Charissa Campbell

Dust devils are believed to be a significant contributor to the martian global dust cycle, supplying as much as 50% of dust to the atmosphere annually. However, their in situ meteorological properties remain poorly known, specifically horizontal and vertical wind speeds. We use Mars Science Laboratory Curiosity rover Navigation Camera images to track image‐to‐image changes in dust devil heights as a proxy for their vertical wind speeds. Across 54 dust devils, we find an average vertical wind speed of ∼1.79 m s ⁻¹ with a standard deviation of 0.39 m s ⁻¹ , nearly identical to the 1.8 m s ⁻¹ found in Gusev Crater by the Spirit rover (Greeley et al., 2006, https://doi.org/10.1029/2006je002743 ). Using typical dust devil opacities and occurrence rates in Gale Crater, we find the areal dust mass flux into the atmosphere in Gale Crater to be 3.5*10 ⁻¹¹ kg s ⁻¹ m ⁻² , 1–2 orders of magnitude below that of other regions on Mars.

Dust storms near Mars' South Pole during the perihelion and summer solstice seasons ( L _s ∼ 250–290°) are phenomenologically distinct from martian dust storms in other locations and seasons. While they have previously been observed to increase atmospheric dust loading and warm the south polar atmosphere, they have no notable impact on the middle latitudes, tropics, or northern hemisphere. Here, we use a combination of multiple remote sensing instruments, atmosphere reanalyses, and general circulation modeling to study the evolution of dust storms near the Martian South Pole during Mars Year 30 (a prototypical year for such storms). Dust lifting preferentially occurs in the eastern hemisphere (180°–320°E) and follows the retreating seasonal CO ₂ polar cap to higher latitudes as the season progresses—implicating the strong cap edge thermal forcing and katabatic flows as well as surface dust availability in driving dust lifting. Dust remains confined to the high southern latitudes during this season with strong diurnal variability in both latitudinal and vertical extents. Dust is entrained in the southern circumpolar jet stream, creating filamentation and longitudinal mixing and controlling the flow of dust back onto the polar cap. The diurnally varying atmospheric circulation limits the mixing of dust to lower latitudes and prevents a global thermal response.

. INTRODUCTIONThe Mars Science Laboratory (MSL) is located in Gale Crater (4.5°S, 137.4°E), and has been performing cloud observations for the entirety of its mission, since its landing in 2012 [eg. 1,2,3]. One such observation is the Phase Function Sky Survey (PFSS), developed by Cooper et al [3] and instituted in Mars Year (MY) 34 to determine the scattering phase function of Martian water-ice clouds. The clouds of interest form during the Aphelion Cloud Belt (ACB) season (Ls=50°-150°), a period of time during which there is an increase in the formation of water-ice clouds around the Martian equator [4]. The PFSS observation was also performed during the MY 35 ACB season and the current MY 36 ACB season.Following the MY 34 ACB season, Mars experienced a global dust storm which lasted from Ls~188° to Ls~250° of that Mars year [5]. Global dust storms are planet-encircling storms which occur every few Mars years and can significantly impact the atmosphere leading to increased dust aerosol sizes [6], an increase in middle atmosphere water vapour [7], and the formation of unseasonal water-ice clouds [8]. While the decrease in visibility during the global dust storm itself made cloud observation difficult, comparing the scattering phase function prior to and following the global dust storm can help to understand the long-term impacts of global dust storms on water-ice clouds.2. METHODSThe PFSS consists of 9 cloud movies of three frames each, taken using MSL’s navigation cameras, at a variety of pointings in order to observe a large range of scattering angles. The goal of the PFSS is to characterise the scattering properties of water-ice clouds and to determine ice crystal geometry. In each movie, clouds are identified using mean frame subtraction, and the phase function is computed using the formula derived by Cooper et al [3]. An average phase function can then be computed for the entirety of the ACB season.Figure 1 shows the temporal distributions of PFSS observations taken during MYs 34 and 35. We aim to capture both morning and afternoon observations in order to study any diurnal variability in water-ice clouds.3. RESULTS AND DISCUSSIONThere were a total of 26 PFSS observations taken in MY 35 between Ls~50°-160°, evenly distributed between AM and PM observations. Typically, times further from local noon (i.e. earlier in the morning or later in the afternoon) show stronger cloud features, and run less risk of being obscured by the presence of the sun. In all movies in which clouds are detected, a phase function can be calculated, and an average phase function determined for the whole ACB season. Future work will look at the water-ice cloud scattering properties for the MY 36 ACB season, allowing us to get more information about the interannual variability of the ACB and to further constrain the ice crystal habit. The PFSS observations will not only assist in our understanding of the long-term atmospheric impacts of global dust storms but also add to a more complete image of time-varying water-ice cloud properties.

The altitudes of Martian water-ice clouds above Gale crater are being calculated through the Cloud Altitude Observation (CAO) taken by the Mars Science Laboratory (MSL, Curiosity). Typically, a mission needs a lidar onboard to calculate altitude, but this observation utilizes local topography – Aeolis Mons – to follow shadows and determine an absolute wind velocity. Comparing this velocity to the angular wind velocity found in a paired vertical movie determines the cloud’s altitude. This method is useful for the community when a mission doesn’t have a lidar but nearby terrain to observe shadows. During Mars Year (MY) 35, 28 observations were collected and will be compared to Campbell et al. [1] results from MY 34 to determine any changes. IntroductionThe Aphelion Cloud Belt (ACB) season is a yearly equatorial cloud belt caused by cooler atmospheric temperatures. Occurring between solar longitude (Ls) 45°-150°, water-ice clouds are typically visible during two times of each sol, early morning and late afternoon [2,11]. Since 2012, MSL has been observing the ACB season in Gale crater (4.5°S, 137.4°E) with a variety of cloud observations to study seasonal and diurnal changes of different parameters such as seasonal and diurnal changes [1,8,9], cloud altitude [1], opacity [8,9,10] and ice crystal geometry [3].One parameter, altitude, is of interest to the atmospheric science community to determine the cloud’s relation to the Planetary Boundary Layer (PBL). Orbiters are able to constrain cloud altitudes through limb measurements but observations within 10 km of the surface are hampered by dust opacity [7]. This is difficult for studying the PBL in Gale crater from orbit as the PBL is very shallow during the ACB season, ranging between 2-3 km [5,12]. From the surface, cloud altitudes could be measured with a lidar [4, 13], but only a single Mars mission has carried this instrument and therefore we must employ other methods to find altitude.As described in Campbell et al. [1], a new atmospheric observation was implemented on MSL during the MY 34 ACB season. Typically used for dust-devil searching, the Navigation Camera (Navcam) took a movie pointed at Aeolis Mons observed shadows moving along the mountain cast by clouds above. These shadows were georeferenced with a digital terrain model and the distance travelled was used to determine an absolute wind velocity. Angular wind velocity is already determined using a vertically pointed movie that follows how clouds move with respect to the camera’s field of view [1]. When these two movies are paired together, the comparison between velocities is used to calculate altitude via trigonometry. Therefore, this paired movie combination facilitates a way to directly calculate cloud altitudes at Gale crater for the first time. For the remainder of MY 34 season, these observations were paired whenever possible and resulted in 9 detections of clouds and their shadows, with the results reported in [1].Following this success, a new MSL observation was designed, optimized, and implemented to observe water-ice cloud shadows and deduce the water-ice cloud altitudes for the MY 35 ACB season and named the Cloud Altitude Observation (CAO). Cloud Altitude ObservationThe CAO is performed in the ACB season on a weekly cadence. It consists of two movies – Cloud Shadow Movie (CSM) and Zenith Movie (ZM) – taken one after the other by the Navcam. Both are made up of eight 1022x1022 pixeled images and span 240 seconds each. The CSM (Figure 1, top) points directly at Aeolis Mons to capture shadow motion which is used to determine an absolute wind velocity. The ZM (Figure 1, bottom) points directly above the rover and is used to determine an angular velocity based on the camera’s field of view. Comparing these two velocities allows the altitude of the clouds to be calculated.Results and DiscussionA total of 28 observations were collected between Ls 47°-163° and will be compared to MY 34 results [1] to determine any seasonal changes. Processing and analysis of these observations are ongoing, but the entire MY 35 ACB season's data will be processed and analyzed by the time of the conference.The MY 35 ACB season is unique in that it followed a Global Dust Storm (GDS) that occurred from Ls 188°-250° [6]. These storms are large enough to envelope the entire planet and affect the local and global environment. MSL has the unique ability to continue operations during a GDS and detected dust optical depths of up to an order of magnitude larger than normally observed [6]. Studying how the cloud altitudes change before/after the GDS will help determine if there are any long-term effects to the water-ice clouds. Once the MY 36 ACB season concludes, it will be added to the current data set to further determine any seasonal changes.References[1] Campbell et al., Planet. Space Sci. 182. 104785. 2020.[2] Clancy et al., Icarus. 122, 36-62. 1996.[3] Cooper et al., P&SS. 168, 62-72. 2019.[4] Dickinson et al., Geophys. Res. Lett. 37, L18203. 2010.[5] Guzewich et al., JGR-Planets, 122, 2779–2792. 2017.[6] Guzewich et al., Geophys. Res. Lett., 46, 71-79. 2019.[7] Kleinbohl et al., JGR-Planets, 114, E10006. 2009.[8] Kloos et al., Adv Sp Res. 58, 1223-1240. 2016.[9] Kloos et al., JGR-Planets. 123, 233–245. 2018.[10] Moores, et al., Adv. Space Res. 55, 2217–2238. 2015b.[11] Tamppari et al., JGR-Planets. 108, 5073. 2003.[12] Tyler, Barnes, Int. J. Mars Sci. Explor. 8, 58–77. 2013.[13] Whiteway et al., Science. 325. 68. 2009.

The Global Dust Storm of Mars Year 34 was observed by the Mars Science Laboratory (MSL) in Gale crater, and many atmospheric effects were seen during its duration. Using atmospheric observations taken by MSL, water-ice cloud opacity and scattering phase function are calculated, allowing comparison of these quantities before and after the dust storm. The average phase function was calculated for the Mars Year 34 Aphelion Cloud Belt (ACB) season, before the global dust storm, and will be calculated for the ACB season following. Opacity calculations are computed throughout the Martian year. The results of these calculations will indicate any impacts the global dust storm may have on Martian water-ice clouds.1. INTRODUCTIONThe Martian atmosphere has long been the subject of study, both in situ and remotely, and as a result we are able to observe its dynamic nature. Water-ice clouds are one such dynamic feature, and their similarities to Earth-based clouds provide an opportunity to learn more not just about Martian weather but our own climate processes.The Mars Science Laboratory (MSL) in Gale Crater has been observing cloud activity in the region for the entirety of its mission to date. Cloud movies captured by MSL can be analysed in order to determine optical parameters such as opacity[1] and scattering phase function[2]. The most important time of year for cloud activity around Gale Crater is the Aphelion Cloud Belt season, which tends to peak between solar longitudes (Ls) 60-100° and is characterised by an increase in equatorial cloud formation[3].MSL has been in Gale Crater for nearly four Martian Years, and as a result has been able to observe the time-varying nature of the Martian climate on daily, seasonal, and annual scales. The presence of a global dust storm in 2018 (Mars Year 34) gave further opportunity to study not only the event itself, but its lasting effect on the Martian atmosphere.2. BACKGROUND2.1 The Mars Year 34 Global Dust StormThe global dust storm of Mars Year (MY) 34 was a planet encircling dust storm which lasted from about Ls~188° to Ls~250°[4]. Global dust storms typically occur every 3-4 years[5] and significantly impact the atmosphere, causing temperature and pressure fluctuations[4] and an increase in water vapour in the middle atmosphere[6]. While the decrease in visibility made it difficult to observe clouds during the global dust storm, a before-and-after comparison of cloud opacity and scattering phase function provide understanding about longstanding atmospheric effects of global dust storms.2.2 MSL Atmospheric MoviesThe navigation cameras onboard MSL take two types of movies year-round, Zenith Movies (ZM) and Suprahorizon Movies (SHM), and a third observation during the ACB season, the Phase Function Sky Survey (PFSS). ZM and SHM are both 8 frame movies taken over 6 minutes, with ZMs pointing to the Zenith and SHMs pointing just above the crater rim. The PFSS is a mosaic of 9 3-frame movies at a variety of pointings making a dome surrounding the rover. In order to better identify clouds, all three types of movies undergo mean frame subtraction, a process by which an average frame is subtracted from each frame of the movie, leaving only the time variable component. This can be seen in Figure 1.3. METHODSZM and SHM can be analysed to determine cloud optical thickness, which has been computed for the first two Martian Years of MSL’s mission (Ls=160° of MY 31 to Ls=160° of MY 33)[7]. After mean frame subtraction, a radiance map (Figure 2) is made of a single frame of the movie, which is examined manually for a region containing cloud and empty sky, giving a variation in spectral radiance (Iλ,VAR) which is then used to compute the optical thickness. During the ACB season, the high-cloud formula, derived by Kloos et al, is used to estimate opacity, which assumes clouds are high, optically thin, and composed of water-ice crystals[1].The PFSS was first instituted in MY 34, and an average scattering phase function determined for that year[2]. As with optical thickness, Iλ,VAR is determined from a radiance map for each movie and used to compute the phase function, using the formula derived by Cooper et al[2]. With the MY 35 ACB season ending in February of 2020, there is now another complete set of 26 PFSS observations, giving 234 total movies.4. RESULTS AND DISCUSSIONCurrently ZM and SHM have been processed and have radiance maps from sol 1706 to sol 2748, a total of 466 movies. All of these must be checked manually for quality, and to identify high and low radiance points. This process is ongoing. Following the calculation of optical thickness for all cloud movies, the catalogue of MSL cloud opacities will be extended from two Martian years to four.In addition, the 234 PFSS movies must likewise be checked, a process that is nearing completion at time of writing. By the time of the conference, it is anticipated that an average phase function for MY 35 will be calculated, allowing comparison with that of MY 34. Due to the larger size of the opacity dataset, opacity calculations are anticipated to take more time.These observations will allow us to better understand not only the lasting atmospheric effects of global dust storms, but also will extend our understanding of variability in water-ice cloud optical properties over larger timescales.REFERENCES[1] Kloos, J.L, et al, Adv. Space Res., 2016.[2] Cooper, B.A., et al, P&SS, 2019.[3] Wolff, M.J., et al, JGR, 1999.[4] Guzewich, S.D. et al, Geophys. Res. Lett., 2019.[5] Zurek, R.W., and Martin, L.J., JGR, 1993.[6] Aoki, S.A., et al, JGR-Planets, 2019.[7] Kloos, J.L. et al, JGR-Planets, 2018.

We examine 3 yr of phase-function observations of water-ice clouds taken during the Aphelion Cloud Belt season by the Mars Science Laboratory (MSL). We derive lower-bound single-scattering phase functions for Mars years (MYs) 34, 35, and 36, over a range of scattering angles from 45° to 155°, expanding on the MY 34 phase function previously derived from MSL observations using the same method. We also modify the procedure used for MY 34 to make use of cloud opacity values derived from other MSL observations, often taken in conjunction with the phase-function observations. From these, we see little variability, both interannually and diurnally in the phase function at Gale Crater. We use our derived phase functions to attempt to constrain a dominant ice-crystal geometry by fitting a two-term Henyey–Greenstein function. In comparing to HG functions of Martian dust and modeled water-ice crystals, we see agreement especially with droxtal water-ice crystals, dust at Gale crater, and irregular volcanic glasses. This could be indicative of crystals composed of some irregular shape.

We update the record of cloud opacity observations conducted by the Mars Science Laboratory (MSL) Curiosity rover to cover the first five Mars Years (MYs) of the mission (L _s = 160° of MY 31 to L _s = 160° of MY 36). Over the three MY period that we add to the previously analyzed two MY record, we achieve good diurnal coverage between 07:00 and 17:00 with nearly 1200 new observations. We derive a new scattering phase function for the clouds of the Aphelion Cloud Belt (ACB) using results from the Zenith and Suprahorizon movie data sets. Our phase function is generally smooth and featureless, which is consistent with the overall lack of atmospheric optical phenomena on Mars aside from a single instance of an observed halo. Applying our new phase function to the data, we find that there is very minimal variability in the ACB's opacity, either diurnally, intraseasonally, or interannually, noting that our observations are only sensitive to ice clouds and cannot detect any ice hazes that may be present over Gale. This contrasts with previous results, which observed a 57% difference in the opacity of morning and afternoon clouds in MY 33. The MY 33 results now appear to be an outlier, not replicated at any point during the MSL mission. We conclude that the higher morning opacities in MY 33 were a consequence of an incomplete understanding of the nature of the scattering phase function close to the Sun.

Using a variety of instruments onboard the Perseverance Rover, a substantial amount of data has been collected and analyzed to understand the environment and atmosphere of the landing site, Jezero Crater. A vital component in understanding this unique environment is studying the local and regional variations of water over diurnal and seasonal timescales. We specifically study the variations of water ice in the atmosphere by observing cloud activity over Jezero Crater using data collected by the Navigation Cameras, NavCam, on board the Mars2020 Rover, Perseverance. IntroductionWater Ice clouds on Mars have been observed for a long time from both the surface and from orbit to understand the role of water in the atmosphere. Using Viking era data, Tamppari (2000, 2003) observed water ice clouds with peak activity seen during the northern spring and summer time. Opportunity Rover acquired images of water ice clouds during the Aphelion Cloud Belt (ACB) season with peak activity seen from Ls 50o to 115o (Lemmon et al., 2015). From the past activity, we have seen that between the solar longitude (Ls) of 45o to 150o, the Aphelion Cloud Belt (ACB) forms every Mars Year (MY) in the equatorial region (10oS–30oN). (Tamppari et al., 2000) (Smith et al., 2004)  Furthermore, the Surface Stereo Imager (SSI) on-board the Phoenix Lander showed a large variety of cloud types over the course of the 151-sol mission (Moores et al., 2010). The navigation camera, NavCam aboard the Mars Science Laboratory (MSL, Curiosity) rover has observed cloud activity at Gale crater through cloud movies and surveys. The diurnal and seasonal patterns were comparable with previous observations. The peak cloud activities were observed in the morning/afternoon time and around the ACB period. (Kloos et al., 2018) They have also been reported to leave a thermal signature at night on Gale’s surface (Cooper et al. 2021). NAVCAM dataAt Jezero Crater, water ice clouds have been observed through cloud movies and cloud surveys collected using the Perseverance NavCam instrument that collects colour stereo images with a 96ox73o field of view at 0.33 mrad/pixel. (Maki et al., 2020) Compared to the NavCam on Curiosity, the NavCam on Perseverance holds the capability of imaging in colour which may provide advantages in highlighting water ice clouds in the atmosphere.  Cloud surveys are single images taken facing the horizon, generally twice a week, at various times of the day. Similarly, cloud movies are collected by taking 8 frames of images looking towards the horizon with an interval of 15 seconds, producing a movie. A combination of cloud surveys and movies collected by the mission are used to study water ice clouds. Data ProcessingVarious image processing techniques are used on these data sets to emphasize the presence of water ice clouds in the atmosphere. For example, the NavCam Cloud movies can be processed using an imaging technique known as Mean Frame Subtraction (MFS) to enhance cloud features (Campbell et al., 2020) (Moores et al., 2015). This technique takes the average frame of the whole movie and subtracts it from each frame (Moores et al., 2015). This reveals the time-variable component, displaying the cloud movement within the movie. Figure 1 shows an example of raw (left) versus MFS processed frames (right) of a NavCam cloud movie. In addition to these processed images and movies, we have explored techniques such as two-dimensional fast Fourier transforms to recognize patterns in clouds. Figure 1: Comparison between raw and Mean Frame Subtraction frames. The raw frame (left) only shows faint clouds but when the Mean Frame Subtraction method is applied (right), the cloud features become visible. We will be presenting the water-ice cloud activity at Jezero Crater with details on various image processing techniques used to determine the characteristics of Martian clouds. The full data set from the start of the Mars2020 mission up to now will be analyzed to determine any seasonal patterns in the cloud activity seen at Jezero.  BibliographyCampbell et al., 2020. Planet. Space Sci. 182 104785, http://dx.doi.org/10.1016/j.pss.2019. 104785.Cooper, B., et al., 221 JGR: Planets, 126(12), doi:10.1029/2020JE006737, 2021Kloos et al., 2018. Geophys. Res. Planets. (2018)123, 233–245. https://doi.org/10.1002/2017JE005314Lemmon et al., 2015. Icarus 251, 96–111. https://doi.org/10.1016Moores et al., 2010. J. Geophys. Res. Planets. https://doi.org/10.1029/ Maki et al., 2020. Space Science Reviews https://doi.org/10.1007/s11214-020-00765-9Moores et al., 2015. Adv. Space Res. 55 2217–2238, http://dx.doi.org/10.1016/j.asr.2015.02.007.Smith, M. D. 2004, Icarus, 167, 148–165. https://doi.org/10.1016/j.icarus.2003.09.010Tamppari et al., 2000, J. Geophys. Res., 105, E2, 4087–4107.  https://doi.org/10.1029/1999JE001133Tamppari et al., 2003, J. Geophys. Res., 108, https://doi.org/10.1029/2002JE001911