Part 7 - Searching for Jussi Peltola

The morning of the 5th of July finally came and I started my long drive to the north with a fully packed car. Not only did I have all the search gear: the RC-boat, the ROV, and all the tools and spare parts needed for repairing the things, but I also had my personal stuff for the four week vacation.

The drive was slow, as it seemed that most of the Finnish population had decided to hit the road at the same time. It is said that the whole of Finland closes down in July as everyone goes on vacation.

Around noon, I was at the site and met my brother there. We had agreed to start from the canal and if we couldn’t find the car in there, we would continue to the other spots.

By the Canal

It was a hot day with the temperature above +30C. Not that it made a difference as I would have been sweating even if it was freezing. I was nervous as hell.

The spot we were searching was a canal that leads to a dam. The current was strong and I wasn’t sure the RC-boat had enough oomph to fight the current. In my mind I saw the strong current sucking the RC-boat to the dam and the hydropower plant spitting out the parts from the other side.

I started the RC-boat, the sonar, and attached a phone. The phone was used to stream the screen of the sonar to a tablet which was on shore, so we could watch the sonar image in real time.

We pushed the RC-boat into the canal. I slammed the throttle 1/3 ahead and the RC-boat easily fought the current. It was clear it packed some real power.

Cruising around for a while proved that everything was working just fine. The RC-boat handled the current well, the streaming worked fine, and the image was sharp. There were only two problems. First, the RC-boat didn’t have a proper hull as the sewer pipe hulls were round. It was easily solved with some planks and zip ties. The second problem was that there was no car in the canal.

Piloting the RC-boat Piloting the RC-boat from the dam.

The place was quickly searched but I refused to believe it. We searched the place again and again. And once again for good measure but there still was nothing in the canal.

After way too long, I was able to finally admit it. The theory was good, but the car was not in the canal.

RC-boat at the canal The RC-boat at work on the canal.

Searching by the River

After I was able to admit being wrong, we packed the gear and drove to the 2nd site, next to the bridge, about 8 km east, towards the E75. We found a small “harbour”, ie. a place the locals used to get their boats onto the river. We parked our cars and started unloading the stuff.

I drove the boat up and down the river while glancing at the sonar video from the tablet every now and then. Once again it worked flawlessly.

The bottom of the river looked funky. It was completely smooth, as if it had been dredged. Also, the current was strong, stronger than at the dam. The riverbed was featureless, except for a water pipeline and a pile of rocks. The pile of rocks was weird as it kept changing its form every time in every image. But there was no car. We searched the place from all angles, but we couldn’t find the car.

The Next Day

We spent the night in a nearby hotel, and returned to the spot in the morning. We searched the river again. This time we also searched downriver, as far as we could see. It was hard to drive the RC-boat once you couldn’t hear the engines. After almost eight hours of working, we finally gave up and packed up the gear.

It felt good to see that the equipment was working so well but I couldn’t shake the feeling of disappointment as we didn’t find anything meaningful.

Searching at the Desk

The next few days I spent at my computer. I had bought a software which could be used to analyze the sonar recordings. During the spring I didn’t have much time or material to learn the software nor how to interpret side scan sonar images. I had the basic understanding but no practical experience.

Once again, I watched Youtube videos and learned a trick or two. Much of the side scan sonar information is actually in the shadows that the objects cast, and especially in unclear images it is more helpful to watch the shadows instead of the actual object. The trick I learned was that side scan sonar works much like a flashlight traveling under the boat and interpreting the image is like interpreting the shadows of the flashlight. Learning the basics, and how to use the software and seeing what difference adjusting the brightness, contrast, and gain made, was a revelation. With this software, I could also measure objects and distances rather easily.

Analyzing on a computer was so much easier than while out searching. The streaming video came with at least a 10 second delay, so it was difficult to know which bottom feature was where. Driving the RC-boat took up most of my focus. Fortunately, we had recorded all the searches so we had lots of material to analyze.

Analyzing the images on my computer verified that the first site was clear. It was not possible for the car to hide there.

Canal The canal was empty. Only rocks and small trees on the bottom.

The second site was also empty, expect for the pile of rocks. It started to look really interesting. First of all, the size of the rock pile was approximately 5m x 2m. Second, the place it was located was exactly right. It was approximately 5 meters from the shore and a few meters upriver from the pylon of the bridge.

It looked promising, but it wasn’t certain by any means. The biggest problem was that it looked so different from different angles. In one image, it looked like Santa’s sleigh, in another it looked like it could be a car standing upright, and different one it looked like a car was upside down. There was also one where it resembled having a giant wing like a Formula 1 car, and one where it simply looked like a pile of rocks.

What caught my attention was that in a couple of the images there was something that cast a shadow of what could be a tire. I tried to keep calm and not get ahead of my skis, but I knew that we really, really needed to check this out.

By the bridge One of the clearer images. Shadow from tyres suggesting it might be a car. The river bottom otherwise smooth.

Checking the weather forecast, the 16th of July looked promising, as it was forecast to be the coolest day in a long time - only +28C.

The night before, I was at a local lake testing and tuning the ROV. Everything else was working fine, except the video feed. I couldn’t get all three cams to start. I suspected the USB connectors of the cameras were faulty but there was nothing I could do except pray. It was going to be the first real test for the ROV.