⇓ More from ICTworks

We Need to Improve GIS Boundary Data Accuracy for Better Development Decisions

By Guest Writer on October 12, 2017

tanzania administrative boundry map

Administrative areas are key to how development data, and development itself, is organized, yet in many countries, simply getting lists of the current administrative areas – provinces, regions, districts, counties, townships, wards or villages- is a problem, especially for lower-level areas.

Even when lists are available, it is not easy for stakeholders to get the boundary data (shape files) needed to map and display the data in Geographic Information Systems (GIS).  In many countries, the administrative areas change frequently – areas are split up or merged, or moved from one “parent” area to another.

Furthermore, occasionally the administrative area structure is different, or is coded differently, across different sectors in the same country. This makes comparing and combining data from different systems and sources very difficult and impedes efforts to display trends over time.

Even when we have maps and data sets which contain geo-coordinates or “points” (e.g. health facilities, schools or water points), it can be difficult to maintain up-to-date links between these “points” and the changing administrative areas they fall under.

One-off is Easy – Ongoing Accuracy is the Challenge

In Tanzania, the Government and stakeholders, with support from PATH, are beginning work to create a publicly available “administrative area registry”.

If successful, this registry would facilitate use of a range of boundary data sets by the government, NGOs and development practitioners, who will be able to display the data in easy-to-understand maps by administrative area.

We realize that the key challenge will not be producing the list, or making it available on a once-off-basis. That has been done before.

Rather, the challenge is maintaining it over time, and ensuring that governance processes are in place to update the registry whenever changes are made, to correct errors as they are identified and to evolve the accuracy of shape files over time.

We Need Your Help to Improve Boundary Data

Administrative area data issues are broader than any one organization, country, or sector and we would like to hear a variety of perspectives:

  • In what ways have you experienced these issues in your work?
  • How have you mitigated, or overcome these challenges?
  • What are some best practices you employ?
  • What initiatives are trying to solve these problems?
  • Who are the key institutions that are innovating practical solutions?
  • What useful open source software platforms could help solve these problems?
  • Which software and GIS groups are already working on these problems?

Please help kick start our boundary data discussion in the comments section of this post. We aim to develop a discussion in collaboration with the Digital Impact Alliance (DIAL) on this topic. If you would like to join our conversation, join our geo areas discussion group.

Together, we can find sustainable solutions to gather, maintain, and share crucial administrative area data for development.

By Elaine Baker, PATH Senior Program Officer – Digital Health

Filed Under: Data
More About: , , , ,

Written by
This Guest Post is an ICTworks community knowledge-sharing effort. We actively solicit original content and search for and re-publish quality ICT-related posts we find online. Please suggest a post (even your own) to add to our collective insight.
Stay Current with ICTworksGet Regular Updates via Email

3 Comments to “We Need to Improve GIS Boundary Data Accuracy for Better Development Decisions”

  1. Reid Porter says:

    Really interesting work, and the administrative area registry idea strikes me as both fairly complicated to pull off (basically a data standard for geography, something others have called for for years) and immensely valuable to all kinds of people in all kinds of sectors if it is pulled off.

    Forgive the dumb question, but is that not essentially the mission behind GADM (http://www.gadm.org/)? As a former GIS guy I’m quite familiar with the limitations, especially in areas that experience frequent and rapid boundary changes, but how would this work extend rather than duplicate that long-standing effort?


  2. Elaine Baker says:

    Thanks for kicking off the discussion Reid ! Yes efforts such as http://www.gadm.org and work by UN agencies ( https://sites.google.com/site/commonoperationaldataset/home and https://unsalb.dfs.un.org/ ) are important international compilers and publishers of this data from multiple countries. However these international data compilers currently struggle to get accurate timely granular updates from many of the countries they gather this data from. The Administrative Area Registry is intended to facilitate a country in keeping its admin area data up to date, by allowing the Government to codify the update process and assign update roles and responsibilities to different officials and institutions within the country (for example different local authorities, Bureau of Statistics, Ministry of Lands, eGovernment Agency, sector Ministries etc).

  3. Mikel Maron says:

    Very cool to see PATH taking this up. At Mapbox, we’ve been working with PATH in the VisualizeNoMalaria initiative, and a similar issue came up for Zambia. I am a long time participant in OpenStreetMap and one of the founders of Humanitarian OpenStreetMap Team — and accessing and managing boundaries is a perennial problem.

    There are a couple issues with GADM. One is that updates are closely managed by a small group, which limits timeliness. The other is the non-commercial use license — which is well meaning, but does prevent many broadly valuable usages, including incorporation into OpenStreetMap.

    OpenStreetMap itself is the home for many boundaries, but it is not necessarily an ideal platform for managing data sets that overlap at multiple scales. Still, it’s model of cooperation and open license makes it a useful starting point for considering how to build a multi-stakeholder collaborative effort around boundaries.

    Tanzania is a really interesting country to start with, given the number of efforts at data collaboration focused there now.

    Would love to discuss more Elaine.