During disasters like Cyclone Idai in 2019, the 2018 Ebola outbreak, or the 2015 Nepal earthquake, improving and updating OpenStreetMap data has proven to be an effective way to help response efforts. Disaster Ninja is a tool for the OpenStreetMap community that makes it easier to prepare and prioritize mapping tasks and save precious time during a crisis.

The team at Kontur built Disaster Ninja based on their experience as a technical solutions provider for disaster management organizations. An active contributor to the OpenStreetMap community, Kontur partnered with the Humanitarian OpenStreetMap Team (HOT) to create a tool to support the rapid deployment of emergency mapping campaigns.

Rapid intelligence

The Kontur team studied HOT’s activation process during a 2019 earthquake in South America, identifying components that could be assisted with automation. Today, HOT Response Coordinators use Disaster Ninja to prepare Disaster Size-Ups documents and mapping tasks in minutes instead of hours, thanks to having critical information all in one place. 

“Disaster Ninja is cutting the size-up time in half. Now it’s possible within an hour to have a good feeling if there is a reason to activate or not. I truly appreciate having very helpful layers into one package because it’s definitely saving time.” - Russell Deffner, HOT Response Coordinator

Ready-to-use data layers show how well population density corresponds to mapped infrastructure to help prioritize populated areas that need the most attention. Information about tasks from HOT’s Tasking Manager helps response teams decide whether to start a new mapping task or use existing tasks. Kontur also set up an automated alert system that connects the Global Disaster Alerting Coordination System to HOT's Slack. With one click, a coordinator can start analyzing an event in the Disaster Ninja interface.

Supporting OpenStreetMap communities

Disaster Ninja is also useful for the wider OpenStreetMap community. Data layers help mappers to determine how recent OpenStreetMap data is in a given area and identify areas where there is a mismatch between population density and expected feature density. 

To celebrate its 16th anniversary, the Ukrainian OpenStreetMap community organized a mapping party to improve OpenStreetMap coverage in the country. Using Disaster Ninja to spot under-mapped areas, they identified that the city of Polohy was practically absent on the map and organized mapping tasks to add it.

The global OpenStreetMap community is large and welcomes both local and remote mappers, but local communities differ in how they communicate and coordinate. Disaster Ninja suggests local mapping communities and ways to connect with them. It also includes an ‘Active Contributors’ layer that identifies the OpenStreetMap username of the most active map editors in an area.

Investing in intuitive analysis

A tour of the analytical layers available in Disaster Ninja illustrates the depth of analysis and design that the Kontur team has invested into this social good tool.

OpenStreetMap ‘Quantity’: Shows the relative distribution of OpenStreetMap objects and Population density, visualized with a bivariate choropleth (inspired by cartographer Joshua Stevens) to visually explore the relationship between the two indicators and draw attention to populated places with lower-than-expected OSM coverage.

Building Quantity and Road Length: These layers analyze the extent to which all populated houses are mapped and all populated places have roads connecting them. Buildings and roads are critical data during disaster response to identify where people may need help and to navigate to locations, so these are typically the priority mapping tasks during HOT activations.

OpenStreetMap ‘Activity’: Measures the level of mapping activity in a given area over the last two years, based on edit history and mapping hours by active local mappers. A ‘mapping hour’ is counted as an hour during which a user uploaded at least one tagged object. Mappers are considered active if they contributed more than 30 mapping hours during the last two years. Whether or not a mapper is considered ‘local’ is based on where they have mapped most actively.

OpenStreetMap ‘Antiquity’: Shows how old the OpenStreetMap data is in a given region and how many times users have viewed OpenStreetMap in that region over the last 30 days. This layer is useful for understanding where people are looking at the map and if the map data there might be due for updates. Identifying the least edited but the most popular areas can help to prioritize update efforts.

Urban Core and Settled Periphery: Mapping in densely developed areas requires smaller task extents and more experienced mappers, while less densely populated areas can have larger task extents and are easier for beginner mappers. The Urban Core layer highlights the most densely populated areas affected by a disaster and the Settled Periphery layer shows more spread-out population areas. These shapes can be downloaded as GeoJSON files for further analysis or to outline mapping tasks in HOT Tasking Manager.

Population: The creation and maintenance of a publicly available global population dataset is a fundamental component of Disaster Ninja. This data is continually updated by processing publicly available data, including: Facebook’s High Resolution Settlement Data; Global Human Settlement Layer; Microsoft Buildings Footprints; and Copernicus Global Land Cover. The latest version of Kontur Population is available at data.humdata.org.

Disaster Ninja is driven by OpenStreetMap community needs, so if you have ideas for how it can improve or be more useful for particular use cases, connect with the Kontur team - they even have a chatbox directly within Disaster Ninja interface.

Explore Disaster Ninja to inspire your next - or your first! - OpenStreetMap contributions. And if you’re building similar tools for disaster response and positive impact, connect with the Mapbox Community team.

‍