How is the time of a sighting generated?

I’ve noticed a series of rather unlikely sightings by a recorder. Part of what makes them unlikely is how close together they are in time. But then I looked at some of my own records and I see iNaturalist has put a time on the observation even though I only entered the date, and the photo of the specimen was taken many years after the sighting. Is iNaturalist getting the time from data embedded in the photo? In which case it will be wrong when the photo is of a preserved specimen.
John Bratton

1 Like

Screenshot%20from%202019-09-27%2013-32-53

Unless you specify the time by clicking on the little clock below the calendar, I believe it will keep the “default” time: so either the time of upload or the timestamp of the photo itself if there is one.

If there is no date/time attached to the photo and you enter a date, the time then becomes the time you entered said date (slightly different to the time of upload). When this happens for me, I just delete the time and keep the date.

It has been suggested that we need an accuracy value for time in the same way as we have for location. ie I saw it on this date +/- 9 days, or at 3pm +/- 60 mins

1 Like

I don’t really understand this - it’s bound to create inaccuracies if you don’t pay attention, and it’s never going to be accurate if you’re using the website to upload observations. Wouldn’t it be better to leave it blank or force you to input a time manually?

3 Likes

It only occurs if the photo’s metadata does not include a date/time. In 99% of cases it isn’t a problem but almost all photos do include this metadata. The rare times they don’t are in cases like you downloaded the photo onto your computer from Facebook, the photo is a screenshot, etc.

1 Like

This is also an issue for audio uploads, which can’t have their metadata read. I’ve uploaded a large amount of audio and didn’t even know I could erase the time, so all of the times for my recordings are wrong.

2 Likes

Almost certainly, yes, if the photo has those data embedded. Observations based on photos of preserved specimens will always need to have their location and date/time edited manually to reflect those of the original collection, leaving time blank if only date is known.

Welcome to the forum @jhbratton!

Thanks for all the responses. I’ll edit the time out of mine where it is incorrect.

1 Like