Remote GIS Volunteer Work
DAN SPECHT, retired gis practitioner
I recently retired, and wanted to work as a volunteer so I could use my GIS skills to do something useful, and because I really like doing GIS. I joined a group that helps people contribute to relief efforts using GIS from their home computers, and it has been a great match for my time and interests.
I registered with OpenStreetMap (www.openstreetmap.org), which is like Google Maps--a simple web map interface to a global geospatial database covering the earth. Also it’s like Wikipedia in that it can be edited by anyone. The data can be licensed for use at no cost, which some commercial users, such as Apple, have done. The data is donated by governments (for example the US Census donated TIGER data), NGOs, and corporations (Automotive Navigation Data donated road data covering the Netherlands, China and India). In addition, OSM volunteers go to various places where they work with local volunteers. The locals walk, bike, or drive around with GPS units and make notes on what they see. Volunteers create feature geometry from the GPS tracks and attributes from the notes.
Figure 1 is an OSM map showing OSM users in my community. Some are regular contributors, others are geography students who added a feature to OSM for a homework assignment. About 600,000 people have contributed content to OSM, and this number is growing rapidly.
San Francisco Bay Area roads are pretty much mapped in OSM, but some parts of the world are barely mapped at all. My guess is that the most underserved 5% to 10% of the world’s population lives in these unmapped areas--places with weak governments and bad roads.
When disaster strikes and relief organizations go in, they need maps for planning. Often they need maps showing each footpath and hut because huts are the only indication of population distribution and footpaths are the only mode of transportation. Humanitarian OpenStreetMap Team (HOT) (http://hotosm.org/) provides maps to these groups, and they crowdsource the mapping effort.
Figure 2 shows a screenshot of a group of huts tagged as a residential area.
HOT supports relief efforts following disasters such as the recent earthquakes in Nepal, a cyclone in Vanuatu, ebola in West Africa and war in Central Africa. I’ve been to none of these places, but working from home I’ve learned something about these parts of the world. Other HOT volunteers go to the sites, where they work with local volunteers. (There’s an article describing how local volunteers on the ground are working with HOT athttps://opensource.com/life/15/5/nepal-earthquake-hfoss).
Imagery providers such as Bing loan imagery for these efforts. Remote mappers use this imagery to create whatever features the relief agencies request. Initially these tend to be roads and urban areas, and later building footprints, schools, and flat open areas for landing helicopters on. Figure 3 shows screenshots of the Java editor (JOSM) and the HTML editor (iD). The editors include pre-defined feature types and lots of editing tools.
If you are interested in helping, register with OSM and go to the HOT tasking manager (http://tasks.hotosm.org/), which lists dozens of projects in order of priority. Each task is tiled. You pick a project, check out a tile and start editing. You can do as much or as little work as you want. You can also review completed tiles and mark them approved or not. You learn a lot faster from reviewing than by editing, because you see what features others are mapping.
Figure 4 shows a screenshot of the tasking manager, which is open source OSM software.
One downside to this approach is that different mappers and different reviewers have different skill levels and also that they interpret the instructions differently. For example, some projects require that areas with buildings be classified as residential, but there is no precise definition of “residential.” Beginning editors make lots of mistakes. The customer needs to look at the map as it’s being made, provide feedback, and if necessary, change the instructions.
However it’s hard not to be impressed when you see an area 100 miles across going from being mapped at the level of one or two major highways to being mapped down to the level of footpaths and small groups of huts.
Summer 2015 Volume 8 Issue 1