Overflow —

Ancient lake sites suggest lots of precipitation on Mars

How much water did it take to fill a lake bed on Mars?

A simulated view of Gale Crater Lake, measuring about 150km across, on Mars about 3 billion years ago.
Enlarge / A simulated view of Gale Crater Lake, measuring about 150km across, on Mars about 3 billion years ago.
NASA/JPL-Caltech/ESA/DLR/FU Berlin/MSSS

Mars clearly once had a lot of water—there are simply far too many features that clearly formed in a watery environment for that to be a matter of debate. What's less clear is how much of that water was liquid and for how long. While some features clearly indicate that liquid water was present for a long time, others likely formed under glacial ice.

It's not clear whether the differences are a matter of timing—a wet period followed by an icy one, for example—or due to regional differences in Mars' climate. It's difficult to tell in part because we can't get climate models of Mars to produce a climate that's wet enough for long enough to form a lot of watery features.

To try to put some constraints on what the ancient Martian climate might have looked like, a team of planetary scientists decided to take a careful look at some of the once-watery features identified on the surface of the red planet. Timothy Goudge, Caleb Fassett, and Gaia Stucky de Quay (yes, that's a planetary scientist named Gaia) identified a series of lakes, and used the features of the lakes to put some constraints on the precipitation that fed them.

Top to bottom

How does a lake tell us about precipitation? By studying the elevation of the land around the basin that the lake lies within, it's possible to figure out how large an area it once drained. Given that and the volume of the lake, it's possible to determine how much precipitation was needed to fill it. But how do you determine the volume of a lake that hasn't existed for billions of years?

While the researchers can't determine the exact volume, they can set limits. A number of the lakes were drained by water flowing out, indicating that their volume must have at least reached that level, providing a lower bounds on lake volume. Others never drained, thus providing a maximum possible volume for this lake—any higher than the surrounding terrain and it would have overflowed. Overall, they analyzed 54 of the former type of lake and another 42 of the latter.

The researchers then did a simple balancing between the precipitation and evaporation rate, with the math being agnostic about whether the precipitation formed as rain or snow. The evaporation rate would also depend on how dry or humid the climate was, with dry climates setting an upper limit on how much water must be input and wetter ones allowing less precipitation to accomplish the same thing.

The researchers then figured out how large an excess precipitation event would have to be to overcome the evaporation and fill the basin. This isn't the sort of steady-state supply of water needed to balance out evaporation and possibly feed an outlet stream; rather, it's the amount of water that would need to flow in excess of that in order to fill the lakes to somewhere between these upper and lower bounds.

The researchers call these "runoff episodes" and refer to their duration as "unconstrained." All they can say is that "this episode must have been sufficiently continuous and supplied enough water to fill and breach open-basin lakes [those with runoff streams] but not closed-basin lakes."

Within limits

Calculating these upper and lower bounds for different lakes produces a frequency distribution for both the upper and lower bounds of the precipitation needed to produce appropriately sized runoff events. And those tell us a couple of significant things. The first is that there was lot of precipitation involved in these events. The lower bound is a bit over four meters, and the upper bound is 159 meters. Again, this was over a completely undefined period of time, but it's still a significant amount of precipitation.

The other thing that's very clear is that the precipitation wasn't evenly distributed—Mars didn't have a single climate. The researchers were able to identify areas that were likely to have received greater or lesser amounts of precipitation.

But a number of key things still aren't clear. Again, one is how long these events took—the authors call them a "quasi-continuous runoff episode." Another is how many of them there were. While we have some data for Gale Crater, where we happen to have dropped a rover, it's generally not clear how long most of these basins remained filled with water. Finally, we don't know whether the precipitation fell as rain or whether some or all of it was snow that then washed into the basins during periods of temporary warmth.

Still, by providing some limits on what must have been going on early in Mars' history, the study provides some constraints that climate modelers will need to meet as they try to understand the red planet's past. With enough of these limits, it will be easier for a clear picture of the distant past to emerge.

Geology, 2020. DOI: 10.1130/G47886.1  (About DOIs).

Channel Ars Technica