[makerofspoons]

With a decent (perhaps high speed?) camera and a trained computer vision model, couldn't a contactless study be performed of flying insects? It should be possible to count the common species in an area with reasonable accuracy and flag photographs of unknown species for human review. Definitely not as cheap as a trap, but perhaps with solar power could be just as easy to place. Unless someone with experience with these studies can make the case that the number of insects killed by these traps is negligible and all this effort isn't worth it of course.

[veddox]

The problem is you simply cannot identify most insect species from photographs of them flying. Depending on which taxon you're working on, you have to do things like count the number of tarsi (foot segments) on each leg, look at the vein pattern on their wings, count the hairs on their thorax, or other minutiae. Also remember that most insects are only a few millimeters in size.

Some taxa can readily be identified in the field (butterflies, most grasshoppers, some beetles and true bugs). Others (dipterans and hymenopterans and pretty much anything smaller than 5mm) need to be done in the lab, with a good stereoscope and big identification books. It helps to have a steady hand and lots of experience...

(In fact, several taxa are so difficult to identify down to the species level that there might only be one or two experts in the whole country that can reliably do so.)