Skip to content

Author

M. Sathianarayanan

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Open access Jul 2026

Climate Transition Zones as Emerging Hotspots for Natural Hazards: Insights from Land Use- Climate Feedbacks Amplify Disaster Risk in Taiwan

Abstract. Human-driven land use change and global climate variability are jointly reshaping regional environmental processes, yet their combined influence on climate classification and hazard dynamics remains underexplored. This study integrates satellite-derived land use/land cover (LULC) data, Köppen–Geiger climate classifications derived from CHIRPS precipitation and MODIS land surface temperature, and a disaster inventory (2001–2020) to examine land–climate–hazard interactions in Taiwan. Results indicate notable LULC transitions, including a 0.98% increase in woody savannas and a 0.72% decline in wetlands, coinciding with a significant expansion of tropical climate zones (Aw +14,172 km²; Af +1,038 km²) and a contraction of temperate regions (−15,365 km²). Approximately 68% of observed LULC changes spatially correspond with shifting climate zones, suggesting strong land–climate coupling. Transition zones, covering only 15% of Taiwan’s land area, exhibit disaster frequencies 2.8 times higher than stable regions. Warming transition areas account for over half of typhoon-induced landslides and a substantial proportion of major flood events in recent years. Localized warming exceeding 1°C is closely linked to albedo reduction from deforestation, while intensified precipitation extremes correlate with agricultural expansion (r = 0.73). These findings highlight climate transition zones as emerging hazard hotspots and underscore the need for dynamic, climate-informed land-use planning strategies.

M. Sathianarayanan, Pai-Hui Hsu · 0 citations
Open access Jul 2026

Street-Level Disaster Location Detection Using Image Matching of Social Media Images

Abstract. Rapid identification of disaster locations is essential for effective emergency response and situational awareness. However, a large proportion of images shared on social media during disasters lack geographic metadata, limiting their usefulness for operational decision-making. Existing approaches mainly rely on geotags or textual geoparsing, which often fail when metadata is missing or ambiguous. As a result, valuable visual information from social media images remains underutilized for disaster mapping. This study proposes a deep learning–based framework to estimate the geographic location of disaster images using visual scene matching. The approach compares query images from social media with georeferenced Google Street View imagery to infer their potential locations. The framework integrates image pre-processing, deep feature extraction, and similarity-based matching to identify the most likely geographic correspondence. Preliminary experiments demonstrate promising results in detecting key scene elements within disaster imagery, providing a foundation for reliable visual matching. By leveraging visual cues rather than textual metadata, the proposed framework aims to improve the usability of non-geotagged social media images for disaster response. The proposed approach has the potential to support rapid disaster mapping by transforming citizen-generated imagery into spatially actionable information for emergency management.

Pai-Hui Hsu, M. Sathianarayanan · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.