Disaster monitoring and damage assessment using multi-sensor data

  DATE：2015-10-10    

The increasing availability of multi-sensor data and related spatial data provides abundant data sources for diverse applications. To meet vital scientific and economic demands, we have been working on extraction of disaster damage information, emergency response, drought monitoring, and achieved the following results.

(1)    Several effective methods of extracting damages to buildings and roads caused by earthquake disasters using multi-source spatial data have been presented, including an object-based hierarchical method for extracting building damage using multitemporal very high resolution (VHR) satellite data and LiDAR data, a method of extracting building damage using oblique airborne imagery, and a method of extracting urban road damage using VHR satellite data and road vector data. These proposed methods have been successfully used in many cases of damage information extraction and assessment in many earthquake disaster events, including Wenchuan Earthquake (2008), Zhouqu Earthquake (2010), Ludian Earthquake (2014), as well as Haiti Earthquake (2010) and Nepal Earthquake (2015).

(2)    The method and software system of snow disaster monitoring and assessment in Xinjiang were studied and developed based on the integrative use of optical and passive microwave remote sensor data in the retrieval of snow covers parameters. The system was successfully used in the monitoring of the cold outbreak and snow disaster in north Xinjiang in April, 2014, and a timely research report was proposed to the governmental department for aiding disaster rescue decision making.

(3)    The compatible system based on the Compass and GPS navigation technologies, and low-altitude unmanned aerial vehicle (UAV) remote sensing system were developed and used in the real-time acquisition of in-situ information of emergency events. An emergency rescue and response system was further established, which was successfully applied in the emergency relief of the earthquake disasters in Ludian (China) and Nepal.

(4)    The endmember variability in the arid areas was deliberately considered and incorporated into the spectral mixture analysis and modeling. The co-inversion soil moisture datasets were assimilated into the distributed hydrological model, to build the data assimilation approach. These strategies much improved the retrieval of landcover parameters such as soil moisture and vegetation coverage. An assemblage of remote sensing based monitoring technologies for ecological and environmental status in the arid areas was established, and the corresponding software application was also developed to implement these proposed algorithms and models. The software system was successfully utilized in Xinjiang, China, and the achievement won the first prize of Provincial Award for Science and Technology Progress of Xinjiang Production and Construction Corps, China, and specially reported by Guangming Daily (section six) on January 8, 2015.

(5)    Taking Ningxia arid area as example, key techniques for monitoring crop growth and assessing farmland drought using combined ground observations, meteorological data and multi-sensor data (ENVISAT-ASAR and EOS-MODIS) were explored, and a system for remote sensing monitoring and rapid assessment of farmland drought was established and has been successfully used in monitoring, assessment and drought defense in the Ningxia area.

We obtained two China invention patents, 15 Software Copyrights. We also won one first prize for scientific and technological progress from Ministry of education (2014), one first prize for scientific and technological progress from the Xinjiang Production and Construction Corps (2014), one second prize for scientific and technological progress in mapping and surveying (2008), and one first prize for scientific and technological progress in mapping and surveying (2014).