Settings

Where and when it is said to be

The clock offset is the single commonest way one of these calculations goes an hour out. A picture taken in London in July is +60, not 0, because the clock was on summer time. Neither the place nor the date leaves this device.

The shadow you measured

Height and shadow can be in any units at all as long as both are in the same one: only the ratio matters, which is why a shadow can be measured off a photograph with no scale in it. Leave the direction blank if you cannot tell which way north is; the answer is then wider, and honestly so.

When it could have been

The curve is how high the sun stood through the day. The shaded bands are the spans your measurement is consistent with: there are usually two, one as the sun climbed and one as it fell, and the direction the shadow points is what tells them apart. The dashed line is the time claimed.

Against the claim


    
How it works

Photo metadata is stripped or easily edited, but the sun is not. SHADOW estimates when a photograph was taken, or checks a claimed time, from the direction and length of a shadow and the place it was taken. It compares the shadow with where the sun would have been. It gives a range, not an exact time. The photo stays in your browser.

The guide: How to work out the time of day from a shadow in a photo

Questions people ask about SHADOW

Can this tell me the date a photograph was taken?

No, and anything that says it can is wrong. The sun retraces its path either side of a solstice, so every shadow except the two extremes happens on two dates a year, and near the equinoxes the sun changes so slowly that a fortnight either way is indistinguishable. What a shadow does narrow is the time of day, which is usually the disputed part anyway. To get at the date you need a second shadow at a known interval, a sunrise or sunset in frame, or something else entirely.

How accurate is it?

The astronomy is good to about a hundredth of a degree over any year you are likely to ask about, which is far finer than a shadow can be measured from a photograph. So the error that matters is always yours, never the arithmetic: measuring a shadow off an image is good to perhaps five per cent on a good day, and the page turns whatever tolerance you set into the span of times that are actually consistent. It gives you a band because a single time quoted to the minute from a measurement that loose would be a lie with a decimal point in it.

Why is refraction not applied?

Because it would be false precision. Refraction lifts the apparent sun by about half a degree at the horizon and by almost nothing above twenty degrees, and a shadow cast by a sun low enough for it to matter is so long that it cannot be measured off a photograph usefully anyway. Adding a correction that is negligible where the input is good, and swamped by measurement error where it is not, buys nothing but a more confident-looking number.

The shadow agrees with the claimed time. Does that prove it?

It supports it, which is weaker and worth saying out loud. A shadow consistent with a claim rules out a great deal (every other hour of that day, for a start), but the same shadow falls again on many other dates, and a person choosing a time to claim can choose one that fits. Disagreement is the strong result here: a shadow that could not have been cast at the time claimed is a fact about the photograph, not an opinion about it.

It disagrees. Does that mean the photograph is fake?

It means the photograph and the story do not match, which has several innocent explanations before you get to a dishonest one. A camera clock never set, a time zone taken as local when the file is in UTC, daylight saving, or a place a few streets from the one assumed will all produce a disagreement. The page tells you which times the shadow actually fits; comparing that to the claim is usually what identifies which of those it was.

What do I need to measure off the picture?

An upright object and its shadow, in the same units; pixels are fine. A lamp post, a bollard, a person standing straight. The direction the shadow points is worth having if anything in frame lets you establish north, because it halves the answer: the same shadow length happens once in the morning and once in the afternoon, and only the direction separates them.

Doing this often? OBSCURA Case keeps photographs with the rest of the evidence: searched, tagged and in the chronology. Everything is US$5 a month or US$48 a year.