Using OpenStreetMap in Disaster Response

Using OpenStreetMap in disaster response: free map data

Using OpenStreetMap in disaster response means downloading the road and building data of a free map built together by volunteers and using it in analysis. However, quality differs from place to place, so you need to check it yourself before you use it.

In short

  • OSM is an open map that anyone can edit and use.
  • Get roads and buildings with Overpass turbo or QuickOSM.
  • Quality varies by region, so compare it with satellite imagery.

Suppose you are asked to support an area abroad hit by an earthquake or flood, and when you open the map, there are almost no roads or buildings. You cannot tell which roads relief supplies can take, or even how many houses are in a village. Volunteers around the world do fill in such blanks within a few days.

The map behind this is OpenStreetMap (OSM). It is also an option you can use for free in domestic work when you urgently need a basic road network or building locations. If you know what kind of data this map is, how to download it and how far to trust it, you will lose much less time in the field.

Black and white vintage photo camera and paper map composing with pen and protractor on table in bright room
Stock photo. Photo: Wallace Chuck / Pexels

What is OpenStreetMap, and who makes it?

OpenStreetMap (OSM) is an open world map that anyone can edit, download and use. It is easy to think of it as the Wikipedia of maps. Volunteers draw roads and buildings over satellite imagery and enter GPS traces and field survey notes. Some organizations and companies also contribute data. It is not a picture file but data with points, lines and areas that carry attribute tags (road type, whether something is a building, and so on), so you can use it directly in analysis. The license is ODbL, and you must show "© OpenStreetMap contributors" as the source. Derived data may carry share-alike conditions, so check before you distribute it. It is a starting point for departments without a commercial map contract, and for work that urgently needs a basic road network, such as disaster support abroad.

How do you download road and building data?

There are three ways, depending on the extent. For a small area, press the Export button at the top of openstreetmap.org, choose the area manually and press Export. A large area hits a limit. For a medium area, enter "building in [area name]" in the Wizard of Overpass turbo (overpass-turbo.eu), run it, and choose GeoJSON under Export. If the area is wider than a municipality, files for a country or region on the Geofabrik download site are stable. In QGIS, go to Vector > QuickOSM > QuickOSM..., enter a key (building) and an area, and a layer is created right away. Menu names can differ with the version and language settings, and if the menu is missing, first install QuickOSM from the plugin manager. If you overlay the downloaded building layer on an expected flood zone, you can count the buildings inside the hazard area right away.

How does the world draw a map together when a disaster happens?

HOT (the Humanitarian OpenStreetMap Team) is a nonprofit organization and community that works to make maps of disaster-hit and poor regions. When a disaster strikes, the map of the place that needs help is often empty. Volunteers then draw roads and buildings remotely while looking at publicly available satellite and aerial imagery. A tool called Tasking Manager is used to manage the work. It divides the target area into small cells so that several people take them without overlapping, and an experienced mapper reviews each finished cell once more. Speed varies a great deal with how quickly imagery is released and how cloudy it is. It is also important that local residents correct names and locations. Road networks gathered this way are used right away in relief route planning and international cooperation work.

Quality differs by region. How do you check it?

Because it is volunteer-based, completeness differs between cities where many people gather and remote areas. It is common to see missing buildings, wrong road types, old edit dates, and empty attributes such as number of floors and names. Here is what you learn only by trying. First, choose five or six different places in the target area, overlay them on satellite imagery and count the missing buildings by eye. Second, check that roads are not broken in the middle. A broken line throws off route analysis. Third, select an object and check the last edit date in its History. If you notice omissions, fix them, or cross-check against official data, before you use the data. When you estimate evacuees, missing buildings make the at-risk population look smaller than it really is.

What should you be careful about when using OSM in disaster work?

OSM is everyday location data, not a real-time status board. Flooded roads, closed sections and collapsed bridges are not reflected automatically. So evacuation routes that come from analysis are only candidates, and you must check them against on-site closure information. You also need care when you edit the public map yourself. As a rule, do not enter information that should not be public, such as facilities for older or vulnerable people or sensitive locations. Do not forget the license attribution either. The usage splits like this: in normal times, put the road network and building distribution down as the base background, and in a disaster, overlay official closure information on top. If you run through this flow once in a drill, you will not be flustered in the field.

Step by step on screen

Steps to follow

  1. Open the map
    In a web browser, go to openstreetmap.org, type [your area name] in the search box and zoom in on the map.
  2. Read the tags
    Click the Query features icon on the toolbar at the right, then click a building or a road to read its tags.
  3. Run Overpass
    Enter "building in [area name]" in the Wizard of overpass-turbo.eu and press the build and run button.
  4. Save the file
    Under Export > Data at the top, press the GeoJSON download and save the file.
  5. Load it in QGIS
    With Layer > Add Layer > Add Vector Layer, choose the downloaded file and press Add.
  6. Compare the quality
    Turn on QuickOSM or a satellite basemap and count the places where buildings are missing. If a button name differs, check the help (?) on the screen.

Menu names may differ slightly between versions. The figure above is a diagram of the sequence.

Prompts you can copy

Have it make a quality checklist for the OSM data of your area

You are a GIS practice trainer in the disaster safety field.
[Organization name] wants to use OpenStreetMap road and building data for [task, for example: evacuation route analysis].
The target area is [municipality name], and the tool to use is [QGIS or a web map].
Please make a checklist table for checking the quality of the OSM data in this area.
Include the items missing buildings, road connectivity, attribute completeness and last edit date, and for each item write one line each for "what to look at, how to check, and an example pass criterion."
Do not make up unconfirmed figures or agency criteria; where a criterion is needed, leave it blank as [value to set myself].
Explain technical terms in one line the first time they appear, so that a beginner can understand.
What a person must check: set the pass criterion values yourself to suit your organization, not by the AI, and test whether the checklist works by comparing against real satellite imagery.

Make a draft of the query to enter in Overpass turbo

You are a GIS helper familiar with OpenStreetMap and Overpass QL.
I want to download [the target to extract, for example: hospitals, schools, main roads] inside the administrative area of [area name] with Overpass turbo.
Write an Overpass QL query and explain what each line means with short comments.
Write the query as a single code block only, and limit the search extent to inside the administrative area of [area name].
Do not make a nationwide search, because too many results can put a load on the server.
Do not guess tag names you are not sure of; mark them "needs confirmation."
What a person must check: run the query in Overpass turbo, see whether the results shown on the map match the real targets, and check the tag names yourself on the OSM wiki (wiki.openstreetmap.org).

Replace the bracketed parts for your situation. Do not use the answer as it is; check the facts first.

Try it yourself

  • Open your neighborhood on openstreetmap.org, click a nearby school or shelter building and read how its tags are written.
  • Download the building data of your neighborhood from Overpass turbo, overlay it on a satellite basemap, and count missing buildings in at least five places.
  • Choose one building, open its History and check when it was last edited and how many people have edited it.

Common questions

Can I use OpenStreetMap data commercially or for work?
You can use it for work as long as you follow the ODbL license conditions. You must show "© OpenStreetMap contributors" as the source, and share-alike conditions can apply when you publish derived data. For the specific scope of use, check the official OSM copyright guidance yourself.

Is OSM data more accurate than other map data?
You cannot generalize about which is more accurate. Areas where many volunteers gather are detailed, but remote areas can be empty. The surest way is to compare the area you will use with satellite imagery or official data yourself.

Can ordinary people take part in drawing maps when a disaster happens?
Sometimes projects that accept volunteers open, such as HOT's Tasking Manager. Create an OSM account, read the guide and the practice course, and start with a small cell. How to take part and the procedure differ by project, so check each project's guide.

What comes next

OpenStreetMap is a map that anyone can make and anyone can use, so its power to fill in blanks is large, but errors can also get mixed in. That is why the habit of checking matters as much as knowing how to download. If you download the data for your own area and overlay it on satellite imagery, you get a sense of how far to trust it and where to add to it. So when a road network downloaded this way and official data disagree, which one should you take as the reference? A later post in this series will take this up.

This is an English adaptation of a post in the Korean GeoAI basics series. The original is here, and more learning materials are on GeoAI for Disaster Safety.

This post was drafted with the help of AI tools and reviewed by the author. Tool features and data portals change over time, so check the official documentation when you practice.

Comments

Popular posts from this blog

What Is GeoAI, and How Does It Help Disaster Safety?

Getting Started with QGIS: What to Do First

Whitebox QGIS Practice 26 — Elevation Statistics of Two Basins: Means of 2,000.1 m and 1,924.3 m, a 76 m Gap