Background

A humanitarian mapping project is an exercise where you need to collect a great a mount of data in the least possible time and it needs to be practical, precise and easy to replicate. In a crisis situation the top priority is to guarantee the victims access to shelter, food, basic sanitation services, social support and many other services. Providing information to make decisions is of vital importance. Those decisions need to be taken by analyzing all the necessities versus the available resources.

It is in this process that an application like Mapillary can support OpenStreetMap's tasks in the collection data about a geographical area and the visual aspect of the terrain.

It has been demonstrated that geographical information is of vital importance in managing a humanitarian crisis situation. The earthquake in Haiti, the storm Katrina or the typhoon in the Philippines are just a few examples where humanitarian mapping activities have been activated.

In Colombia, a team of volunteers from the OpenStreetMap community has been working for years to train a group of mappers to help in crisis situations and retrieve geographical data for the decision making.

Mapping #MapatónxGuajira

The Guajira region is experiencing a drought crisis and the OpenStreetMap Colombia community is gathering information in the Riohacha and Nazareth region. The mission is to recollect geographical data that can provide enough information to make decisions around basic sanitation and water for the community.

We decided to use three different tools for data collection; satellite images, street photos with Mapillary and images from a drone flight.

With the support of OCHA Colombia and #BrigadaDigital we started an initial phase of mapping remotely with satellite images. The satellite information was provided by the Missing Map project and Secretary of La Guajira provided information regarding natural wells and Jagüeyes (ponds). Most of the mapping was done in the Ministry of Communications building with more than 60 volunteers. We currently have active HOT tasks from Riohacha.

As a second phase the Colombian team mapping collected on the ground data using Mapillary to collect images. In addition to collecting images with Mapillary, we also did a drone flight with a video camera over the areas with flooding risk in the Riohacha, the drones where provided by EcoExplora and other institutions.

With students from UNIGUAJIRA Maicao in the Computer Science fields and other non tech students we organized three workshops. One was a field trip to collect information in an urban area of Maicao. The commissioned mappers did two trips. The first one to was to Carraipia to map aqueduct and the other was a farm called “El Limón” on the outskirts of the city.

The Result

On the workshops in Maicao in the UNIGUAJIRA we collected traces and photographies in three blocks of the neighborhood in Barrio Buenos with Mapillary. We used FieldPapers to gather cadastral information of the neighborhood.

Both applications where stable on the ground. Though, we had to adjust the methodology since a couple of traces was done parallel to the direction of the street and we lost information on the numeric description of the buildings. Our suggestion was to do it in perpendicular mode so you could get the information of the buildings and also have enough view of the street. Some shots where captured in manual mode.

On the field trip to Carraipia we used Mapillary with the Mapillary app and a smartphone. We places the phone in the front window with a car mount and we captured the whole tip from Maicao to Bocatoma. In this case the app performed without any problems. You can see the whole trip on this trace of Carraipia.

The trip to the farm of “El Limón” was done on a motor bike, all the images where captured with an action camera that was fixed on the helmets of one of the mappers. The geotagging of the photographs is done through synchronization of the photos and the gpx traces of the GPS. We use a plugin called Photo Geotagging. The field trip's goal was to discover magueys (ponds) and understand the topology of the terrain.

In the use case of data collection, all three applications did their job but we found that the bottle neck is the uploading of the photos due to the lack of broadband connections in these areas.

Further, it is important to consider powerbanks when mapping crisis situations with Mapillary since you are far away from a power outlet. Also, it is good to bring extra memory SD cards since you will be collecting a lot of images and photo storage will be crucial to get all the desired information.

We would like to thank Claudio Cossio and the rest of the Mapillary team for the support with equipment, accessories and resources to use it on the #MapatónxGuajira initiative and we hope that they continue their support on future activities.

A big thank you to Juan Carlos Pachón for allowing us to share his story.