Компании Spectee и Synspective объединили данные спутников SAR и соцсетей для точного картирования паводков

Spectee, Synspective test Kumamoto flood mapping with SAR and social media KEY POINTS - Spectee and Synspective develop flood-mapping method combining SAR satellite data with social media-derived inundation depth information - Joint verification uses August 2025 heavy rain data from Kumamoto's Amakusa region - Estimated flood area qualitatively matches a Kumamoto prefectural survey of actual inundation zones TOKYO, Aug 12 - Spectee and Synspective have developed a method to estimate flood inundation areas more precisely by combining synthetic aperture radar, or SAR, satellite data with information derived from social media posts, the companies said on August 12. The method integrates wide-area inundation zones extracted from SAR imagery with location-specific flood depth data identified from social media posts. The companies said joint verification using data from past heavy rain events confirmed the method's effectiveness. The work is part of a collaboration announced in July 2025 to speed up and improve disaster response. Spectee provides AI-based real-time disaster prevention and crisis management information services, while Synspective develops and operates small SAR satellites and offers related solutions. According to the release, the approach links the two different datasets through geospatial information including elevation and land-use data. Synspective automatically interprets SAR data from frequent satellite observations to extract area-wide inundation ranges, while Spectee uses AI to identify flooded locations and inundation depth from images, videos and text posted on social media. The combined analysis is intended to address the limits of each dataset when used alone. Synspective's StriX small SAR satellites can observe wide surface areas regardless of weather conditions or time of day, but satellite-only analysis can struggle to identify localized flooding in dense urban districts. Spectee's system is suited to pinpointing on-the-ground flooding conditions in real time, but coverage is more limited in areas with few social media posts. The companies conducted joint verification using SAR satellite data and social media post data collected during heavy rain in the Amakusa region of Kumamoto Prefecture in August 2025. In areas covered by both methods, the SAR-based and social media-based analyses produced consistent results. The companies also said social media data helped supplement inundation mapping in urban areas where flood presence is harder to determine from SAR interpretation alone, while SAR-derived flood boundaries and terrain characteristics such as river basins and gradients helped estimate broader river flooding that is harder to capture through social media information alone. The inundation area estimated under the new method was qualitatively consistent with actual flood zones identified in a survey by the civil engineering department of the Kumamoto Prefecture Amakusa Regional Promotion Bureau. Kamiamakusa city has separately published an overview of flood damage related to the August 2025 heavy rain. SAR is a remote-sensing technology that can observe the Earth's surface at night and through cloud cover, making it widely used in disaster monitoring when optical satellite imagery is constrained by weather. In Japan, local governments and central agencies use hazard and damage assessments to support evacuation, rescue and recovery operations after floods and landslides. © Copyright 2026 IBTimes JP. All rights reserved.