Potential way to target places most in need of identification

[nathantaylor]

I was looking through some of the plants of India trying to find Euphorbias this morning when I started to wonder if there would be a way to rank places by how badly they need identification of specific taxa. A way of doing this might be to select a general taxon (let’s say Plantae for example) and take the number of observations in that taxon to divide by. You could then take the number of research grade observations and divide that by the number of observations in the general taxon (or all observations if you’re a super generalist identifier). You could even get more specific and try to calculate what percentage of observations are IDed below family or another lower-level taxon. The reason this is important is that I would be lucky to get to family on the vast majority of the plants of India (for example), so if the majority of observations are already IDed to family, but aren’t research grade, it would be a waste of time for me to target the area. However, if the vast majority of plants are IDed at the Kingdom level, I could go through and at least get everything to Monocots or Dicots. Perhaps a pie chart could even be made to display the various taxonomic levels. Based on these numbers, one could rank the locations most in need of identification starting at continent or country level and working ones way down. Anyway, it was a thought, and hopefully a thought worth sharing.