Diana Hayes

🇨🇦🏳️‍🌈🏳️‍⚧️

Mars

Cloud Movies

We have been taking movies of clouds on Mars using the Curiosity rover’s Navigation Cameras (Navcams) since shortly after the rover landed in 2012. These movies can broadly be divided into two categories: Suprahorizon Movies (SHMs), which look out over the horizon; and Zenith Movies (ZMs), which look straight up above the rover.

A SHM taken on Sol 3063.
A ZM taken on Sol 3063.

These movies may look a bit fuzzy or noisy. This is because the clouds are typically sub-optical (not visible in the raw images). To enhance their visibility, we apply a method known as “mean-frame subtraction” (MFS). In this method, we determine the average value of every pixel across the entire movie, and then subtract that average from each frame. The average approximately represents everything in the image that doesn’t change over time, and by subtracting it, we leave behind everything that does change, primarily the clouds.

Figure 4 of Hayes et al. (2024), showing four frames of a ZM at different points in the processing pipeline. In the top row, we have the raw images. This was a particularly exceptional ZM, so the clouds are actually just about visible. In the middle row, we have applied MFS, which makes the clouds much more visible. In the bottom frame, we have further enhanced the clouds’ visibility by removing the highest and lowest 2% of values and stretching the remaining values to fill the full range.

We can use these movies to determine various properties of the clouds. My primary focus has been their opacity, or how thick they are. By tracking how the opacity changes over each sol and over the entire year, we can get insights into the dynamics of Mars’s atmosphere over Gale Crater.

We’re mostly interested in a period of the martian year known as the “Aphelion Cloud Belt” (ACB) season. This period spans a range of times surrounding aphelion, when Mars is furthest from the Sun. Due to the larger distance from the Sun, the atmosphere is generally colder. At the same time, it’s spring/summer in the northern hemisphere, which means that the north polar ice cap is sublimating away, releasing a lot of water vapour into the atmosphere. Because cold air cannot hold as much moisture as warm air, the atmosphere at this time of year becomes rapidly saturated with water vapour, making it much easier for clouds to form. Clouds of the ACB are generally observed between latitudes 10°S and 40°N. Curiosity is located around 4.5°S, so we can observe the southern edge of the ACB.

An illustration of Mars’s orbit around the Sun. The time of year is measured by the planet’s “solar longitude,” which is determined by dividing the orbit up into 360 degrees. The colours in the orbit indicate the season in the southern hemisphere (where Curiosity landed): green = spring, yellow = summer, red = fall, blue = winter. The cyan region indicates the range of solar longitudes during which we observe the ACB. The high eccentricity of Mars’s orbit is also obvious here, meaning that it’s much closer to the Sun at perihelion than at aphelion. This means that spring and summer in the southern hemisphere are much warmer than they are in the northern hemisphere, which is illustrated by the colours on the planet representing the intensity of sunlight on the surface.

The ACB has been observed to be a very repeatable feature, meaning that it doesn’t change much from year-to-year. Previous work looking at opacities from the first two Mars years of Curiosity’s mission suggested that cloud opacities over Gale in the morning are generally higher than those in the afternoon, similar to the pattern that has been seen in the ACB elsewhere.

We found that this apparent diurnal variability was actually an artifact of a bad assumption made in the original model that was used to derive the opacities. One of the key components in the model is the “scattering phase function,” which tells us how much light is scattered in each direction when the light interacts with a particle (such as dust or ice crystals) in the atmosphere. Previously, it had been assumed that the phase function was flat, with a single value at all scattering angles. While this assumption is good at large scattering angles (far from the Sun), the phase function increases rapidly at small scattering angles (close to the Sun). On average, movies taken in the morning are looking at smaller scattering angles than those in the afternoon, so this assumption was artifically increasing the morning opacity values.

Once we used a more plausible phase function in our model, we found that the ACB over Gale doesn’t actually change that much. Opacities are pretty constant over each sol, over the entire ACB season, and from one year to the next. We verified this with opacity measurements taken from orbital images, which agreed with our results.

Figure 16 of Hayes et al. (2024), illustrating our measured cloud opacities (the coloured points) over the five ACB seasons that we examined, compared with measurements from the MARs Color Imager (MARCI) instrument onboard the Mars Reconnaissance Orbiter (MRO). We can see that the two datasets agree reasonably well, indicating that cloud opacities over Gale don’t change very much over the course of the ACB season.
Figure 17 of Hayes et al. (2024), illustrating our measured cloud opacities (the coloured points) over the Mars Year 36 ACB season, compared with opacities derived from observations using the Emerates Exploration Imager (EXI) onboard the Emirate Mars Mission’s Hope orbiter (EMM). Unlike MRO, which is in a “sun-synchronous” orbit that observes each location at about the same time every sol, EMM is in an elliptical “supersynchronous” orbit, which allows for observations to be taken at all times of day. This allowed us to validate our observation of minimal opacity variability at different times of day (the bottom graph), as well as over the entire ACB season (the top graph).