[Paper Review] Floods impact dynamics quantified from big data sources

David Pastor-Escuredo, Yolanda Torres+2 more|arXiv (Cornell University)|Apr 24, 2018

Human Mobility and Location-Based Analysis13 references3 citations

TL;DR

This paper proposes a multi-source big data framework integrating environmental, social media, remote sensing, digital topography, and mobile phone data to quantify flood impact dynamics across multiple granularities. By dynamically requesting data based on indicators, it enables real-time humanitarian response and recovery planning while addressing privacy and data heterogeneity challenges, demonstrated through three diverse flood case studies with actionable insights for resilience building.

ABSTRACT

Natural disasters affect hundreds of millions of people worldwide every year. Early warning, humanitarian response and recovery mechanisms can be improved by using big data sources. Measuring the different dimensions of the impact of natural disasters is critical for designing policies and building up resilience. Detailed quantification of the movement and behaviours of affected populations requires the use of high granularity data that entails privacy risks. Leveraging all this data is costly and has to be done ensuring privacy and security of large amounts of data. Proxies based on social media and data aggregates would streamline this process by providing evidences and narrowing requirements. We propose a framework that integrates environmental data, social media, remote sensing, digital topography and mobile phone data to understand different types of floods and how data can provide insights useful for managing humanitarian action and recovery plans. Thus, data is dynamically requested upon data-based indicators forming a multi-granularity and multi-access data pipeline. We present a composed study of three cases to show potential variability in the natures of floodings,as well as the impact and applicability of data sources. Critical heterogeneity of the available data in the different cases has to be addressed in order to design systematic approaches based on data. The proposed framework establishes the foundation to relate the physical and socio-economical impacts of floods.

Motivation & Objective

• To develop a systematic, privacy-aware approach for quantifying flood impacts using diverse big data sources.
• To address the challenge of data heterogeneity and privacy risks in large-scale disaster impact assessment.
• To enable dynamic, multi-granularity data pipelines for real-time humanitarian action and recovery planning.
• To demonstrate the applicability of data-driven insights across varied flood event types and geographies.

Proposed method

• The framework integrates environmental data, social media, remote sensing, digital topography, and mobile phone data into a multi-access, multi-granularity data pipeline.
• Data is dynamically requested based on data-based indicators, enabling adaptive data acquisition during flood events.
• A composite data pipeline processes heterogeneous sources, ensuring privacy and security through proxy-based aggregation.
• The approach uses proxies from social media and aggregated mobility data to reduce privacy risks while maintaining analytical utility.
• Case studies are used to validate the framework across different flood types and data availability levels.
• The system is designed to relate physical flood characteristics with socio-economic impacts through integrated data analysis.

Experimental results

Research questions

• RQ1 How can big data sources be systematically integrated to quantify flood impact dynamics across multiple spatial and temporal granularities?
• RQ2 What role do social media and aggregated mobile phone data play in reducing privacy risks while maintaining impact assessment accuracy?
• RQ3 How does data heterogeneity across different flood events affect the reliability and scalability of big data-based impact assessment frameworks?
• RQ4 To what extent can dynamic data request mechanisms improve the timeliness and relevance of humanitarian response?
• RQ5 How can the framework relate physical flood characteristics to socio-economic impacts for evidence-based policy and resilience planning?

Key findings

• The framework successfully integrates diverse data sources—environmental, social media, remote sensing, topographic, and mobile phone data—into a dynamic, multi-granularity pipeline for flood impact analysis.
• Proxies derived from social media and aggregated mobile data effectively reduce privacy risks while enabling actionable insights into population movement and behavior during floods.
• Case studies revealed significant variability in flood characteristics and data availability, highlighting the need for adaptive, context-sensitive data integration strategies.
• The dynamic data request mechanism based on indicators improved data relevance and reduced unnecessary data collection, enhancing efficiency.
• The framework demonstrated practical applicability in three distinct flood events, showing consistent potential for humanitarian response and recovery planning.
• The integration of physical and socio-economic impact data enables more informed policy design and resilience-building strategies.