Disaster Preparation

Blog Co-Author: Samantha Fox, Senior Consultant, IBM Global Business Services

The first edition of the blog discussed currently observable climate change effects, and how they are beginning to impact all facets of human life. This installment will focus on methods of preparation for those effects.

Costs associated with natural disasters on the rise. Since 1980 there have been 254 weather and climate disasters exceeding $1.7 trillion in remediation in the United States. In the last decade, the frequency of disaster events and their costs are on the rise, and unfortunately the U.S. is one of the most natural disaster prone countries in the world. This is large in part due to its geographic size, diversity of its landscape, and varied weather and environmental patterns. And according to NASA, this exposure to natural disasters also makes each U.S region vulnerable to the future effects of climate change in a myriad of ways:

Northeast. Heat waves, heavy downpours and sea level rise pose growing challenges to infrastructure, agriculture, fisheries and ecosystems.

Northwest. Sea level rise, erosion, inundation, risks to infrastructure and increasing ocean acidity pose major threats, alongside land focused challenges of increasing wildfires, insect outbreaks and tree diseases causing widespread tree die-off.

Southeast. Extreme heat will affect health, energy, agriculture; and decreased water availability will have economic and environmental impacts.

Midwest. Extreme heat, heavy downpours and flooding will affect infrastructure, health, agriculture, forestry, transportation, air and water quality, and more.

Southwest. Increased heat, declining water supplies, reduced agricultural yields, health impacts in cities due to heat, and flooding and erosion in coastal areas are additional concerns.

The Earth has already warmed 1° Celsius since the Industrial Revolution, and the major effects of climate change, like the ones mentioned above, are estimated by the Intergovernmental Panel on Climate Change to occur beginning at an increase of global temperature of 1.5° Celsius. According to the Yale School of Forestry & Environmental Studies,“ by some estimates, curbing warming at 1.5 degrees could be sufficient to prevent an ice-free Arctic in summer, to save the Amazon rainforest, and to prevent the Siberian tundra from melting and releasing planet-warming methane from its frozen depths.” However according to the Intergovernmental Panel on Climate Change and Climate Scientists “limiting warming to 1.5° Celsius above pre-industrial levels would require transformative systematic change” from countries, regions, and businesses that are underway, but not on track to reach the 1.5° Celsius increase limit stated in preventative initiatives like the 2015 Paris Agreement.

Scientists agree that a 2° Celsius increase is the cap to prevent “dangerous climate change,” yet the United Nations says that we are on track for 1.4 - 4.3° Celsius of warming by the end of the century. With the current trajectory, it is clear we will see the effects of climate change within our and/or our children’s lifetimes. And while there is much debate about how to prevent global warming, logic shows clear that we need data to better predict needed responses and improve recovery times from natural disasters.

Digitalization of data, data-sharing via the cloud, and management and use of big data through analytics, including augmented intelligence have played a critical role in “making sense” of critical aspects within the vast amount of information collected on natural disasters. Currently analysts gather and analyze the data they receive and produce reports.  However, the addition of cognitive technologies like text analytics, descriptive and predictive analytics with machine learning, visualization, and data pedigree tracking can augment these reports.  This would allow for analysts to brief decision makers on proactive solutions for situational awareness and augment intelligence instead of producing reactive reports. The ability to be proactive in emergency response scenarios will become increasingly important as disasters continue to worsen.

An example of how big data and analytics assisted with a disaster relief effort is the Rebuild Texas program.  Blockchain technology was used to process, resource, and track payments associated with disaster relief assistance when Hurricane Harvey struck.  Blockchain technology allowed for the disaster relief assistance process to be streamlined, where survivors only had to submit one application for the whole ecosystem. Stakeholders had visibility into aid distribution across the entire process, data was exchanged quickly, and disputes could be submitted directly to FEMA. The technology simplified an extremely complicated form management system plaguing the disaster aid response process.  

Disasters create big data problems, and the ability to integrate structured and unstructured data from multiple sources in real time or near-real time is essential to create a common operating picture and visualize situational awareness. To address this challenge, enhanced methods for utilizing advanced analytics that consume all data from authorized sources and applies machine learning were created. The machine learning allows the analyst to enhance reports with automated documenting sources, methods, entity net maps, patterns of life, and applicable context for their management that may assist in warning of emerging conditions, trends, and threats from multiple large data sets. This results in more accurate common operating pictures, which allows for better planning and response, thus creating a new capability - cognitive command and control (C2). Weather, utility, and infrastructure data is also integrated to create new situational awareness for Emergency Operating Centers (EOCs).

A demonstration of these C2 capabilities are at Texas A&M where they are developing and delivering a system for monitoring the conditions, impacts, and emergency operations associated with extreme weather events. Statistical data from infrastructure location data, power grid asset failure and associated weather condition histories are collected and integrated.  The result is a light, clickable, prototype tool that can be used in emergency operations centers. By including some of the architecture from cognitive C2, it is possible to create real-time situational awareness and apply predictive analysis. By enhancing emergency operation centers in this way, the ability to respond and recover from disasters will improve exponentially. 

As climate change continues, natural disasters will occur more frequently, and disaster management will grow in complexity and remain a Big Data problem requiring a public private partnership (P3) solution. Technology will connect organizations with capabilities and revolutionize disaster management.