I have a specific use-case that I cannot seem to find any solutions for online, so any and all advice would be most welcome. Some basic details:

I am part of a small academic organization (a research lab), and our research leans heavily on GIS. We are all field scientists who specialize in our particular discipline, NOT computer scientists, so while we are all competent at the analytical side of desktop GIS, as soon as we get into the weeds of database programs, web maps, online servers, etc. we are lost. Because we are a bunch of broke academics, we mostly use QGIS (currently 2.18 but switching to 3.x as soon as there's an LTR) and I'd like to keep it that way.

Our research requires us to be offline (or with only tenuous and expensive internet access) for MONTHS at a time. Most GIS multi-user solutions seem to presume that people are coming back to the office and syncing their data every night. For us, this is not realistic.

As such, currently I spend about a week after every field campaign manually integrating 3 or 4 updated/edited versions of the same shapefile to reflect everyone's changes. We also have an absurd proliferation of shapefiles that I need to get under control.

It looks like Spatialite and GeoPackage both allow multiple geometries to live in the same file/database, so that is very appealing for dealing with the second problem (shapefile proliferation). However, how do these file formats handle data syncronization?

Is there a way for us to work with a single, simple database of features (e.g. Spatialite) while in our home GIS lab, then take 3 or 4 local copies of that database to our field sites and make a few hundred edits per user, and then sync all of those edited versions back together at the end of the field season?

  • 1
    Do the spatial extents of the field sites overlap? That is, can each researcher work with an extracted portion of the main database/shapefile?
    – mkennedy
    Aug 23, 2018 at 19:48
  • They don't overlap. That sounds like a good idea--would it present the same issue with synchronization that we currently have?
    – LAT
    Aug 23, 2018 at 19:57
  • QField? qfield.org/docs/index.html
    – Mapperz
    Aug 23, 2018 at 20:42
  • 1
    You would still have to merge the data back in somehow, but at least you would know that changes can be in the researcher's extracted area only.
    – mkennedy
    Aug 23, 2018 at 21:20
  • 1
    break your data into chunks per researcher and use a db to extract and load deltas will keep it in check.
    – Mapperz
    Aug 23, 2018 at 21:44

2 Answers 2


For anybody who happens to find this post, I want to mention the solution we settled on.

As @mkennedy suggested, before going into the field we split our main feature layer (a shapefile, for now) into various ROIs, and give each member of the team the ROI where they will be working. Each person preps their local data the way they want it on their own computer, and then migrates it to QField using the QFieldSync plugin for QGIS.

All offline data collection is done in QField, and synced with the investigator's computer via USB every evening.

At the end of our field season, each of the updated ROI-specific subsets of the main file is re-integrated to produce a new version of the master file.

So far this has worked quite well. Thanks to those who posted here for the suggestions!


A approach that is available now that was not available at the time of this question is the Mergin plugin https://public.cloudmergin.com/#. The plugin uses GeoDiff to manage edits by simulations users of a geopackage. https://github.com/lutraconsulting/geodiff These users can either be accessing the geopackage through QGIS or the Input app.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.