems and real-time aerial monitoring platforms [27] are bridging the gap between theoretical advances and engineering practice.

The innovations presented in this Special Issue are particularly vital in mountainous regions experiencing accelerated infrastructure development. In the context of China’s major national strategies—such as the Belt and Road Initiative—projects like the Sichuan–Tibet Railway, the China–Myanmar oil and gas pipeline, and the Wudongde Hydropower Station are advancing across geologically complex alpine canyon zones. The increasing prevalence of high and steep slope constructions has heightened both the frequency and intensity of rockfall hazards. However, continued reliance on passive mitigation—due to the lack of effective early warning technologies—often results in considerable economic losses and ecological impacts.

Against this backdrop, advancing robust, field-ready monitoring and early warning technologies is not only a scientific imperative but also a national strategic necessity. The convergence of cutting-edge geomechanical theories, AI-enhanced predictive modeling, real-time micro-sensing, and low-latency satellite–UAV platforms marks a paradigm shift in our capacity to detect, analyze, and mitigate geological risks. Looking ahead, realizing the full potential of these innovations—particularly in terrain-sensitive, infrastructure-intensive settings—will require sustained technological breakthroughs, cross-disciplinary collaboration, and equitable international knowledge exchange. Only through such collective efforts can we strengthen global resilience, safeguard critical infrastructure, and protect vulnerable communities in the face of escalating geohazard challenges.