I've been looking into chronolocation and one part I find particularly difficult is determining when a photo or video was actually recorded when there's no reliable timestamp attached to it.
There can be clues in the image itself — weather, shadows and sun position, vehicles, construction work, vegetation, signs or other changes in the environment. Historical satellite imagery and archived sources can sometimes help too.
What methods do you find most useful for narrowing down the date of an old photo or video, and how do you decide when the evidence is strong enough to support a specific time period?
6 answers
Earnest Researcher 329
I usually check the shadows first, if I have some idea where the photo was taken. The direction and length can give you a rough idea of the time of year, and sometimes the time of day too.
I wouldn't take that as proof though. The same shadow can probably fit more than one date, so I'd use it mainly to narrow things down before looking for other clues.
Quiet Cartographer 196
I often try to work backwards from changes in the surroundings. If there's a building, road, bridge or some other distinctive feature in the image, I’ll check older satellite images or archived photos to see when it looked like that. What works for me is to first narrow it down to a period, for example that the photo was probably taken between 2018 and 2020.Once you have that, it’s much easier to compare it with other clues. I rarely try to pin down an exact date from just one clue.
Wandering Researcher 260
It also depends on what form the image is in. If you have the original digital file, it’s worth checking the metadata first, but that may not help with an old scanned photo
Careful Cartographer 231
I agree that the key is to compare several different clues. A shadow on its own may not be enough, but if a building or an old satellite image also supports the same time period, I find it much more convincing.
That does seem more convincing than just using the shadow. If a building or an old satellite image also points to the same period, I'd trust the estimate more.
Careful Cartographer 231
Thanks everyone, these answers have really helped! Combining the shadow angle with changes in the surroundings does seem to work well for narrowing down a time period. On the metadata point though, I'd add that I wouldn't take it at face value either it can be edited or stripped, so on its own it's not enough to confirm a date.