A storm hits. Trees down power lines, water levels rise. Emergency teams scramble to figure out the damage, who’s vulnerable? First responders think visually so answering these questions often starts with a map. And those maps are frequently powered by Esri, a leader in Geographic Information Systems (GIS). A couple weeks ago I visited their Redlands, CA headquarters on what just happened to be the day after Hurricane Sandy hit the Eastern seaboard. Historically GIS solutions have been delivered as complicated desktop apps, which required trained GIS specialists to use. But for about a year now Esri has included a cloud-based component to the system called ArcGIS Online. It lets users, even laymen like you and me, create and instantly publish and share interactive maps. During the storm they experienced an intense spike in demand. The system scaled 3x in one day going from 50 million maps to 150 million just after the peak of the storm, all hosted on Windows Azure. To learn more about how cloud is affecting their industry, I chatted with Russ Johnson, Esri’s Director of Public Safety and National Security Solutions, as well as Paul Ross, Product Manager for ArcGIS Online.
Before joining Esri, Russ used to be a Commander in National Incident Response teams. He described how knowing what was in place before the disaster is so critical, especially when leading a diverse team that spans local, state and federal organizations. This information would typically be printed and distributed to emergency workers. One of the biggest challenges was to find the data, because local government is often in disarray due to the disaster. For Sept 11th it took at least a week to find the data, bring it together, normalize it, create a database, then produce maps. Once maps were produced people came out of the woodwork with asks. “What bridges can carry big loads? What government buildings are vacant, damaged, etc…”
After initial assignments are made, you typically plan and refresh every twelve hours. That could require four to five hours in a helicopter surveying the incident. Then a couple hours with the planning team to generate and print new assignments for the response teams.
Enter the cloud which is beginning to fundamentally change a thirty year process in a couple of key ways. First there’s access to unbelievable amounts of quality base data already online. Then there’s ease of use – the new app allows people to quickly create maps almost anyone can use, no specialists required. And instead of printed maps, or at best converted to static PDFs and then posted online, now data and intelligence is coming in real time, maps are dynamically updating - you can get the current status of roads, gas stations, shelters. These maps can be almost instantaneously accurate and easily shared broadly, across multiple agencies. Resource assignments can be made digitally and sent directly to emergency responders. The public can access the same maps. The result is more informed decisions, better outcomes, and greater continuity of governance.
Another new capability enabled by cloud is dynamic mash ups. For example take a map of flood zones, then bring in real time stream gauges to show current water levels. Or take a map of shelters and layer on the open commercial stores nearby. One map generates another. Laymen users combine static and dynamic data by themselves. The possibilities are endless. For example… here’s a relevant question many people had - how will Superstorm Sandy affect voter turnout in the 2012 US presidential election? This map took precinct-level historical voting data and overlaid FEMA impact zones for the disaster.
Darker shaded counties were most damaged by the storm. By mashing up storm and voting data, one could assess impact of the storm on expected voter turnout by political party.
We’re on the cusp of something even more powerful - crowd sourcing for damage assessment. Leading up to the storm, on the spur of the moment Esri created a cloud-based mobile solution allowing individuals to report and upload conditions on the ground into a central database. With the wide and growing availability of smart phones, this raises the notion of every member of the public as a possible sensor for damage assessment. Powerful stuff.
Even though we’re not quite there yet, the cloud has already been a game changer in this industry. Paul Ross explained that for the first time with Hurricane Sandy there was no hesitation to use the cloud for mission critical information. Utility companies posted outage maps – identifying where power was out and when you can expect it to come back. State and local governments posted evacuation maps and impact areas.
What I find inspiring is that Esri is not just doing the same things in the cloud as they did on premises. They are doing things that are only possible in the cloud. It shows the power of software to help solve real human problems and the power of the cloud to deliver step-change improvements in how it’s done.
About Esri and the Microsoft Disaster Response Program
Esri and Microsoft partner to provide public and private agencies and communities information maps during disasters in a cloud computing infrastructure. Microsoft directs citizens to the maps via online platforms such as MSN, Bing, and Microsoft.com.