granularity of indicators vs. topics

[abrin]

if NABONE has evidence at sheep at a site, and SEAD has evidence of pasture at that site, can we have an indicator of "Sheep pasture" or do we do this at the topic level.

[abrin]

this might be better done at the topic level, and then we need to look at what join / overlap looks like. e.g. spatial proximity and temporal overlap