摘要
全国第一次自然灾害综合风险普查是一项重大的国情国力调查,是提升自然灾害防治能力的基础性工作。普查数据质量是风险普查的“生命线”,直接决定普查工作的质量。全国气象灾害综合风险普查调查对象由传统的“数据表格”向“多维灾体”转变,数据质量控制不仅只是单点极值阈值的质控,还需考虑整个灾害事件致灾因子选取的准确性。针对数据本身属性以及站点空间比对属性异常和重复记录等问题,通过质检规则、管理约束、质量核查分析以及评估结果验证等,建立动态数据质控方法。该方法包括11个质控规则136个细化项,经过数据质控后,气象致灾因子的填报准确率明显提升,数据重复率显著下降,数据上报完成率和数据质量形式审核通过率均达100%,为评估区划和未来普查成果高效应用提供有力的基础数据支撑。
The first national natural disaster survey programme is a great effort of China to collect data regarding natural hazards,exposure,vulnerability,disaster reduction capacity of major disasters,including data for nine types of meteorological disasters.The forms of survey dataset in this programme range from tabular data,GIS data to unstructured data.Data quality is of great importance which directly influences the reliability of risk assessment results.Due to the large volume and complexity of the survey data,it is of great challenge to implement quality control,considering correctness,completeness,consistency,and standardization of data.Besides the capacity of checking ordinary static data,inspection rules shall also be able to check and ensure that extreme events are extracted properly or not with station dataset.We proposed a set of data quality control methods,including automatic quality inspection rules,management operation requirements,quality check analysis,and verification of assessment results.For inspections,we developed 11 quality control rules with 136 detailed items,which improved both the reliability and speed of data quality review.After the quality control,data revision and resubmission,the ratios of completeness and correctness for final data both achieved 100%.The data quality improved and assured through quality control process can provide a consolidate basis for further risk analysis in the future.
作者
陈东辉
熊安元
唐为安
CHEN Donghui;XIONG Anyuan;TANG Weian(National Meteorological Information Center,Beijing 100081,China;Anhui Climate Center,Hefei 230031,China)
出处
《灾害学》
CSCD
北大核心
2022年第4期135-142,共8页
Journal of Catastrophology
基金
国家重点研发计划项目(2019YFA0606904)。
关键词
气象灾害
风险普查
数据质量控制
数据审核
质检规则
危险性评估
meteorological disasters
risk survey
data quality control
quality inspection rule
hazard assessment
