A small subset of observations of animals–not including hair, feathers, or scat–involve animals which have been dead for some time. A case in point is this Arid Eudesmia Moth found “deceased”: https://www.inaturalist.org/observations/194578833
This species doesn’t fly in December in Texas, so the moth had apparently been dead for some time. So the date of the observation is correct, but it offers a data point which might complicate the examination of seasonality.
Q: Since there is usually no way in such cases to estimate the date of death, is there any other way to treat such observations? Marking it as “No” for “Date is accurate?” and thus religating the observation to Casual status seems overly heavy-handed.
Is the number of such cases so vanishingly small that I shouldn’t worry about it?!
This particular study has lead to initiation of remediation efforts to save Pacific newts from local extinction due to excessive roadkill . You would not want to harm the effort by messing with the data.
iNat observations record encounters with organisms or recent evidence of organisms and “Date is accurate” refers to the date of the encounter, so that shouldn’t be used. I wouldn’t worry about it too much. Annotate as Dead and hopefully anyone taking a close look will see that.
This is comparable to animal sign, such as photos of beaver dams or lodges that might no longer be active (e.g., abandoned years prior). A long-dead carcass is the equivalent to sign, where a live animal might no longer be present but was there. Same for winter photos of plants that are dead but were alive earlier in the season.
Yes, roadkills are a whole other topic and worthy of special study and recognition. My question pertained to a smaller subset such as insects found dead (and thereafter documented) inside a home, etc., etc. In the end, it revolves around the interpretation of “Recent evidence of organism” and I’m very willing to interpret that quite generously. As @tiwane suggests, it’s a matter of annotating such observations properly and recognizing that they may not represent “current” occurrence/behavior.