These piles dot the northern half of Thunder Bay near Alpena. Each circle is roughly 15m in diameter and some contain lumps of coal. Click image for larger view and image credit.
Man-made “tracks” in Thunder Bay: In the northern half of Thunder Bay there are numerous “tracks” that appear similar to anchor drags, but they are often 17m wide. Some are raised a few meters above the lake bottom. Click image for larger view and image credit.
Mission Summary
Charles Loeffler
Principal Investigator
Senior Engineering Scientist
Applied Research Laboratories
The University of Texas at Austin
The NOAA Thunder Bay National Marine Sanctuary (TB-NMS) is a large national sanctuary (448 square miles), which includes all of Thunder Bay and a part of Lake Huron. It is the site of hundreds of well-preserved shipwrecks. The locations of numerous wrecks are known but many others have not been found. NOAA is charged with maintaining and surveying this area, and they are always looking for new approaches to efficiently survey this large area. Their problem is compounded since the sanctuary may be expanded to over 2000 square miles. Back in 2008, NOAA and The University of Texas at Austin Applied Research Laboratories (ARL:UT) felt that the ATLAS-AUV would be a good sensor system to survey this area for known and unknown shipwrecks sites. Since the time when the project was originally proposed, Russ Green (Thunder Bay’s Deputy Superintendent/Research Coordinator) decided to expand the role of ATLAS to do more than just finding shipwrecks. He wanted to try using it to identify bottom characteristics, locate marine habitats, and survey prehistoric archeological sites.
Lake Floor in the Deep Water Site: The lake bottom in this area has a brush-stroke appearance that might be glacial scrapings and brighter lines that could be drag marks. Click image for larger view and image credit.
The team of ARL:UT and NOAA scientists had 10-1/2 days to run surveys. Due to the weather, missions were run on 7 days for a total of 52-1/2 mission hours in shallow and deep-water areas. We surveyed approximately 104 square miles for an average rate of two square miles per hour.. In deep areas, the rate increased to three mi2/hr. While in the shallow areas, where the SVP limited the sonar’s operating range, the rate was 0.4 mi2/hr (which is still higher that many other survey methods). The whole team was pleased with the area covered, the discoveries, and the quality of the maps. We’ve all realized that there are many applications for this sensor and/or AUV technology, including large surveys of ship-wreck sites, bottom characterizations, archeological sites, and for environmental evaluation.
AUV Swims back in Dangerous Seas: Increasingly bad weather forced a mission abort. For safety considerations, the AUV was reprogrammed swim back into to the protected waters of Thunder Bay. Click image for larger view and image credit.
Ultimately, it was truly a successful test. The AUV, sonar, and processing ran without any incidents. (There were no software crashes or AUV crashes into the lake floor or ship wrecks). The weather was unusually poor, and we did not get out to the deep-water area as much as we would have liked, but we still covered a lot of area. The deep-water area is difficult for NOAA to survey at this time, but even with our limited on-site time, we did survey a larger area than they had ever covered before. In the shallow area, we surveyed nearly half of Thunder Bay and the ATLAS maps revealed unexpected features and human activities. These discoveries have renewed NOAA’s interest in the region within the bay. For example, there are hundreds of 15 meter diameter circular piles throughout the bay. A dive-team inspected one of them during our second week and discovered that they contain piles of large hunks of coal. There are also numerous manmade features that look like snowmobile tracks in the approach area to Alpena. Some are raised a few meters and some are depressed. All are roughly 17m wide. There is quite a bit of speculation on their origins. In the deep-water areas, one of the dominant visible features are possibly glacial scrapings, which looked like “brush strokes” on the maps.
In addition to the amount of area surveyed, there were a number of other results from the test.
- We realized and developed a back-up safety plan for mission planning. Since the AUV had a second battery pack, it could be programmed to swim back into the protected waters of the bay on its own, which eliminated any AUV recoveries in high wind and waves. It is now part of ARL:UT’s standard planning procedure to include a safety back-up transit to safe waters for the AUV.
- We used a new method to get the AUV into the water after assembly by simply launching it from a boat trailer.
- ARL:UT collected tera-bytes of data in the deeper depths and at higher speeds than ever before for this system. This data will be used to develop improved sonar data processing schemes (i.e. computer programs).
- New techniques to pull data off of the AUV during a mission were successfully tested. This is used to immediately evaluate the performance of the sonar and to provide an early glance at any potential discoveries.
- Additional challenges with strongly refracting sound velocity profile were observed and will need to be solved with new processing schemes.
- A method to launch and recovery AUV from the R/V Storm was developed and tested.
- The ATLAS maps were used to locate interesting objects and features on the lake-floor, then divers and a small side-scan sonar were sent to those specified locations to inspect them.
- NOAA’s Thunder Bay facility and sanctuary have been realized as a good location to operate, test, and develop AUV systems.
The Alpena Marina is within a few minutes of the NOAA facility. A very simple and effective technique to transport the AUV to the marina and launch it was to use a standard boat trailer. The procedure was straight forward with no high crane lifts or jarring motions. After launch, the AUV’s remote control was used to motor it over to the support ship where it was lifted on-board with the ship’s crane. Click image for larger view and image credit.
All of these results will help us develop or define a more robust AUV system to run increasingly independent missions, and will help us determine better ways to use AUVs to survey for objects, like those in the Thunder Bay Sanctuary.
One “result” that was personally satisfying for the ARL scientists was the ease at which the ATLAS maps were understood and interpreted by the NOAA staff and other non-engineering visitors. Every one who came into the mapping work area would study the maps for just a couple of minutes to get oriented, then ask a few questions and begin talking about the geology of the seafloor and not about anomalies due to the sonar or processing. This group included maritime historians, archeologists, SLS operators, and divers. Everyone seemed to understand the data.
