Shadows and Suncalc
I received a question from Micah Hoffman, who was doing some writing for his SEC487 OSINT class in regards to determining the time and date a Google Streetview image was taken. He was able to determine the time pretty close but did ask me to verify his findings and whether I had some additional hints. Honoured to be included in a SANS course, I will take you on a small hike along the several steps taken to determine the most probable day this scenery was captured by Google Streetview, followed by finding the exact time within that day.
Location and Surroundings
The location we are looking at is the Israel National Trail, next to the Reading Power station in Tel Aviv. The link to this exact location in Google Streetview is: https://goo.gl/maps/3bEuT2yJcTR2
On the left of this picture we see the power plant and to the northeast we see Sde Dov Airport. The shadows seem to be coming from the east, but that is simply not accurate enough to determine the exact time. We need to determine the direction of the sun and get as close as possible to the exact angle, and preferably we need the exact day this photo was taken.
Finding the Day
Looking at Google Streetview, we see it was captured in June 2015, but the exact day is not published. As far as I know it is not possible to find it in the meta data or responses that are sent to the browser or to Google Earth. So we‘ll be looking at something else.
If we look upwards, we see there is a heavy overcast on this particular day. Going over historical weather data of Tel Aviv in the month of June 2015 we find that only on one specific morning (yes morning, remember the shadow was coming from the east?) there was a heavier overcast compared to other days that month. Using that piece of information we have reason to believe this was the exact day this Streetview scene was captured.
The whole month of June 2015 can be viewed at the ‘Time and Date’ website here:
Finding the Time
Now that we have the day and the location, finding the exact time is the easiest part. It consists of nothing more than finding out where the shadow is coming from, or in other words: Where the sun was located at the moment of capture. To do this, we go back to Google Streetview. I am going to move two ‘clicks’ (as in mouse clicks, not as in kilometres) towards the bridge and turning slightly left. By doing this, we find a shadow that is lining up perfectly with buildings in the background, therefore making it possible to find the direction the sun is in as accurate as possible.
To find where exactly the shadow is pointing at we need to find out which buildings can be seen in the background, so we are going to have a closer look at that. In the green we see a hangar or shed in the foreground. Behind that we see a line of multiple apartment buildings highlighted in orange, and left of it a few more buildings that seem to be closer together, highlighted in red.
Since we know exactly where we are standing (to be exact, we are standing here: https://goo.gl/maps/eHwAEf4Xfw32) we can draw a line towards the direction we are looking at. By using clear satellite imagery that is at least somewhat close to the date we are looking at, we find we are looking at the buildings highlighted in red and orange.
And the green circle is the building the direction of the sun seems to be in, which can be found by using the shadow we found earlier on and extending it all the way towards the buildings in the background. It is a bit difficult to see in the following screen capture, but do try it yourself!
Let Tools do the Talking
We now have an exact location, the day this capture was made and very strong visible clues to line up the direction of the sun. We can use Suncalc to line up the sun as close as possible, but sometimes this can be somewhat difficult in certain situations, since we need to zoom out considerably to be able to do that. Later on we’ll show how to use Google Earth to help with this.
In Suncalc we first set the date on June 28, 2015 and then place the marker right on the spot we were standing, which is the location where the concrete path changes to wood. Then we change the time until it lines up with the second building from the left within the ‘orange’ block. After sliding a bit back and forth, we find the following: June 28, 2015 at 8:29 UTC+3
To double check our findings, we will have a close look at the ‘Azimuth’, or the direction of the sun, which is highlighted in red in the image above. If we go to Google Earth, we can determine the direction the sun is in, by using the ruler tool. For that you click on the ruler on the menu bar, then first click on the location we are standing (near the bridge), then we click on the second building where the shadow is pointing at. Google Earth will give us an angle of 81.86° which is pretty close to the 81.78° we found by using Suncalc.
If you are able to find the direction via such an accurate way, then I personally would recommend you use that before using Suncalc. This goes especially in cases where the points that need to be lined up are too far away from each other, forcing you to zoom out a lot within Suncalc which will make any findings less accurate.
General Tips and Hints
When trying to find the day and time of a photo, try to find objects that line up in the exact same direction as the shadow go. The direction of the sun alone is enough to find the time, but only as long as you have an exact day that the photo was captured. If only the month and year are known, the accuracy will be roughly within 15 minutes or so, depending on where you are and what time of year it is. If nothing is known, then you first have to use other methods to determine the date as accurately as possible, but that is for another blog.
Since in this case it was Google Streetview, it was easy to find a suitable spot that fit that criteria. A small warning though when doing this: Google Streetview sometimes stitch together different ‘runs’ that are taking at different moments in time. So do make sure that when moving around in Google Streetview you stay within the same ‘capture-sequence’ of that same day. Since there were people visible just north of the location discussed above I was able to confirm that the location I preferred to work with was indeed taken at the same time and date, and not a few days or weeks later.