Note: I am posting my old blog entries from elsewhere on the internet. This entry was originally posted on 6/21/10.
I’m in Tucson right now, working on assembling two weeks’ worth of images for the HiRISE camera. It’s really hot here, but I’ve barely noticed because I’ve been in heavily air conditioned buildings all the time.
One of my multiple jobs is to represent the interests of the HiRISE science and operations teams to the rest of the people running the Mars Reconnaissance Orbiter at JPL. As part of that, I need to know all about how HiRISE plans its images, takes the images, and processes the data before getting it out to the world (pretty pics at the link!). So I am taking a turn as the scientist for a planning cycle. Since it is my first time at this, I’m out here in Tucson – the camera operations team is based at the University of Arizona, and by being here I can work with them directly.
Each planning cycle has a scientist from the science team and a targeting specialist from the operations team who work together. We have a database of targets from team members and the public (here is where you can enter a target!), and the targeting specialist pulls out what targets will be observable by the spacecraft during the particular two weeks of this cycle.
I spent Friday through Monday culling the list down to a subset of targets that were high-priority or that I thought were interesting. I tried to spread out the targets around the planet, and to get a nice cross-section of science themes. The targeting specialist will now take this file and do additional checks before turning it over to JPL.
This is just the first round – we are doing this stage of planning in complete ignorance of what the other instruments on the spacecraft want to do during the same time. All of the instruments turn in files containing the observations that they want to take that will require rotating the spacecraft in some way. On Thursday, one of my coworkers at JPL will take all these files and see what can actually be scheduled and what conflicts with another instrument. (This is one of my usual jobs; someone else is doing it right now.) Once we have those results, we will try to squeeze in additional images that don’t involve turning the spacecraft (and therefore won’t interfere with anyone else’s observations).
There’s still about 2 weeks of work on this cycle left, but I’ve completed my first chunk of work, and I’m feeling pretty good about it. It is kind of amazing – for these 2 particular weeks (starting July 4), I get a huge amount of control over what HiRISE is doing. If I don’t like your target, it doesn’t get in the plan! I have already used my power to bump up the priority of a couple targets requested by some friends from grad school, and to yank some where we had a bunch of images already. Mua ha ha! (This is why we rotate the job between science team members…)
It’s pretty cool, but lots of painstaking work. Also, living out of a hotel for nearly three weeks kind of sucks.
Questions from a friend on the original post:
Q: How cool! How do you choose which targets are high priority or interesting? Aside from being friends with the requester, I mean. ;)Do you have some kind of measure of potential scientific gain or are you just picking areas here and there to diversify?
A: Since our images are like postage stamps (we’ve been operating for several years, and have covered ~1 % of the martian surface), we have a system to help sort through targets.
The requester enters a description of why he or she thinks the target is scientifically useful, and assigns it a science theme (i.e., volcanism, fluvial processes, polar geology, composition, etc). Each science theme has a HiRISE team member assigned to it. That person is responsible for going through all requests in their theme and prioritizing them. So by the time it gets to me, each target in the database has two pieces of information attached – why the requester wanted it, and how important the theme lead thought that target was in relation to the other targets in that science theme. I take all that into account when I make my selections. If I think a target looks interesting based on other datasets, my own experience, or the description in the database, I can override the theme lead’s prioritization. I also try to select a mix of science themes, so if we had a large number of high-priority volcanism targets but few high-priority gullies targets, I would try to balance that out by removing some of the volcano targets. I also showed favoritism to polar geology, since 1. that’s my field of study and 2. it is currently northern summer, the only time of year when we can get clear images of the north polar layered deposits.
Commentaires