How to determine shooting time using shadow length

We consider the method of determining the shooting time by measuring the length of shadows, using the SunCalc tool to simulate the movement of the sun in the sky. This can be a useful tool for open source researchers, especially when the location and date of shooting are known, and the subject and shadow are perpendicular to the direction of the shot.

Let’s start

Determining when a video or photo was taken can be a daunting task for open source researchers. Ideally, a wall clock or a wristwatch could get into the frame – but most often this does not happen. However, the solution may be to measure the length of the shadows.

It has long been known that the shadow angle can be estimated from the estimation of the shooting time. SunCalc, a tool that simulates the movement of the sun across the sky on different dates at different points on Earth, allows you to perform such an operation quite simply.

However, sometimes the angle of the shadow is not completely clear, or the video or photo may not allow you to accurately estimate the angle.

However, just as the angle of the shadow can be used to determine the angle of the sun relative to north (azimuth), the length of the shadow can be used to calculate the angle of the sun above the horizon (elevation).

As the sun rises above the horizon, the shadows are initially westward and very long. Around noon, the shadows become the shortest. Further, the closer to sunset, the more the shadows are directed to the east and the longer they are. This estimated change in length can be used to estimate time. It is important that you do not need to know the actual height of the object that casts the shadow. Only the ratio of the length of the shadow to the height of the object is important. This means that the measurements in this article are not related to the actual height and length of objects and shadows. Since we only care about their relationship to each other, we are not interested in actual measurements.

For this technique to work correctly, the photo or video must meet several conditions:

1. the place and date of shooting must be known;

2. the object and the shadow should be located approximately perpendicular to the shooting direction;

3. the shadow must fall on a surface parallel to the horizon (for example, on a flat floor);

4. the picture should not be too distorted by the camera lens, such as a fisheye lens.

After all, using this technique, you should remember that shadows have a certain length twice a day: in the morning and in the afternoon. Therefore, it is important to check the direction of the shadow.

Example 1: controlled conditions

In this example, I’m using a photo I took around the Al-Aqsa Mosque compound in Jerusalem on October 14, 2019. Let’s try out the technique of estimating the ratio of shadow length to object height and see what time we get.

In this photo, we see a security guard leaning on a fence. We can see that this photo meets the four conditions mentioned above:

1. the place and date of shooting must be known;

2. the object and the shadow should be located approximately perpendicular to the shooting direction;

3. the shadow must fall on a surface parallel to the horizon (for example, on a flat floor);

4. the picture should not be too distorted by the camera lens, such as a fisheye lens.

Next, we measure the height of the object image and the length of the shadow image in the image. Remember that the ratio of measurements is more important here than the measurements themselves.

Then we go to SunCalc, set this area of the Al-Aqsa complex as a place and enter the object height of 2.33 meters (parameter at an object level SunCalc). Next, we will increase the time until the length of the shadow is as close as possible to the length in the picture (3.53 meters). Data entry into SunCalc is shown in the screenshot below.

This gives us a time of 9:37. After performing the analysis, I checked the EXIF data of my camera and found that the picture was taken at 9:43, that is, the difference between the actual time and the estimated time was only six minutes.

Example 2: Video frame

In the example below, we see a video frame where a group of people watch a cloud of smoke from a Saudi coalition airstrike in the town of Zabid in Yemen on May 12, 2015. As a result of this airstrike, many civilians died.

We see that the shadow of the man in the white shirt (on the left) falls on a flat surface roughly perpendicular to the direction of the shot. So, we can measure the ratio of the length of the shadow to the height of the object casting it, and calculate the shooting time.

In this case, the height of the person in the picture is 5.67 centimeters, and the length of his shadow, which appears to end just beyond the leg of the person on the right, is 9.84 cm.

Next, we’ll use Suncalc: set the “object level” (the height of the object casting the shadow) to 5.67m and experiment with time until the shadow length is as close as possible to 9.84m. This gives us a time of 16:14.

To check the accuracy of the method in this case, let’s change the time by 15 minutes on both sides from 16:14 and see what the length of the shadow will be.

• At 16:14, the ratio of the length of the shadow to the height of the object casting it is 1.74:1.

• 15 minutes earlier, at 15:59, the shadow is 1.51 times longer than the object that casts it.

• 15 minutes later, at 16:29, the shadow is 2.02 times longer than the object casting it.

As you can see, although the time established using this calculation of the length of the shadow should be considered quite approximate, we can be sure that the deviation from 16:14 is no more than 15 minutes in one direction or another.

When using this technique, it is often more useful to set an approximate time frame than a specific time. It is quite simple: it is enough to shift the time in SunCalc until the length of the shadow no longer corresponds to the calculated one.

Conclusion

This technique is a bit more complicated to use than calculating the time from the shadow angle, but it can still be useful if estimating the shadow angle is difficult. Under the right conditions, this technique can be quite accurate, sometimes even surprisingly accurate.

The information was taken from Bellingcat’s open sources