< Back to Blogs


by Mark Lemmon

April 17, 2008 - I'm still recovering from the latest exercise in simulating operations on Mars, last week. We spent 3 days, running about 2 sols per day, doing an event called the Ice Thread Test. A thread test is a chance to focus on some aspect of operations that needs attention, without bringing the whole project to a focus on simulating all aspects of operations.

In this case, the "thread" was the discovery of ice. A handful of us were involved, representing mission management, science, the robotic arm team, and the camera team. Behind the scenes, the PIT crew had prepared a great set-up and kept things rolling. We started with a pristine workspace with a few intriguing areas to dig ("those 4 rocks mark the corners of one of the allowed dig zones"). Some things about the test were comforting: we had a small tilt, and the workspace was aligned north. We’ve practiced with tilts up to around 15 degrees, which are tilts we can deal with, but they complicate operations and are not at all likely. We’re landing in an area with regional slopes less than 2 degrees and few boulders at the scale where they could cause big tilts. We are hoping for polygonal ice fractures, but the polygons are 3-6 m across and
the relief in the fractures is expected to be centimeter scale.

We dug the first sol (Wednesday morning) and hit something hard and bright. The ABM ("Anomalous Bright Material"--we figured this is NASA, so a 3-letter acronym was mandatory) turned out not to be ice. We were told to pay no attention to the edge of the box ("PVC" was the more accurate acronym).

Our second dig resulted in hitting something more interesting. We had a gray-green layer partly filling the trench bottom, with some red stuff in the walls and on part of the bottom. At this
point, we started trying to characterize the cool stuff we had found. It didn't sublimate away overnight -- OK, for the testbed in Tucson in April, they weren't going to use real ice. We were waiting for all the data, since we didn't know if we had a thin green layer and then red stuff, or if the green layer was the ice-like goal of the test. So we spent a couple sols characterizing it, showing that it was hard stuff. It turned out to be hard enough to require the robotic arm (RA) scraping tools.

The RA has a few ways of digging with the scoop (sorry, the Icy Soil Acquisition Device). Normal digging relies on the front blade, and only got a little dust. For sampling the ice we can use the rasp (our little drill on the back of the scoop). In between, the scoop (ahem, ISAD) has scraper blades mounted on the bottom. We found that, at least at first, we did not need the rasp and the scraper blades could do quite well.

Some of the things we learned by doing this test involved readiness (that is, we still have a to-do list). We weren't quite to the point of being able to rasp the trench Friday night as planned. A bigger lesson was playing out the decision cycle and trying to act in a flight-like way. In a test, you know there is an answer, a goal. You proceed toward the goal, perform the tasks required, and are done. In real life, we will be exploring. We don't know what we will find, or how many
different interesting things will be unearthed ("unmarsed" just doesn't work for me). It was relatively easy to say that a greenish layer in the trench was going to be analyzed. But deciding if it was soft or hard, thin or thick, plentiful or in short supply -- these decisions take time and
information. Looking at this trench through the eyes (and sense of touch) of a robot will complicate decisions. We tend to think that "we'll know it when we see it" about the ice and any other interesting layers -- but that's only true once we get all the information. So, I think we're looking forward to a very interesting process of discovery. And now, after the Ice Thread, I think we have a better understanding of what our limitations are, and what information
needs to be prioritized to get around those limitations.

Below is an animation of 4 images of the trench from the Surface Stereo Imager. The camera did a great job capturing the size and shape of the trench. The colors are a little off due to the
limitations of the PIT (warm camera, different lighting on the calibration targets and the trench). The collapse features along the left side are very intriguing. The end result of the scraping (between the last two frames) was a nice smooth surface where we could have used the rasp the next sol.