Gehirn and ESRI Japan's Challenge to Display Real-Time Weather Information on a Map to Support Disaster Prevention and Business Continuity Planning.

ESRI Japan
Katsuhiko Akimoto, Osamu Sugawara, Nobuyuki Tanaka

Everything that exists in the real world has geographical information associated with it. When various natural disasters strike, geographical information is indispensable in understanding the actual damage and working towards recovery.

ESRI Japan, a leading provider of GIS (Geographic Information Systems), has partnered with Gehirn to offer NADIAct, a cloud GIS and real-time disaster prevention weather information service provided by Gehirn through ArcGIS. “ESRI Japan Data Contents Online Suite Weather Online Service (Gehirn Edition)”, hereinafter referred to as the “Online Weather Service (Gehirn Edition)”, has been released. We asked about the background of this initiative to support disaster prevention and mitigation by combining real-time disaster prevention weather information with geographic information.

Graphical display of “Geographic Information” associated with all kinds of events to support decision making

Q: Please tell us about the company ESRI Japan.

Akimoto: The U.S.-based ESRI was founded in 1969 and released the world’s first commercial GIS software in 1981. The company has held a global share of the GIS market ever since. The Japanese subsidiary was established in 2002 and provides products and services to government agencies, municipalities, and businesses, and has been installed in approximately 25,000 organisations. In addition, ESRI has distributors not only in Japan but also in Asia, Europe and other countries in order to adapt to the circumstances of each country, while collecting any useful and relevant geographic information in/about each country.

ESRI Japan Platform Solutions and Content Division, General Manager, Katsuhiko Akimoto

ESRI Japan Platform Solutions and Content Division, General Manager, Katsuhiko Akimoto

Q: What kind of system is GIS?

Akimoto: Everything in the world, such as cities, buildings, electricity and gas infrastructure, traffic lights, etc., all have location information associated with it. GIS is a system that represents various types of information in the real world in the form of layers on a map.

For example, if we overlay layers such as “population distribution” based on census data, “hazard maps showing the risk of flood damage” published by the Ministry of Land, Infrastructure, Transport and Tourism, etc., and “road networks” on top of each other on a map, the formulation of optimal evacuation routes in the event of flood damage can be realised. In this way, GIS facilitates the analysis of something or produces better results by displaying real-world events with “geographic information” as the key.

Q: What is the difference between GIS and a typical “Maps” application?

Akimoto: I see general map applications such as Google Maps as a way of displaying content such as “here’s a good restaurant” on a map. GIS manages such information but can be viewed in different ways depending on the purpose and can be analysed and visualised however the client needs it.

Q: In what fields is GIS being used?

Akimoto: We have a wide range of customers in public offices and utilities, such as electricity, gas, and water, and we’ve also received interest from the insurance and distribution industries. Retailers are also using it to develop strategies for opening new stores and as a material for area marketing. One unique example is the linkage with the “Monster Strike” game, which utilises location information.

Q: Is it also used in the field of disaster prevention?

Akimoto: Disaster prevention can be divided into various aspects, such as initial response, prevention and preparedness. GIS is also used for “hazard maps”, which you may be familiar with. It’s also used in the formulation of Business Continuity Planning (BCP) for companies, in issuing evacuation orders from local governments, and in the issuance of disaster certificates after a disaster occurs, just to name a few.

Based on location information, GIS can piece together “how many people are in this area, how flooded is the area, and what is the condition of the road network?” This is perhaps the greatest strength of GIS.

Q: At the entrance to your office, there was a row of “thank you” letters from local governments from all over the country…

Sugawara: Starting with the Niigata Chuetsu Earthquake (2004), we’ve been supporting systems for issuing disaster victim certificates in the aftermath of the Great East Japan Earthquake (2011) and the torrential rains in Western Japan. Only on a map can we link three pieces of information: the person who suffered damage, the house they own, and the extent of the damage. It’s difficult to know “who lives where” just by looking at a ledger, but if you put it all on a map and overlay it with information like “which areas are flooded by more than 2 metres”, you can easily see it.

ESRI Japan Data Solutions Group, General Manager and Technician, Osamu Sugawara

ESRI Japan Data Solutions Group, General Manager and Technician, Osamu Sugawara

Q: Do you provide assistance in areas after a disaster?

Tanaka: When GIS operators are needed for restoration, it’s sometimes necessary to visit the site ourselves. In one instance, I worked side-by-side with Self-Defense Forces personnel who were dispatched to a disaster site. Quick decisions are required at the site of a disaster, and I think that in such cases, people are reminded of the importance of maps.

ESRI Japan Platform Product Group, GIS Product Specialist, Nobuyuki Tanaka

ESRI Japan Platform Product Group, GIS Product Specialist, Nobuyuki Tanaka

Akimoto: After the Great East Japan Earthquake, the “Emergency Mapping Team” was formed under the leadership of the Cabinet Office. ESRI and some professors from the Disaster Prevention Research Institute of Kyoto University formed a team to create various maps. We visualised a wide range of information, such as “how much of the water supply was damaged”, “where are cultural assets and how were they damaged”, “where are evacuation centres located and how many people can they accommodate”, “where are stockpiles of supplies and how much is stored there”, etc. We also created a map of the areas that were affected by the disaster, and we used this map as a guide for which people needed help.

A very large area was affected by the Great East Japan Earthquake, and one of the means of giving accurate and efficient instructions was with a map. Maps were used to show people with decision-making authority the full situation, and to determine the destination of relief supplies.

Tanaka: I remember extracting and inputting data from Excel and PDF files sent by fax from various municipal governments in different formats, and working tirelessly to make the data visible on a map in a centralised, bird’s-eye view kind of way.

Q: It sounds like you’re working on a tally of the recent number of COVID-19 infections?

Akimoto: Incidentally, this system is also used in our COVID-19 dashboard, which uses data published by Johns Hopkins University (USA)

“Deliver more value from a BCP point of view” through Gehirn

Q: GIS forms a close relationship with disaster management and prevention. It felt like this was also the background of the collaboration with Gehirn. How did you find out about Gehirn?

Akimoto: I heard from the vice-president of a beverage company that there was a company called Gehirn, which has a very good disaster prevention app. I decided to contact them immediately and found out that another person in the company was in contact with them, so I was introduced to them and met Mr Ishimori and his team.

Q: Did you have the idea of creating a service like NADIAct or the Online Weather Service (Gehirn Edition) from then on?

Akimoto: No, I don’t think we were thinking of anything that concrete at the time. ESRI Japan has a team that think up “tasks” for new projects, and one of them, the BCP task, came up with the idea of using map data for companies’ BCP activities. As a result, we had already created a system to map the seismic intensity and liquefaction information in the event of a Nankai Trough earthquake, or an earthquake directly under the capital, as was issued by the Cabinet Office, and to distribute this information via the cloud.

When we spoke to Gehirn, this project had come to a close, and we were discussing what to do next. We had the idea that since real-time weather information is now available, we could provide some value there, but that would require live data. In talking with Gehirn, we agreed that they could provide us with exactly that kind of information.

Q: Did the conversation take shape from there?

Akimoto: Yes. The NADIAct project was born out of the decisions we made while receiving sample data, such as how to implement the system, the format of the data and where to put it, and it took shape with help from Tokio Marine. Additionally, as each piece of data provided for NADIAct can be used as a layer, we also released it in a form that can be displayed in ArcGIS, which forms part of the Online Weather Service (Gehirn Edition).

Q: On the Gehirn side, when you first heard about it, did you think “I can do it”?

Ishimori: Yes. Since they said they wanted them as layers of GIS, we thought we could do that by distributing what was already available as data for our disaster prevention app.

Gehirn Inc., Daiki Ishimori

Gehirn Inc., Daiki Ishimori

Akimoto: When we spoke to Gehirn, we found that their data was in a format that was easy to incorporate into GIS, and had already been coordinated, so we were able to make adjustments without hesitation. The fact that Gehirn had knowledge of GIS in the first place was a big factor - we were able to agree on the GIS coordinate system and how it was to be attached, which made for smooth communication.

Tanaka: I thought, “How complicated can it be?” and when I got hold of the data for the first time, all I found was JSON and GeoTIFF, and I thought, “This’ll be a piece of cake.” Gehirn doing the conversion from the original data was very helpful.

Accurate information distribution, supported by Gehirn’s know-how on the background of weather information, and even the law

Ishimori: Although they use Gehirn as the data provider, if they wanted to, I think ESRI Japan could get the data directly from the JMA and process it for GIS themselves…

Akimoto: As I said, when it comes to weather-related know-how, Gehirn is a cut above the rest. It was very reassuring to receive data from someone with whom we could consult the Meteorological Services Act and the background to the development of the data, such as “Why does the JMA distribute data in this way?”

It would probably have been tough if ESRI Japan tried to do the same thing alone. Even if we could have purchased information from the Weather Service Support Centre, it would have been difficult to provide the service in such a short time with our in-house knowledge, because we would have had to think about the rules for how to distribute the information.

Ishimori: Yeah. For example, with regard to earthquakes, the JMA data has separate telegrams for seismic intensity bulletins and epicentre bulletins, so the receiving side needs to overlap them and combine them into one. Also, if a tsunami advisory (or higher) is announced, the epicentre data is included in the tsunami information without an epicentre bulletin, so the data has to be taken from multiple places. If we forget to take care of these details, data can be lost, and NERV Disaster Prevention takes great care in this process so that data that is updated in pieces is always presented in a single piece. It’s kind of an obsession.

Sugawara: When I started talking about this service with Gehirn, I was surprised to find out how many different types of information are handled just for “rain”. For example, how many layers of information correspond to the category of “rain”?

Nukaya: By “rain”, do you mean “analytical rainfall”?

Sugawara: This is exactly what I mean - When you say “rain” like this, it’s not just the amount of rainfall you’re talking about.

Ishimori: In addition to real-time weather observations (AMeDAS), there is also the precipitation radar forecast and a soil rainfall index. The keyword “rain” alone covers such a wide range, including heavy rainfall warnings, advisories and record-breaking short-period heavy rainfall information types. And since landslide warnings, floods and inundation damage are also calculated from rainfall models; they can also be considered part of the rain category…

Sugawara: Which rainfall information the map user wants to use depends on the application. If we are talking about sloped surfaces, they would need soil rainfall indices rather than warning information. I think the ability to carefully provide information down to each of these specialist layers is a major value that comes from working with Gehirn on this service.

For example, some users have their own models in the field of disaster prevention, such as “calculating the degree of danger based on the cumulative rainfall over a 24-hour period”. The fact that we are now able to provide suitable data when we receive a pinpoint request for this kind of data is very significant.

Q: In which areas are you starting to see the Online Weather Service (Gehirn Edition) and NADIAct be used?

Akimoto: A motorway management company has started to use analytical rainfall from the Online Weather Service (Gehirn Edition) to foresee motorway slopes collapsing. A manufacturing company is using NADIAct to maintain and manage its procurement network and supply chain, plotting information of 200 or 300 suppliers on a map and overlaying it with risk data to determine “which rivers are likely to overflow and which sites are likely to be flooded in the event of a typhoon”. The system can then send out alerts.

In the future, we are considering not only using the weather information received from Gehirn on its own but also combining it with, for example, road traffic congestion information and social networking information to further increase its value. When disasters such as typhoons or major earthquakes occur, traffic volumes also change. We are considering the possibility of using information such as where and how traffic jams are occurring and, conversely, where traffic volumes have dropped drastically for use in the transport industry.

The project is the fruit of both parties’ desire to make things “more useful to the world, more quickly”

Sugawara: Another thing I really liked about being part of this project is that they organise weather information that is difficult to use and put it out there in a very particular way in order to be useful to the world. If we asked other companies to do the same thing, a lot of people would say, “Well, in that case, it would probably take about a month…” but Gehirn is not like that. They say, “We'll look into it,” and a week later, they say “It's ready”. I think NADIAct is the result of people with passion and technical skills, who are willing to frankly exchange their opinions without ties or discovery, and with the single-minded desire to create something useful for the world.

Nukaya: Gehirn’s strength is that almost all of us are engineers. I believe this also contributes to our speed. What we deal with in this service is disaster information. As Mr Ishimori often says, we want to implement it faster and to provide it to as many users as possible, so that when a disaster happens while we are taking our time, we won’t have to regret that we should have released the information sooner.

Gehirn Inc., Takashi Nukaya

Gehirn Inc., Takashi Nukaya

Q: What has been the response to the service, and what future developments do you have in mind for the distribution of disaster management information?

Tanaka: We want to increase the information we can provide to meet the diverse needs of our customers. At the same time, I think we need to make a proper appeal for how effective this kind of information is.

Akimoto: Although we are a private company, we have always had a passion for disaster prevention and social contribution, as evidenced by our free Disaster Response Programme for municipalities and companies affected by disasters. We hope to continue to cooperate with Gehirn, which is in a similar position to us.

Contact Us

If you have any questions about Gehirn's vulnerability analysis or disaster prevention information distribution system, feel free to contact us via the form below. One of our representatives will respond as soon as they can.