Study On Risk Early Warning Of Geo-hazards In Yunnan Province Based On Big Data

Geological disasters often occur where there are the development of geological structure、complex and diverse terrain and strong human activities,the multiplicity of disasters seriously affects the safety of people’s life and property.In Honghe prefecture and Nujiang prefecture,there are Similar geological environment characteristics — high and steep mountain and ntensive structure activities,in recent years,beause frequent geological disasters threaten the population and social stability in the area,Honghe prefecture and Nujiang prefecture are listed as the demonstration areas for Monitoring and early warning of geological disasters.Acorrding to Job requirements of Geological disaster warning and Emergency decision support in the demonstration area,this is a problem how to synthesize spatial data and attribute data of geological disaster in the area and establish a perfect warning system for geological disasters with refering to the relevant technology and thinking of big data,While ensuring the data processing and analysis efficiency of geological disasters.The government needs to solve the urgent problem during advancing the early warning of geological disasters in Honghe prefecture and Nujiang prefecture.In the paper,around two core requirements of geological disaster warning and early warning efficiency,we have explored a set of feasible Display approach for geological hazard data and system of Theory and method for geological hazard assessment with using big data technology,at the same time,we have developed a geological hazard warning system used actually in the actual geological environment of two demonstration areas.These results provide a direct basis for geological disaster prevention and relief on one area,enhance coping capacity,and reduce disaster losses to a large extent.The main work of this paper is as follows:(1)Based on analysis of the data characteristics of geological disasters in Yunnan province,the geological disasters data warehouse is designed,large data cluster multidimensional analysis platform based on hadoop+Kylin is constructionthed,Cube data for two topics of threat of geological disasters and geological disaster monitoring is analyzed quickly online,the results can provide support for geological disaster risk warning.(2)Combining the characteristics of the geological environment in the two research areas,the laws of the distribution and development of geological disasters are summarized,and on this basis,geological environment influencing factors are selected,and the susceptibility of geological disaster is systematically studied by using the hierarchical analysis-information method process and information method.The results show that the hierarchical analysis-information method is good and simple.The influence of geological environment on geological disasters is reflected objectively with no distractions of too many subjective factors.(3)The characteristics of rainfall induced geological hazards in different areas are identified through the statistical analysis of historical disaster data and rainfall induced geological hazard data and considering the correlation between rainfall and geological disasters.Base on the the characteristics,rainfall induced model is further constructed in two demonstration zone.The risk of future rainfall to the research area is determined by using the model.(4)The warning units of 100 m×100m were used in two monitoring demonstration zones.based on information method,we have established geological hazard risk warning models of Geological hazard factor and rainfall induced factor by using the methods of generalized linear operation and probability operation.After Historical disaster verification,the evaluation accuracy of model is above 75%,the range of forecast accuracy can be improved to village level.The results show that the model is highly available.(5)Considering the impact of geological disasters,On the basis of the hazard warning of geological hazards,the study on the risk of geological disaster risk is extended,focusing on the social attribute of geological disasters.By selecting four indicators including land type、artificial facilities、population and GDP,this paper analyzes the differences in the resilience of different affected bodies to geological disasters and comprehensively evaluates the vulnerability of the study area.In the end,the disaster risk warning model of geological disaster is constructed in combination with the vulnerability of the disaster body;the theory and method of geological hazard assessment are enriched.(6)Acording to the characteristics of data-intensive and computation intensive for geological disaster and distributed task scheduling of MapReduce,we has found a Parallel processing method to analyzing geological hazard risk in Cluster parallel computing environment.The methord has broke through core and key technologies about task decomposition,parallel scheduling,result fusion during analysising Vector space data from the network space data source and realized parallel acceleration of geological hazard risk analysis.