[ Edit:  video, still images and text updated to reflect additional detected satellites ]

This image plots satellite trails across the sky during a one-hour period, superimposed on a star trails image taken at the same time.

Details

I travelled to Mount Seymour on 12 August this year to take wide-field images of the Perseid meteor shower. For practical and safety reasons, I set up in a car parking area (part of the way up the mountain) to take the images. This also allowed for the lighting of foreground trees by passing vehicles. 

The usual issues conspired to make the night a challenge – skyglow from the city below, clouds, too many cars with people looking to watch the same meteor shower, and sadly smoke from the extensive wildfires across the Province. Fortunately, there was a one-hour time slice of better conditions, allowing for a star trails image. However, my focus transitioned to the effects of satellite trails.

Equipment used included a Nikon D750 camera and a Sigma Art 20mm f/1.4 lens on a fixed tripod. Settings were f/1.8, ISO 1600.  The angle of view was 84 x 62 degrees. 259 x 13 seconds subframe exposures were taken.

In the one-hour period, at least 26 satellites passed through the camera’s field-of-view; with trails able to be seen in 58% of all subframe images. The levels of satellite visibility in the images varied from ‘easily-seen’ to ‘slight’. Most of the visible satellite trails could be readily seen in the uncorrected images crossing from one end of the image to the other. A few satellites could only be seen for portions of their movements across the sky. It is likely that more satellites passed by unseen in my images. No aircraft trails were present. And yes, there were a few meteors.

In the end, 258 satellite trail segments needed to be erased from 151 (out of 259) images to obtain the final image (without satellite trails). 

Of course, it is standard practice when processing deep space images to eliminate these sorts of trails with pixel rejection during the stacking process. However, this can be a much more arduous manual process for wide-field images that capture numerous satellite trails, and where many subframe images may need to be reviewed and corrected if necessary.

The initial image (i.e. still with satellite trails) is provided in the version history. The fully processed image, without satellite trails, is posted separately.

In addition, the linked video presents a timelapse of the satellite trails during the one-hour portion of the night. The timelapse is also superimposed on a wide-field star trails image.