Geospatial Clean Up Recovering Lost Data | Case Study | GEMS
Recovering lost positional data in your research
Geolocation is a pivotal component in agricultural research studies. Inaccuracies and missing data can weaken or even invalidate conclusions that are reached. Our geospatial cleaning tools facilitate the imputation of missing locational data, and correct those values that are inconsistent with the evidence.
Impact and partners
Through an innovative partnership with Macalester College, students of Assistant Professor Getiria Onsongo received scholarships to work on developing GEMS tools along side core GEMS developers. In this manner, the GEMS community benefits from the availability of these useful tools, and the students gain valuable experience working in a larger professional software development team, presenting their work, and having their work used by others.
Process and problem solved
Agriculture by its very nature is a location sensitive activity. Knowing precisely where an experimental trial is performed or a farm field is located is often the key to matching trial and farm production data with information on soils, climate, terrain and so forth. Unfortunately, much of the available experimental or farm data have geographic (i.e., latitude and longitude) coordinates that are incomplete, missing altogether, or represented by geographical names and not the analytically more useful lat-long coordinates. In some cases the reported names and coordinates do not jive (such as field trials located in the Pacific Ocean!).
GEMS Services Used to support Geospatial Clean Up
GEMS Platform
(annual subscription)
Farming is all about location, location, location
Virtually every public, private or non-governmental organization that has conducted field trials or agricultural production surveys has run into issues of accuracy and completeness with the geolocations in their records. In some cases, a field technician has entered incorrect lat-long values on a piece of paper or a spreadsheet. Perhaps (s)he transposed a digit, left out a digit, flipped the two coordinates, or omitted a minus sign. Or, maybe nothing was entered at all. Regardless, that data point is lost to subsequent meaningful research projects such as (i) those looking at the interplay of genetics with rainfall and solar radiation on crop growth, or (ii) tracking the effect of the soil characteristics, the slope of the land, the weather, and the growth of this plant on the leaching of nutrients in runoff to nearby rivers and lakes.
Fortunately, all is not lost. With GEMS Geospatial cleaning tools, available in the GEMS Platform, researchers can upload files and correct and impute missing geocoordinates with an easy-to-use interface. The algorithm is published in a peer-reviewed journal, bolstering confidence in the results obtained.
Results
- Missing lat-long coordinates are imputed using city or regional names.
- Incorrect geocoordinates are flagged and suggested corrections are made.
- Easy mapping visual interface allows quick verification of corrections.