An approach to constructing a system for processing open earth remote sensing data for flood monitoring using big data technologies

Every year up to 70 major floods occur in the territory of the Russian Federation. Tens of thousands of communities with the total population of up to 5 million may find themselves within a disaster area. To avoid human casualties and economic losses, early threat detection is needed in order to implement preventive measures. Earth Remote Sensing (ERS) data from space are one of the major sources of reliable and timely information about floods. For overall assessment of the flood situation, ERS data of various resolutions and scales (territorial coverage) are needed. At present, users can access an extensive list of open ERS data sources, including those that allow direct reception of data at the user's own hardware and software receiving systems. The authors propose a methodology for constructing a flood monitoring ERS data processing system, which is based on open ERS data and open-source software packages, including those for big data processing. This system makes it possible to build an ERS data processing pipeline from data reception to the thematic end product.