You are talking about something that occurred 12 years ago. I no longer live in that area, nor do I have the telescope. It was a refracting Celestron that was advertising itself as 500x equivalent. The experiment was conducted from several different spots in that area.
Tom, I have several telescopes, Celestron CPC800 and 1100, Meade LightSwitch and LX200. To achieve 500x magnification with humanly visible optical resolution, you need a huge aperture, never produced on a refracting unit for popular use. It would need to have a more than 10 inches of refracting glass as objective, you will not be able to carry it handy. Also, the smaller eyepiece you can see anything from 12 years ago technology is no less than 6mm. To have a 500x magnification such refractive telescope must have an objetive with a focal length of 3000mm (3 meters long), the whole telescope with the focuser would be longer than 3200mm. Celestron never produced such best.
You may had a 50~70mm objective refracting unit, handly transportable, 800 to 1000m long, but it needs a nice mount (tripod), you can not use it on the floor. Somebody may stick a label with "500x" on it, but no cigar for that. That unit can give you a maximum of 200x magnification in the limit of optical resolution. Optical resolution formula is d/1.22Lm, where d is aperture and Lm is the green light wavelength, often used on astronomy calculations. Regular 15mm~24mm objectives (popular for that telescope) can give you a somehow visible image with 40x~60x magnification. With that magnification, a 2m tall image at 38km distance can be seen with an aparent size of around 0.02°, what is 1/20 of the size of the Moon. It will be like watching the details of lunar crater Langrenus by naked eye, or a person as seen by naked eye 633 meters away (38000/60). I need to admit, based on naked eye observation I can not even tell if there is a person 6 city blocks away, not even a car.
A 22cm ball (or freesbe) has an apparent size of 1 arcsecond at 46.5km away, that is 1/1800 the size of the Moon. My CPC1100's aperture is able to discriminate 0.5 arcsecond, so it in fact start to lose optical discrimination of a freesbe at 90km away, image fuses with surrounding photons. It means you can not recognize it as a freesbe, the wavelength of the image is higher than the size of the object, you just can actually see a different brighness fuzzy thing, nothing else. Considering a person has similar size head, you can not even say if that fuzzy dot on top of a very tinny little stick is a person's body or a lamp pole, and that with my CPC1100 (11" mirror, schmidt cassegrain, 2800mm focal length, 85lb of weight, resolution 0.5 arcseconds, $3k) without any extra features. Over many kilometers of water, moisture a lot, waves spraying it becomes really difficult.
Next time take pictures, from the telescope, from the scene, from the scene through the telescope with different eyepieces so you can have a progressive image magnifications.
There is an easy experiments, not even need image or photos, it uses three lasers, two powerful (minimum 1W units) red and one green lasers. I wonder why nobody made it before. Mount them side by side over a wooden base, green in the middle, with screw for vertical alignment. Shot them at night over dry land against a building wall few miles away, as far as possible. A person close to the buildings cell phone the one with the lasers, and tell him to adjust the screws for the three lasers to be aligned, no matter if they are angled or horizontal, just aligned, green as better is possible in the middle of reds. Then go to the beach and shoot them over the 48km patch of water against a big building on the other side. Using a cell phone tell the person with the lasers to very slowly tilt down the front of the lasers base until the spots disappears down at the receiving side, then slowly tilt it up until they appear, so you can measure the minimum altitude the lasers hit the building wall. Then, measure or estimate, how big is the difference of alignment between the green and red spots. There will be a difference of alignment, the green laser will be lower than the reds. Red light refracts different from green on a moisture oceanic air. By measuring the misalignment of the beams, meaning diference of refraction from red to green wavelength, we can calculate how much refraction it is actually happening in general, air density, etc. So we can insert this variable on the minimum height of the receiving beams on the building side and calculate the correct numbers. Another blue laser could be used together, since blue bends even more. We can not just assume the light travels straight over a patch of moisture air, it will bend down as if going through a very low density glass. This is specially pronounced over lakes, ocean, etc. It also happens over land with less effect, the thermal difference from the ground and the air creates this cushion of moisture and warm turbulent air, it creates havoc for visible sight. Sharpshooters know that and compensate for plain dirt terrain, moist, water, jungle, dry, rocky, sometimes even the color of the land changes everything, dark color retain more warmth and create uplift air flow. It is very difficult to hit a 20cm target with a bullet at 1500m away, even with a supersonic projectile, they need to know and compensate for everything. That is not only compensation for the bullet travelling, it is also visual compensation on the scope, light refracts easily.