Cost vs. Plausibility, Stingrays, and Lunar Spelunking

The approach of Davies and Wagner (2013) is a good one as far as SETI papers go, so I’ll start with a quick summary of the salient points.

The primary point of the paper is that a search of data from the Lunar Reconnaissance Orbiter should be performed, looking for anything out-of-place that indicates the presence of non-terrestrial artifacts (or NTAs, to borrow a phrase from Haqq-Misra and Kopparapu (2012)) or past non-terrestrial activity. The authors argue that the moon is a good place to search for artifacts for many reasons: it’s close and we have good, high-resolution data of it, the surface is unchanging (on a hundreds of millions of year timescale), and it’s tectonically inactive, so we don’t have to worry about the artifact’s signature being swamped by thermal/radioactive/magnetic processes from geological action (like we would have on the Earth).

Look at this cute lil’ orbiter

The authors then divide potential NTAs into four classes, based on assumptions that they openly admit are anthropocentric (which is refreshing, compared to some of the other papers we’ve read).

The first class is messages, things that are “deliberate” in catching our attention. One thing to keep in mind is how long the message might have been waiting there – the longer it needs to last, the harder it will be for us to find due to the trade-off of detectability and durability.

The second class is scientific instruments, which have a nice symmetric pro and con. Con: they wouldn’t’ve been made for us to find, so they may not be easy to spot or recognize. Pro: instruments need power supplies, and power supplies are more easily detectable (think solar panels or waste heat).

The third class is trash – things left over from prior expeditions and never cleared away – a category that humans are particularly good at. The authors make a case for searching in lava tubes – trash left there would be protected from asteroid impacts and could lay undisturbed for far longer than something on the surface. I’ll be the first to admit that lunar spelunking for alien artifacts sounds like the most epic job posting ever, but it probably isn’t realistic to expect that a search like that would occur any time soon, even if we had any reason to believe it would be successful.

The final class of NTA is “geo-engineering”, or scars on the moon’s surface left behind by some prior alien activity (mining? excavations? who knows). Features created by geo-engineering might be easier to spot with the data based on scale, but the difficulties come in trying to decide which features are natural vs. NTAs, and which features are even interesting in the first place.

At the end of the paper, having defined some idea of what we might be looking for, the authors give some examples of ways to search the already extant LRO photographic dataset for these features. I decided to organize and expand on their suggestions in the following table:

This table is specifically in reference to the problem of searching for NTAs in LRO photographs, but it could be easily generalized to any big-data SETI project, and even many big-data projects in general. I think this is a useful summary for thinking about the problem of big-data, and a good argument for why the multiple-pronged approach that was being tried by the authors is the way to go.

{Side note: I am a huge proponent for citizen science as a way to make scientific progress while educating and engaging the public. I participated in many citizen science projects in middle and high school, and led a Seafloor Explorer competition for 20 middle school students that classified the objects and wildlife in ~10,000 images of the Atlantic seafloor. The gallery below shows some of the images that my students got very excited about in the classifying process. Applying a citizen science strategy to SETI could be very useful…}

This slideshow requires JavaScript.

To end this post, I’m going to change gears and get a little philosophical for a moment. The authors make an interesting case for pursuing SETI in large, already existing databases – many SETI ideas are low cost and high potential reward projects and should be pursued based on cost over plausibility. I still don’t entirely know how I feel about that idea. Could that mentality be politically destructive for SETI in the current funding landscape, and should that matter if science is being done and progress is being made? Does it lend legitimacy to fringe-sounding ideas, like the genomic SETI concept that the authors mention (ex. this paper), or does spending a little bit of effort to test and debunk these ideas actually make the field better in the long run? Are we uncertain enough about the nature of ETI that disregarding plausibility and just prioritizing by cost is actually a more logically consistent way to go about the search?

I don’t have answers, but I think these are interesting questions and we should keep mulling over them.