So in chatting with some of the folks at PHAC (@NHizon @janet.lin ), they highlighted that there is a feature that would be helpful to them given an issue they’re having with sites. I’ll try to summarize the issue here, but if there’s any nuance I’m missing, I invite them to jump in and correct me.
Essentially, when sampling and taking measures from samples, it’s useful to know what geography that sample/measure is intended to reflect. This is why we have the polygons table, which allows users to specify the exact geography and catchment boundaries for a site. For situations in which a polygon shape file is not available, how can we record some of that polygon catchment information in the absence of the polygon? Take for example WWTP X, which is the wastewater treatment plant that services REGION A: a region in which there are three municipalities; and WWTP Y, which is the wastewater treatment plant that services REGION B: a region in which there are four municipalities.
As shown in the diagram above, it is possible to sample WWTP X such that you have measures that reflect each of the municipalities individually, or the whole region. Conversely, WWTP Y can only be sampled and measured in a way that reflects the whole region.
When it comes to analysis and aggregation, it sounds like it would be useful to users to be able to differentiate between the geographic level that is being sampled (whether that’s a region, or something smaller) so that these comparison can be more meaningful. Because comparing the measures from Town A2 and Region B isn’t quite right, but we can’t get any more granular data from Region B at the municipal level, for example.
To address this, I’ve suggested a new field in the sites table: siteLevel
siteLevel will be an optional, categorical field that lets users specify the level at which sampling/measures can be interpreted. The reason for the inclusion of both A Regions and B Region serves to address edge-cases in large metropolitan areas, where the whole region may be made up of many smaller regions, made up of multiple municipalities.
Curious to hear any additional feedback, and to see if anyone has any concerns with this proposed addition.