Twilight Mesospheric Clouds in Jezero as Observed by MEDA Radiation and Dust Sensor (RDS)

1. Introduction The H2O and CO2 ice clouds on Mars are primary constituents studied for understanding the past and present climate of the planet (Forget & Pierrehumbert,1997; Montmessin etal.,2004). Cloud particles can affect the energy balance of the planet (e.g., Wolff etal.,2019), and thus the atmospheric dynamics, as well as influence the vertical distribution of dust particles through dust scavenging. The dust scavenging by H2O clouds has critical consequences in the water cycle of the planet; for example, regions in the atmosphere with insufficient quantity of dust particles (or condensation nuclei) can inhibit the formation of H2O clouds (Määttänen etal.,2005; Montmessin etal.,2002), and thus reach water vapor concentrations in excess of saturation (Maltagliati etal.,2011; Navarro Abstract The Mars Environmental Dynamics Analyzer instrument, on board NASA's Mars 2020 Perseverance rover, includes a number of sensors to characterize the Martian atmosphere. One of these sensors is the Radiation and Dust Sensor (RDS) that measures the solar irradiance at different wavelengths and geometries. We analyzed the RDS observations made during twilight for the period between sol 71 and 492 of the mission (Ls 39°–262°, Mars Year 36) to characterize the clouds over the Perseverance rover site. Using the ratio between the irradiance at zenith at 450 and 750nm, we inferred that the main constituent of the detected high-altitude aerosol layers was ice from Ls=39°–150° (cloudy period), and dust from Ls 150°–262°. A total of 161 twilights were analyzed in the cloudy period using a radiative transfer code and we found: (a) signatures of clouds/hazes in the signals in 58% of the twilights; (b) most of the clouds had altitudes between 40 and 50km, suggesting water ice composition, and had particle sizes between 0.6 and 2µm; (c) the cloud activity at sunrise is slightly higher that at sunset, likely due to the differences in temperature; (d) the time period with more cloud detections and with the greatest cloud opacities is during Ls 120°–150°; and (e) a notable decrease in the cloud activity around aphelion, along with lower cloud altitudes and opacities. This decrease in cloud activity indicates lower concentrations of water vapor or cloud condensation nuclei (dust) around this period in the Martian mesosphere. Plain Language Summary During twilight, ground-based observations of the irradiance allows the detection and characterization of high-altitude clouds (above 30–35km). Because the sun is at or below the horizon, the cloud layers reflect the direct light that only reaches the higher parts of the atmosphere, producing an increase in the sky brightness with respect to the cloud-free scenario. Moreover, the decrease in the intensity with the solar zenith angle highly depends on the cloud altitude and density. Using observations made by the Radiation and Dust Sensor, part of the instrument Mars Environmental Dynamics Analyzer on board Perseverance rover, we present here a study of the twilight clouds detected at the Perseverance landing site for the first 490 sols of the mission (Mars Year 36). By modeling the irradiance at 450 and 950nm with radiative transfer simulations, we constrained the cloud altitude, opacity, and particle radius. The number of twilights analyzed allowed us to study the seasonal trend in the cloud activity. During the cloudy period, Ls 39°–150°, we find a significant decrease in the cloud activity above 30–35km around aphelion (Ls∼70°). This implies that the seasonal distribution of clouds above 30–35km differs from that observed at lower altitudes. TOLEDO ETAL. © 2023 The Authors. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. Twilight Mesospheric Clouds in Jezero as Observed by MEDA Radiation and Dust Sensor (RDS) D. Toledo1 , L. Gómez1 , V. Apéstigue1 , I. Arruego1 , M. Smith2 , A. Munguira3 , G. Martínez4 , P. Patel5 , A. Sanchez-Lavega3 , M. Lemmon6 , L. Tamppari5 , D. Viudez-Moreiras7 , R. Hueso3 , A. Vicente-Retortillo7 , C. Newman8 , R. Lorenz9 , M. Yela1 , M. de la Torre Juarez5 , and J. A. Rodriguez-Manfredi7 1Instituto Nacional de Técnica Aeroespacial (INTA), Madrid, Spain, 2NASA Godard Space Flight Center, Greenbelt, MD, USA, 3Universidad del País Vasco UPV/EHU, Bilbao, Spain, 4Lunar and Planetary Institute, Universities Space Research Association, Houston, TX, USA, 5Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA, 6Space Science Institute, Boulder, CO, USA, 7Centro de Atrobiología (INTA-CSIC), Torrejón de Ardoz, Madrid, Spain, 8Aeolis Research, Chandler, AZ, USA, 9Johns Hopkins Applied Physics Laboratory, Laurel, MD, USA Key Points: • Most of the cloud detected at twilight between sol 71 and 492 of the Mars 2020 mission (Ls 39°–262°) occurred at altitudes between 40 and 50km • Around aphelion (Ls∼70°) we found the minimum in cloud activity and lower cloud opacities • The cloud activity at sunrise is slightly stronger than at sunset and this is likely due to the lower temperatures Correspondence to: D. Toledo, [email protected] Citation: Toledo, D., Gómez, L., Apéstigue, V., Arruego, I., Smith, M., Munguira, A., etal. (2023). Twilight mesospheric clouds in Jezero as observed by MEDA Radiation and Dust Sensor (RDS). Journal of Geophysical Research: Planets, 128, e2023JE007785. https://doi. org/10.1029/2023JE007785 Received 11 FEB 2023 Accepted 22 JUN 2023 10.1029/2023JE007785 RESEARCH ARTICLE 1 of 18 Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 2 of 18 etal.,2014). Although other mechanisms may be responsible for the existence of supersaturation on Mars as well (Fedorova etal.,2020), it is well established that the formation of clouds limits the concentration of water vapor to values below saturation, and this partially controls the amount of water vapor that can be transported to the higher parts of the atmosphere, where the water can be photodissociated into its lighter components H and O. One of the two major cloud regimes on Mars is the aphelion cloud belt (ACB) (Clancy etal.,1996) occurring in the equatorial regions between ∼10°S and ∼30°N and during the northern spring and summer (Ls∼0°–180°). The other main cloud regime is the hoods over the polar cap in both hemispheres (Benson etal.,2010,2011) during late summer and the whole winter (these clouds are not studied in this work). As reported in many previous works, both events are observed every Martian year (MY). Although H2O and CO2 clouds or hazes have been observed and studied from the surface of Mars, directly through images or indirectly through aerosol opacity measurements (e.g., Lemmon et al., 2015; Lorenz et al., 2020; P. H. Smith & Lemmon, 1999), the longest record of cloud events comes from instrumentation onboard orbiters (e.g., Määttänen etal.,2010; McConnochie etal.,2010; Sánchez-Lavega etal.,2018; Tamppari etal.,2003; Wang & Ingersoll,2002; Wolff etal.,2022). When these orbiter observations are made at limb-viewing geometry, information on the cloud vertical profiles can be derived (e.g., Rannou etal.,2006; M. D. Smith etal.,2013). In these particular cases the cloud frequency-of-occurrence or properties (e.g., opacity, particle radius) can be studied as a function of the altitude. On the other hand, if the orbiter observations are obtained at nadir-viewing geometry, in general the cloud vertical profiles cannot be derived and the total ice column opacity is provided (e.g., Giuranna etal.,2021; M. D. Smith,2009). While orbital observations provide a more complete global coverage, landed observations represent a critically important component to: (a) cross validate the orbital observations and retrievals; (b) study the diurnal and seasonal variations of the cloud activity without the impact of the orbiter spatial and temporal sampling; and (c) investigate the atmospheric context in which the clouds were formed (if observations of meteorological time series are available). On 18 February 2021, the Mars 2020 rover Perseverance successfully landed in Jezero crater (latitude 18.44°N and longitude 77.45°E). To provide meteorological context for other observations and for future human exploration, Perseverance carries the Mars Environmental Dynamics Analyzer (MEDA) (Rodriguez-Manfredi etal.,2021) instrument, which includes a set of sensors: two wind sensors to infer wind direction and speed, five thermal sensors at different locations and heights (ATS), an infrared radiometer to measure ground and atmospheric temperature as well as atmospheric IR fluxes and reflected solar fluxes (TIRS), a relativity humidity sensor, a pressure sensor, and the Radiation and Dust Sensor (RDS) that measures the solar radiation at different wavelengths ranges from the UVA to the near infrared (Rodriguez-Manfredi etal.,2021,2023). In this paper we focus on RDS observations at twilight, when the solar zenith angle (SZA) is between 90° and 98°, to detect and characterize high-altitude clouds (above ∼30km) for the first 492 sols of the mission (MY 36). We briefly describe the RDS in Section2, as well as the observations, the principle of measurement and the radiative transfer (RT) modeling. In Section3, we present time series of high-altitude aerosol layers (ALs) detected during twilight, the cloud retrievals and main results. 2. Observations and Radiative Transfer Modeling 2.1. RDS Instrument RDS measures the solar irradiance at different spectral wavelengths and incident geometries. It is comprised of two sets of photodetectors (RDS-DP) and a camera pointing at zenith (RDS-SkyCam). The first set of photodetectors, the Top channels, corresponds to eight zenith-pointed detectors which cover the light spectrum from UVA to Near IR (Top-1 to Top-8: 255, 259, 250–400, 450, 650, 750, 190–1,100, and 950nm). Most of the Top detectors use interferential filters and mechanical masks (Apestigue etal.,2022) to constrain their field of view to ±15° zenith angle, while the Top-7 channel covers the full sky from 0° to 90° zenith angle and for all azimuth angles. The second set corresponds to the eight Lateral channels, which are pointed sideways at 20° (except Lat-8, which is 35°) above the rover deck and are all at 750nm. The Lat-1 channel is blinded to study the photodetector performance degradation. In this work only the observations made by the Top sensors will be used. In general, MEDA sampling is set at 1Hz with all sensors operating for blocks of 1hr and 5min. The disposition of the blocks along the day are selected for each sol based on a cadence that alternates even and odd hours, and the duration and number of block sometimes change depending on power availability and data volume constraints. For this reason, not all the twilights are covered by MEDA. Because of the low levels of irradiance expected during this 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 3 of 18 time of the day, these RDS observations are acquired with an extra 40 gain factor (Apestigue etal.,2022), which is activated when SZA≥90°. Note that because the RDS gain factor was not activated until sol 70, our analysis does not cover the first sols of the mission. 2.2. Principle of Measurement for the Detection of Clouds Since high-altitude ALs (e.g., H2O clouds or detached dust layers) imply an increase in the irradiance during twilight, clouds can be detected by looking at the evolution of the RDS Top observations at SZAs between 90° and 98° (∼30min long). Moreover, during this period only the higher parts of the atmosphere receive Sun direct light (as for altitudes 𝐴𝐴𝐴 R ( 1+tan 2(90◦− SZA) )0.5 , where R is the radius of the planet, the direct light intersects the planet surface), making the variation of the irradiance with SZA very sensitive to cloud properties such as the altitude or the number density. As indicated in Toledo, Rannou, Pommereau, Sarkissian, and Foujols(2016), this technique allows detecting clouds with very low opacities (subvisual opacities), as the pathway of sunlight in a horizontally homogeneous AL of h geometrical thickness is enhanced by a factor >1/sin(SZA−90°). Figure1 shows, as an example, RDS signals measured by Top-4 (450±10nm) and Top-8 (950±10nm) sensors at Figure 1. The upper panels show a comparison between Radiation and Dust Sensor (RDS) observations at 450 (Top 4) and 950nm (Top 8) made under cloud-free conditions (sol 99) and under the presence of clouds (sol 271). The presence of the clouds result in an increase in the irradiance (indicated with the black arrows). Note that each signal was normalized by the signal value at solar zenith angle=90°. By doing so, we diminish the impact of the dust opacity and particle radius on the RDS signals, and make the comparison between signals easier to interpret. As we will see in the following section, the normalization of the signals also allow us to reduce the number of free parameters in the radiative transfer analysis. The presence of clouds (or hazes) on sol 271 at sunrise was also confirmed by images taken by the Visual Monitoring Camera (VMC) (Sánchez-Lavega etal.,2018) onboard Mars Express (lower panel). In the VMC images we see that Jezero crater (indicated with the black arrow) was overcast by bright morning limb clouds or hazes. 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 4 of 18 twilight for a cloud free day and for a day with the presence of clouds (also confirmed by orbiter observations). We observe that the clouds produce an increase in the irradiance (indicated by the black arrows) with respect to the scenario without clouds, and thus by comparing each twilight recorded by the RDS with the signals for the cloud-free scenario, we can infer on what sols of the mission there were clouds (or high dust layers) present during this period of the day. An easy way to determine when cloud features are present in the RDS observations is to compare the Top-4 and Top-8 signals for each twilight with those for the cloud-free day of Figure1 (red lines and referred hereafter as the reference signals). Figure2 shows, as an example, the correlation between the twilight RDS signals at 450 and 950nm and the reference signals (at the same wavelengths and SZAs) for three different sunrises. The two correlation curves (one for Top-4 and another for Top-8) derived for each sunrise were fitted to a straight line, whose slopes were compared with the identity relation (gray dashed line). If the slopes are close to 1, then the observations indicate aerosol conditions similar to those found for the reference signals (cloud-free twilight). On the contrary, if the slopes are <1, then the observations point to the possible presence of high ALs. Note that because the high ALs increase irradiance at surface during twilight and the reference signals are plotted in the x-axis of Figure2, the slopes are expected to be smaller than 1 under cloud conditions. In the examples illustrated in Figure2, we obtained slope values of 1.01, 0.95, and 0.69 at 450nm for sols 190, 178, and 311, respectively. We estimated that the slopes are significantly different to 1 when the slopes are smaller than ∼0.97. This threshold represents the maximum slope value below 1 for which signal differences relative to the reference signal are significant (accounting for signal errors). Therefore, based on these results we can infer that high ALs were potentially present at sunrise only for sols 178 and 311. For determining the composition of the detected high ALs (layers made of ices or just dust), as both high-altitude clouds or detached dust layers are expected to cause similar effects on the slope values, we will make use of the ratio between the intensity at zenith measured at two different wavelengths. In particular, by choosing two wavelengths at which the single scattering albedo (or the imaginary part of the refractive index) of the dust particles is very different but approximately the same for water ice, then the value of the ratio between the intensities at these two wavelengths highly depends on the aerosol composition. We can compute these ratios, defined here as the color index (CI), from the measurements made by different RDS Top channels; a CI from the ratio between Top 3 (250–400) and Top 6 (750nm) channels, and another CI from the ratio between Top 4 (450) and Top 6 (750nm) channels. This selection of channels is based on the fact that dust particles have a much greater imaginary refractive index at 250–400 and 450nm than at 750nm (Wolff etal.,2009,2010). According to this CI definition and since the single scattering albedo of the water (or CO2) ice particles is ∼1 in any of these three RDS channels, we expect greater values of Top3 (250–400nm)/Top6 (750nm) and Top4 (450nm)/Top6 (750nm) when the high-altitude ALs are made of water ice than when they are composed of only dust. Although similar results would be obtained by using the Top 8 (950nm) channel instead of the Top 6, we made this election because the Top 6 wavelength range is the closest one to the minimum in the imaginary refractive index of the dust (Wolff etal.,2009). It is important to note here that the CI is also sensitive to variations in the cloud particle size. In particular, an increase in the particle radius of the high-altitude ALs would also decrease the CI values Figure 2. Correlation between the Radiation and Dust Sensor signals at 450 (blue) and 950nm (red) measured for a cloud-free day, represented in the x-axis, and during the dawn of (a) sol 190, (b) 178, and (c) 311, represented in the y-axis. For each twilight we derived two correlation plots, one per channel, whose measurements are compared with the signals measured under cloud-free conditions for the same solar zenith angles. The correlation curves were fitted to a straight line (solid lines in blue and red for the Top 4 and Top 8 channels, respectively) whose slope is used to infer the presence of aerosol layer. The gray dashed line represents the identity relation. 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 5 of 18 (for a constant opacity). This is due to the dependence of the phase function and single scattering albedo on the particle radius. 2.3. Atmospheric and Radiative Transfer Modeling The cloud properties are derived by modeling the RDS Top measurements with RT simulations. We first investigated the sensitivity of the RDS signals to different cloud properties (e.g., altitude, opacity, geometrical thickness, particle shape), and found that the cloud altitude, number density, and particle radius are the parameters with the greatest impact. RT simulations at twilight are made with a three-dimensional Monte Carlo RT model in spherical geometry (since the plane-parallel approximation breaks down for high SZAs) previously used for cloud properties retrievals on Earth (Gomez-Martin etal.,2021; Toledo, Rannou, Pommereau, Sarkissian, & Foujols,2016), Titan (Rannou etal.,2016; West etal.,2016), and adapted to the Martian atmosphere (Toledo, Rannou, Pommereau, & Foujols,2016; Toledo etal.,2017). Since Monte Carlo RT simulations take a long time to calculate, the retrieval procedure makes use of a pre-computed set of look-up tables, for minimizing the mean square difference between simulated and observed RDS signals. Cloud scattering properties are computed with Mie theory and the refractive index of water ice (Warren,1984). The cloud geometrical thickness was fixed at 2km (we varied this parameter up to values of 6km and did not found significant variations in the simulations), and the cloud spatial distribution density was defined by a Gaussian height profile, scaled to produce the desired opacity. The dust scattering properties were derived from the empirical formulation proposed by Pollack and Cuzzi(1980) and using the spectral refractive index given in Wolff etal.(2009) is used. The RDS signals were normalized by the intensity measured at SZA=90° (or the minimum SZA of the twilight) to reduce the impact of the background dust properties (opacity and reff) on the cloud retrievals. For the vertical distribution of dust particles, we adopted the modified Conrath profile (Conrath,1975) proposed by Forget etal.(1999). 𝜏𝜏 (z) = 𝜏𝜏0⋅𝜎𝜎(z) ⋅exp [ 𝜈𝜈⋅ ( 1−𝜎𝜎(z) −l)] (1) where τ0 is the vertical opacity at surface, σ(z) is the ratio between the pressure at z level and the pressure at surface (here we assume p varies with height as p=p0⋅exp(−z/H), where p0 is the pressure at surface and H the scale height and equal to 11km), ν is a constant set to 0.007 and l is the ration between a reference height (set to 70km) and the altitude of the top of the dust layer (Zmax). We investigated the use of more complex dust vertical distributions in our RT simulations. Based on previous works, we simulated the RDS signals using a dust vertical profile resulting from a Conrath-type profile and a detached dust layer (defined by a Gaussian height profile) with variable altitude. We found no significant differences in the cloud retrievals using this non-monotonic dust vertical distributions for detached-dust layer altitudes less than or equal to 25km. Based on the results reported in McCleese etal.(2010) and Heavens etal.(2011a,2011b), which found the maximum dust mass mixing ratio at altitudes between 15 and 25km (for MY 28–29 and for most of the northern spring and summer), we favored the simple Conrath-type profiles over more complex dust structures for our retrieval analysis. 3. Results 3.1. Presence of Clouds in the Period Ls 39°–262° The slope analysis described in 2.2 was performed for all the twilights available up to sol 492 (Ls=262°), whose results are displayed in the upper panel of Figure3a. An inspection of the slope values reveals 4 obvious periods of different high altitude aerosol activity: 1. Between Ls∼39° and 50°, high-altitude ALs signatures in the RDS signals are found for about ∼40% of the twilights covered by MEDA. In general, the slopes obtained at sunrise are smaller than those during sunset, suggesting greater opacities or altitudes. We cannot establish the start of this period as no RDS data with high gain is available before Ls=39°. In the following section and in AppendixA we will show that these ALs are at altitudes above ∼30km. 2. The second period, between Ls∼50° and 114°, is characterized by a notable drop in the high-altitude aerosol detections: in this period, values of the slopes are close to 1. Only 27 twilights out of 101 present slopes smaller than 0.97 for Top 4 (450nm) channel, and 15 out of 101 for Top 8 (950nm) channel. The particular conditions which led to this decrease are unclear and will be discussed in Section3.3. 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 6 of 18 3. After the second period, we see a decrease in the slopes derived from both channels, and thus an increase in the number of twilights with the presence of high-altitude ALs. In general, the slopes during this period are smaller than those for the first period and decrease with Ls (before reaching the minimum), indicating higher aerosol altitudes or opacities if we assume that the slope in our correlation plots decrease with these two parameters (this will be demonstrated in the next section). The decrease in the correlation slopes lasts up to Ls∼150°, which is when the minimum is found. During this period of maximum high-altitude aerosol activity (between Ls=114° and 162°) is when the cameras of Perseverance rover and MEDA detected the Figure 3. (a) Correlation slopes derived from the procedure described in Figure2 and the Radiation and Dust Sensor (RDS) Top 4 (blue squares) and Top 8 (red squares) observations for the twilights covered by RDS up to sol 492. The black dashed line indicates the time when a regional dust storm was observed in Jezero (Lemmon, Smith, etal.,2022), and the black solid line shows the 0.97 threshold value. (b) color index signals used to discriminate between dust and ice are computed from the ratio between the RDS Top 3 (250–400nm) and Top 6 (750nm) observations (blue dots) and from the ratio between the RDS Top 4 (450) and Top 6 (750nm) observations (red dots). 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 7 of 18 formation of a 22° scattering halo (Lemmon, Toledo, etal.,2022) around the Sun, Ls=142° and when a regional dust storm (MY36/2022A) was actively raising dust in Jezero crater (Lemmon, Smith, etal.,2022; Sánchez-Lavega etal.,2022; M. D. Smith etal.,2023), Ls=153°–156° (indicated by the black dashed line). 4. For Ls>∼156° (after the dust storm), the slopes become closer to 1 but with values smaller than during the second period and highly variable. The variability during the last period could indicate a change in the kind of aerosol (ice or dust) present in the higher parts of the atmosphere. Although the results shown in Figure3a may indicate a change in the aerosol-type present during twilight after the regional dust storm MY36/2022A (dashed black line in Figure3a), from the correlation slopes we cannot directly discriminate between high-altitude detached dust layers or clouds. Indeed, in both scenarios the RDS observations at 450 and 950nm would show an increase respect to the reference signals, and thus have a slope lower than 1 in the correlation plots. To evaluate the possible aerosol composition of the different detected high ALs, Figure3b shows the ratios Top 3 (250–400nm)/Top 6 (750nm) and Top 4 (450nm)/Top 6 (750nm) (CI defined in Section2.2) for the same twilights analyzed in Figure3a. For each twilight, we represent the ratio values given at SZA∼90°. From approximately Ls=39° until Ls=150°, which is within the ACB season, the CI does not show strong variations. At the time around when the regional dust storm passed over the Perseverance rover site, the CI decreased by a factor of 2 in a few sols. This is consistent with the presence of dust layers at high altitude and the increase in the dust opacity as a result of the dust storm. Interestingly, although the CI increased once the dust storm had vanished, it never recovered the values registered before. Moreover, the CI time-series clearly shows a negative trend after Ls∼172°. Therefore, based on these results we can identify two periods with different aerosol scenarios during twilight: (a) a first period from Ls∼39° to ∼150° with high and stable CI values likely produced by the presence of predominantly water ice; (b) a second period with lower CI values that are decreasing with time, mainly dominated by dust. This is also consistent with the results using the observations made by MEDA-TIRS and reported in M. D. Smith etal.(2023) that a systematic change in the diurnal trend of the aerosol opacity occurred around Ls 150°. In that work, a diurnal and a seasonal component in the aerosol opacity variability was derived, and from that it was inferred that after the regional dust storm the dust was the aerosol dominating the opacity. On the basis of these results, we conclude that high-altitude ALs found in the correlation slopes before Ls∼150° were mainly made of ice particles, while the cases after that date corresponded to ALs whose opacity was dominated by dust. From the analysis of the correlation slopes and the CI we cannot infer whether the observed ice particles consisted of clouds (detached layers at a given altitude) or hazes vertically extended over several km. In the next section, we will make use of RT simulations to constrain the cloud properties of the detection cases before Ls 150°. As indicated before, the CI values are also sensitive to variations in the cloud particle radius. Nonetheless, because the decrease in CI coincides with the end of the ACB season and our RT simulations do not indicate a systematic change in the particle radius around Ls 150°, we conclude that the drop in CI is primarily due to the aerosol composition. 3.2. Cloud Altitude, Opacity, and Particle Size Retrievals The cloud altitude, number density and particle size were derived by fitting the RDS Top 4 (450) and Top 8 (950nm) twilight observations simultaneously with the model described in Section2.3. The cloud opacity at each wavelength is derived from the fitted cloud number density and the particle cross section, computed from the fitted reff and the refractive index of water ice. Only the twilights for which the RDS observations covered the minimum SZA range of (91°–97°) were considered in the analysis (a total of 161 twilights). In AppendixA we demonstrate that for clouds above ∼30km, our retrievals are not significantly affected by the vertical extension of the main dust layer (for this reason and to decrease the number of free parameters, our analysis is focused on altitudes above 30km). Assuming that there were not detached dust layers above 25km in our observations for the cloudy period, we used a Zmax=45km for the dust profiles. We did not find significant differences in our cloud retrievals by varying this parameter from 30 to 50km. We also performed a sensitivity analysis for the dust opacity and reff, detailed in AppendixB, to evaluate the impact of these parameters on the cloud retrievals. We found that for opacity and reff values between 0.3 and 0.6, and between 1.2 and 1.4μm, respectively, our cloud retrievals are not significantly affected. For this reason and based on the times series of the dust opacity retrieved from images taken regularly by SkyCam (see AppendixB), these parameters are fixed to 0.4 and 1.4μm. Regarding the cloud particle shape, we investigated the impact in the RDS signals when using different shapes other 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 8 of 18 than spheres (to see if adding an additional free parameter was needed). In particular, similar simulations were made but using spheroid and cylindrical particles and we did not find significant variations (see AppendixC for more information). Therefore, the only free parameters in our inversion analysis are the cloud altitude, opacity and particle radius. Figure4 shows three of our best fits to the data acquired by Top-4 and Top-8 sensors during twilight for sol 78, 99, and 292, and where we can see that we match the normalized signals very well (reduced Chi-square function, χ 2, values<1.2). For these cases, the correlation slopes are 0.97, 1.01, and 0.92 at 450nm, and 0.96, 0.99, and 0.95 at 950nm, respectively. For sol 99, whose slope is ∼1, the observations could be fitted without a cloud in the RT model. This is consistent with our assumption in Section3.1 that slopes ∼1 are indicative of skies free of high-altitude ALs. For the fitted signals for sol 78, whose intensities are greater than those on sol 99 for the same SZAs, we used the cloud model described above and we derived a cloud altitude, opacity and reff of 42.8±4.1km, 0.011±0.004 and 1.14±0.21μm, respectively. We originally attempted to fit the data using only the dust Conrath profiles, with Zmax treated as a free parameter, but could not achieve a fit with a reasonably good χ 2 using this model (reduced χ 2≫1). However, for the same observations (sol 78), a reduced χ 2 similar to that obtained with the detached cloud model was achieved by using a cloud vertically extended over several kilometers and with reff<0.4μm. In this aerosol model, referred as the haze model thereafter, the layer of ice particles is extended over 30km (or more) and centered at an altitude of 40km. Therefore, for these particular observations we could not infer if the deviation with respect to the reference signals was produced by the presence of detached clouds or hazes. For the three twilights, the observations made on sol 292 show the higher deviations with respect to the reference signals, and the cloud model provides fitted cloud altitude, opacity and reff values of 42.0±2.0km, 0.031±0.006 and 1.34±0.52μm, respectively. For these observations, neither the haze model nor the dust model could fit the data with a reasonably good χ 2, thus indicating unequivocally the presence of clouds. Although the opacities derived for sol 78 and 292 are small, it is important to note that these opacities represent the average over the sensor's FOV. If, during the detection, the clouds covered only a few percent of the FOV, then our retrieved cloud opacity would be smaller than that derived from an instrument (e.g. a camera) whose FOV is fully covered by the cloud. Another point to note is that from this analysis we can only infer the cloud or haze opacity above ∼30km (see AppendixA), and thus the opacity contribution from clouds or hazes below this level are not included in the cloud opacity retrievals. A similar analysis was performed for all the twilights covered by MEDA up to Ls=150°, which is the time when the drop in the CI is observed (Figure3). For the complete data set (a total of 161 twilights analyzed with the RT model), the signals shown in Figure4 are representative examples. In 54 twilights, RDS observations indicated the presence of clouds, whose fitted parameters are displayed in the left panels of Figure5 (a, c, and e). In these cases, the cloud model achieved reduced χ 2 values<1.2, and similar results were not obtained (in terms of χ 2) by replacing the cloud layer by a vertically extended haze. That is to say, these cases are like the twilight on sol 292 analyzed in Figure4. On the other hand, for a total of 40 twilights, we found that both the cloud and haze models fitted the data with reduced χ 2 values<1.2. The results obtained for these cases using the cloud model Figure 4. Comparison between simulations and observations at 450 (left) and 950nm (right) for the twilights of sols 78, 99, and 292. The shaded areas represent the errors and the red dashed lines the simulations using the cloud parameters fitted for each case. For each twilight, the observations at 450 and 950nm were fitted simultaneously. 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 9 of 18 are displayed in the right panels of Figure5 (b, d, and f). For the rest of the twilights, we have: (a) 42 cases with cloud opacities below our limit of detection (∼0.004), defined in this work as the minimum opacity to produce a variation of at least 5% respect to the reference signals at SZA=93°; (b) 25 cases for which none of the models achieved a good fit (likely due to changes in the cloud opacity during the twilight period or to complex aerosol scenarios). Therefore, in the 58% of the twilights analyzed we found signatures of clouds or hazes in the RDS signals. In most of the cases the clouds were found at altitudes between 40 and 50km. Based on these altitudes, we assume these clouds are made of water ice. However, we point that from the modeling of the RDS signals, we cannot directly discriminate between clouds made of CO2 or H2O ice. Therefore, we can not rule out the possibility that some of the clouds shown in Figure5 are made of CO2 ice (in particular those with the highest altitudes). In general, the cloud particle sizes were in the range between reff=0.6 and 2μm (accounting for the errors in this Figure 5. Cloud altitude, opacity, and effective radius (reff) retrieved from Radiation and Dust Sensor (RDS) Top-4 and Top-8 twilight observations up to Ls=150° using the cloud model described in Section2.3. The left panels (a, c, and e) represent the cloud cases for which only the cloud model could fit the data with a reduced χ 2<1.2, while the right ones (b, d, and f) the cases for which both the cloud and haze models achieved fits with reduced χ 2<1.2. The purple dots indicate the twilight Ls dates for which the minimum solar zenith angles range (91°–97°) was covered by the RDS observations. 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 16 of 18 Data Availability Statement All Perseverance data used in this study are publicly available via the Planetary Data System (Rodriguez-Manfredi & de la Torre Juarez,2021). The slope and CI analyses, radiative transfer simulations, cloud retrievals, temperatures, and MCD data of Figures1-8,A1,B1,B2,C1, andC2 are available in an archive located at Toledo(2023). References Anderson, E., & Leovy, C. (1978). Mariner 9 television limb observations of dust and ice hazes on Mars. Journal of the Atmospheric Sciences, 35(4), 723–734. https://doi.org/10.1175/1520-0469(1978)035<0723:mtlood>2.0.co;2 Apestigue, V., Gonzalo, A., Jiménez, J. J., Boland, J., Lemmon, M., de Mingo, J. R., etal. (2022). Radiation and dust sensor for Mars environmental dynamic analyzer onboard M2020 rover. Sensors, 22(8), 2907. https://doi.org/10.3390/s22082907 Atwood, S. A., Smith, M. D., Badri, K., Edwards, C. S., Christensen, P.R., Wolff, M. J., etal. (2022). Diurnal variability in emirs daytime observations of water ice clouds during Mars aphelion-season. Geophysical Research Letters, 49(15), e2022GL099654. https://doi. org/10.1029/2022gl099654 Benson, J. L., Kass, D. M., & Kleinböhl, A. (2011). Mars' north polar hood as observed by the Mars climate sounder. Journal of Geophysical Research, 116(E3), E03008. https://doi.org/10.1029/2010je003693 Benson, J. L., Kass, D. M., Kleinböhl, A., McCleese, D. J., Schofield, J. T., & Taylor, F. W. (2010). Mars' south polar hood as observed by the Mars climate sounder. Journal of Geophysical Research, 115(E12), E12015. https://doi.org/10.1029/2009je003554 Clancy, R. T., Grossman, A., Wolff, M., James, P., Rudy, D., Billawala, Y., etal. (1996). Water vapor saturation at low altitudes around Mars aphelion: A key to Mars climate? Icarus, 122(1), 36–62. https://doi.org/10.1006/icar.1996.0108 Clancy, R. T., Wolff, M. J., & Christensen, P.R. (2003). Mars aerosol studies with the MGS TES emission phase function observations: Optical depths, particle sizes, and ice cloud types versus latitude and solar longitude. Journal of Geophysical Research, 108(E9), 5098. https://doi. org/10.1029/2003je002058 Clancy, R. T., Wolff, M. J., Whitney, B. A., Cantor, B. A., & Smith, M. D. (2007). Mars equatorial mesospheric clouds: Global occurrence and physical properties from Mars global surveyor thermal emission spectrometer and Mars orbiter camera limb observations. Journal of Geophysical Research, 112(E4), E04004. https://doi.org/10.1029/2006je002805 Conrath, B. J. (1975). Thermal structure of the Martian atmosphere during the dissipation of the dust storm of 1971. Icarus, 24(1), 36–46. https:// doi.org/10.1016/0019-1035(75)90156-6 Fedorova, A. A., Montmessin, F., Korablev, O., Luginin, M., Trokhimovskiy, A., Belyaev, D. A., etal. (2020). Stormy water on Mars: The distribution and saturation of atmospheric water during the dusty season. Science, 367(6475), 297–300. https://doi.org/10.1126/science.aay9522 Forget, F., Hourdin, F., Fournier, R., Hourdin, C., Talagrand, O., Collins, M., etal. (1999). Improved general circulation models of the Martian atmosphere from the surface to above 80 km. Journal of Geophysical Research, 104(E10), 24155–24175. https://doi.org/10.1029/1999je001025 Forget, F., & Pierrehumbert, R. T. (1997). Warming early Mars with carbon dioxide clouds that scatter infrared radiation. Science, 278(5341), 1273–1276. https://doi.org/10.1126/science.278.5341.1273 Giuranna, M., Wolkenberg, P., Grassi, D., Aronica, A., Aoki, S., Scaccabarozzi, D., etal. (2021). The current weather and climate of Mars: 12 years of atmospheric monitoring by the planetary Fourier spectrometer on Mars express. Icarus, 353, 113406. https://doi.org/10.1016/j. icarus.2019.113406 Gomez-Martin, L., Toledo, D., Prados-Roman, C., Adame, J. A., Ochoa, H., & Yela, M. (2021). Polar stratospheric clouds detection at Belgrano ii Antarctic station with visible ground-based spectroscopic measurements. Remote Sensing, 13(8), 1412. https://doi.org/10.3390/rs13081412 Guzewich, S. D., & Smith, M. (2019). Seasonal variation in Martian water ice cloud particle size. Journal of Geophysical Research: Planets, 124(2), 636–643. https://doi.org/10.1029/2018je005843 Heavens, N., Richardson, M., Kleinböhl, A., Kass, D., McCleese, D., Abdou, W., etal. (2011a). Vertical distribution of dust in the Martian atmosphere during northern spring and summer: High-altitude tropical dust maximum at northern summer solstice. Journal of Geophysical Research, 116(E1), E01007. https://doi.org/10.1029/2010je003692 Heavens, N., Richardson, M., Kleinböhl, A., Kass, D., McCleese, D., Abdou, W., etal. (2011b). The vertical distribution of dust in the Martian atmosphere during northern spring and summer: Observations by the Mars climate sounder and analysis of zonal average vertical dust profiles. Journal of Geophysical Research, 116(E4), E04003. https://doi.org/10.1029/2010je003691 Figure C2. Radiation and Dust Sensor Top-8 signals simulated for a cloud layer at an altitude of 40km and opacity of 0.005, and the phase functions illustrated in FigureC1. For the dust particles, we used the model described in Section2.3 (or in Figure4). Acknowledgments This work has been funded by the Spanish Ministry of Economy and Competitiveness, through the projects no. ESP2014-54256-C4-1-R (also ESP2014-54256-C4-2-R, ESP201454256-C4-3-R, and ESP2014-54256C4-4-R), Spanish Ministry of Science, Innovation and Universities, projects no. ESP2016-79612-C3-1-R (also ESP2016-79612-C3-2-R and ESP201679612-C3-3-R), Spanish Ministry of Science and Innovation/State Agency of Research (10.13039/501100011033), projects no. PID2021-126719OB-C41, ESP2016-80320-C2-1-R, RTI2018098728-B-C31 (also RTI2018-098728B-C32 and RTI2018-098728-B-C33), RTI2018-099825-B-C31. RH and ASL were supported by the Spanish project PID2019-109467GB-I00 funded by MCIN/AEI/10.13039/50110001103 and by Grupos Gobierno Vasco IT1742-22. The US co-authors performed their work under sponsorship from NASA’s Mars 2020 project, from the Game Changing Development programme within the Space Technology Mission Directorate and from the Human Exploration and Operations Directorate. Part of this research was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). G.M. acknowledges JPL funding from USRA Contract Number 1638782. ML is supported by contract 15-712 from Arizona State University and 1607215 from Caltech-JPL. A. V-R. is supported by the Comunidad de Madrid Project S2018/NMT-4291 (TEC2SPACE-CM). 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 17 of 18 Hernández-Bernal, J., Sánchez-Lavega, A., del Río-Gaztelurrutia, T., Hueso, R., Ravanis, E., Cardesín-Moinelo, A., etal. (2021). A long-term study of Mars mesospheric clouds seen at twilight based on Mars express VMC images. Geophysical Research Letters, 48(7), e2020GL092188. https://doi.org/10.1029/2020GL092188 Jaquin, F., Gierasch, P., & Kahn, R. (1986). The vertical structure of limb hazes in the Martian atmosphere. Icarus, 68(3), 442–461. https://doi. org/10.1016/0019-1035(86)90050-3 Lemmon, M. T., Smith, M. D., Viudez-Moreiras, D., de la Torre-Juarez, M., Vicente-Retortillo, A., Munguira, A., et al. (2022). Dust, sand, and winds within an active Martian storm in Jezero Crater. Geophysical Research Letters, 49(17), e2022GL100126. https://doi. org/10.1029/2022gl100126 Lemmon, M. T., Toledo, D., Apestigue, V., Arruego, I., Wolff, M. J., Patel, P., et al. (2022). Hexagonal prisms form in water-ice clouds on Mars, producing halo displays seen by perseverance rover. Geophysical Research Letters, 49(17), e2022GL099776. https://doi. org/10.1029/2022gl099776 Lemmon, M. T., Wolff, M. J., Bell, J. F., III, Smith, M. D., Cantor, B. A., & Smith, P.H. (2015). Dust aerosol, clouds, and the atmospheric optical depth record over 5 Mars years of the Mars exploration rover mission. Icarus, 251, 96–111. https://doi.org/10.1016/j.icarus.2014.03.029 Lorenz, R. D., Lemmon, M. T., Maki, J., Banfield, D., Spiga, A., Charalambous, C., etal. (2020). Scientific observations with the insight solar arrays: Dust, clouds, and eclipses on Mars. Earth and Space Science, 7(5), e2019EA000992. https://doi.org/10.1029/2019ea000992 Määttänen, A., Montmessin, F., Gondet, B., Scholten, F., Hoffmann, H., González-Galindo, F., etal. (2010). Mapping the mesospheric CO2 clouds on Mars: MEx/omega and MEx/HRSC observations and challenges for atmospheric models. Icarus, 209(2), 452–469. https://doi. org/10.1016/j.icarus.2010.05.017 Määttänen, A., Pérot, K., Montmessin, F., & Hauchecorne, A. (2013). Mesospheric clouds on Mars and on Earth. Comparative Climatology of Terrestrial Planets, 393–413. https://doi.org/10.2458/azu_uapress_9780816530595-ch16 Määttänen, A., Vehkamäki, H., Lauri, A., Merikallio, S., Kauhanen, J., Savijärvi, H., & Kulmala, M. (2005). Nucleation studies in the Martian atmosphere. Journal of Geophysical Research, 110(E2), E02002. https://doi.org/10.1029/2004je002308 Madeleine, J.-B., Forget, F., Millour, E., Montabone, L., & Wolff, M. (2011). Revisiting the radiative impact of dust on Mars using the LMD global climate model. Journal of Geophysical Research, 116(E11), E11010. https://doi.org/10.1029/2011je003855 Maltagliati, L., Montmessin, F., Fedorova, A., Korablev, O., Forget, F., & Bertaux, J.-L. (2011). Evidence of water vapor in excess of saturation in the atmosphere of Mars. Science, 333(6051), 1868–1871. https://doi.org/10.1126/science.1207957 McCleese, D., Heavens, N., Schofield, J., Abdou, W., Bandfield, J., Calcutt, S., etal. (2010). Structure and dynamics of the Martian lower and middle atmosphere as observed by the Mars climate sounder: Seasonal variations in zonal mean temperature, dust, and water ice aerosols. Journal of Geophysical Research, 115(E12), E12016. https://doi.org/10.1029/2010je003677 McConnochie, T., Bell, J., III, Savransky, D., Wolff, M., Toigo, A., Wang, H., etal. (2010). Themis-vis observations of clouds in the Martian mesosphere: Altitudes, wind speeds, and decameter-scale morphology. Icarus, 210(2), 545–565. https://doi.org/10.1016/j.icarus.2010.07.021 Mishchenko, M. I. (1991). Light scattering by randomly oriented axially symmetric particles. Journal of the Optical Society of America A, 8(6), 871–882. https://doi.org/10.1364/josaa.8.000871 Mishchenko, M. I., & Travis, L. D. (1994). T-matrix computations of light scattering by large spheroidal particles. Optics Communications, 109(1–2), 16–21. https://doi.org/10.1016/0030-4018(94)90731-5 Montabone, L., Spiga, A., Kass, D. M., Kleinböhl, A., Forget, F., & Millour, E. (2020). Martian year 34 column dust climatology from Mars climate sounder observations: Reconstructed maps and model simulations. Journal of Geophysical Research: Planets, 125(8), e2019JE006111. https://doi.org/10.1029/2019je006111 Montmessin, F., Forget, F., Rannou, P., Cabane, M., & Haberle, R. M. (2004). Origin and role of water ice clouds in the Martian water cycle as inferred from a general circulation model. Journal of Geophysical Research, 109(E10), E10004. https://doi.org/10.1029/2004je002284 Montmessin, F., Quémerais, E., Bertaux, J.-L., Korablev, O., Rannou, P., & Lebonnois, S. (2006). Stellar occultations at UV wavelengths by the SPICAM instrument: Retrieval and analysis of Martian haze profiles. Journal of Geophysical Research, 111(E9), E09S09. https://doi. org/10.1029/2005je002662 Montmessin, F., Rannou, P., & Cabane, M. (2002). New insights into Martian dust distribution and water-ice cloud microphysics. Journal of Geophysical Research, 107(E6), 4-1–4-14. https://doi.org/10.1029/2001JE001520 Munguira, A., Hueso, R., Sánchez-Lavega, A., Torre-Juarez, M. D. L., Martínez, G., Newman, C., etal. (2022). Near surface atmospheric temperatures at Jezero from Mars 2020 MEDA measurements. Journal of Geophysical Research: Planets, 128, e2022JE007559. https://doi. org/10.1029/2022JE007559 Navarro, T., Madeleine, J.-B., Forget, F., Spiga, A., Millour, E., Montmessin, F., & Määttänen, A. (2014). Global climate modeling of the Martian water cycle with improved microphysics and radiatively active water ice clouds. Journal of Geophysical Research: Planets, 119(7), 1479–1495. https://doi.org/10.1002/2013je004550 Pollack, J. B., & Cuzzi, J. N. (1980). Scattering by nonspherical particles of size comparable to a wavelength: A new semi-empirical theory and its application to tropospheric aerosols. Journal of the Atmospheric Sciences, 37(4), 868–881. https://doi. org/10.1175/1520-0469(1980)037<0868:sbnpos>2.0.co;2 Rannou, P., Perrier, S., Bertaux, J.-L., Montmessin, F., Korablev, O., & Reberac, A. (2006). Dust and cloud detection at the Mars limb with UV scattered sunlight with SPICAM. Journal of Geophysical Research, 111(E9), E09S10. https://doi.org/10.1029/2006je002693 Rannou, P., Toledo, D., Lavvas, P., D’aversa, E., Moriconi, M., Adriani, A., etal. (2016). Titan's surface spectra at the Huygens landing site and Shangri-la. Icarus, 270, 291–306. https://doi.org/10.1016/j.icarus.2015.09.016 Rodriguez-Manfredi, J. A., & de la Torre Juarez, M. (2021). Mars 2020 perseverance rover Mars environmental dynamics analyzer (MEDA) experiment data record (EDR) and reduced data record (RDR) data products archive bundle [Dataset]. PDS Atmospheres Node. Retrieved from https://arcnav.psi.edu/urn:nasa:pds:mars2020_meda:browse_skycam Rodriguez-Manfredi, J. A., De la Torre Juárez, M., Alonso, A., Apéstigue, V., Arruego, I., Atienza, T., etal. (2021). The Mars environmental dynamics analyzer, MEDA. A suite of environmental sensors for the Mars 2020 mission. Space Science Reviews, 217(3), 1–86. https://doi. org/10.1007/s11214-021-00816-9 Rodriguez-Manfredi, J. A., de la Torre Juarez, M., Sanchez-Lavega, A., Hueso, R., Martinez, G., Lemmon, M., etal. (2023). The diverse meteorology of jezero crater over the first 250 sols of perseverance on Mars. Nature Geoscience, 16, 1–10. https://doi.org/10.1038/s41561-022-01084-0 Sánchez-Lavega, A., Chen-Chen, H., Ordonez-Etxeberria, I., Hueso, R., del Rio-Gaztelurrutia, T., Garro, A., etal. (2018). Limb clouds and dust on Mars from images obtained by the visual monitoring camera (VMC) onboard Mars express. Icarus, 299, 194–205. https://doi.org/10.1016/j. icarus.2017.07.026 Sánchez-Lavega, A., Rio-Gaztelurrutia, T. D., Hueso, R., Juárez, M. D. L. T., Martínez, G., Harri, A.-M., etal. (2022). Mars 2020 perseverance rover studies of the Martian atmosphere over Jezero from pressure measurements. Journal of Geophysical Research: Planets, 128, e2022JE007480. https://doi.org/10.1029/2022JE007480 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Journal of Geophysical Research: Planets TOLEDO ETAL. 10.1029/2023JE007785 18 of 18 Smith, M. D. (2009). Themis observations of Mars aerosol optical depth from 2002–2008. Icarus, 202(2), 444–452. https://doi.org/10.1016/j. icarus.2009.03.027 Smith, M. D., Badri, K., Atwood, S. A., Edwards, C. S., Christensen, P.R., Wolff, M. J., etal. (2022). EMIRS observations of the aphelion-season Mars atmosphere. Geophysical Research Letters, 49(15), e2022GL099636. https://doi.org/10.1029/2022gl099636 Smith, M. D., Martínez, G. M., Sebastián, E., Lemmon, M. T., Wolff, M. J., Apéstigue, V., etal. (2023). Diurnal and seasonal variations of aerosol optical depth observed by MEDA/TIRS at Jezero Crater, Mars. Journal of Geophysical Research: Planets, 128(1), e2022JE007560. https:// doi.org/10.1029/2022je007560 Smith, M. D., Wolff, M. J., Clancy, R. T., Kleinböhl, A., & Murchie, S. L. (2013). Vertical distribution of dust and water ice aerosols from CRISM limb-geometry observations. Journal of Geophysical Research: Planets, 118(2), 321–334. https://doi.org/10.1002/jgre.20047 Smith, P.H., & Lemmon, M. (1999). Opacity of the Martian atmosphere measured by the imager for Mars pathfinder. Journal of Geophysical Research, 104(E4), 8975–8985. https://doi.org/10.1029/1998je900017 Stcherbinine, A., Montmessin, F., Vincendon, M., Wolff, M. J., Vals, M., Korablev, O., etal. (2022). A two Martian years survey of water ice clouds on Mars with ACS onboard TGO. Journal of Geophysical Research: Planets, 127(12), e2022JE007502. https://doi.org/10.1029/2022je007502 Tamppari, L. K., Zurek, R., & Paige, D. (2003). Viking-era diurnal water-ice clouds. Journal of Geophysical Research, 108(E7), 5073. https:// doi.org/10.1029/2002je001911 Toledo, D. (2023). Twilight mesospheric clouds in Jezero as observed by MEDA radiation and dust sensor (RDS) [Dataset]. Zenodo. https://doi. org/10.5281/zenodo.7932426 Toledo, D., Arruego, I., Apéstigue, V., Jiménez, J., Gómez, L., Yela, M., etal. (2017). Measurement of dust optical depth using the solar irradiance sensor (SIS) onboard the exomars 2016 EDM. Planetary and Space Science, 138, 33–43. https://doi.org/10.1016/j.pss.2017.01.015 Toledo, D., Rannou, P., Pommereau, J.-P., & Foujols, T. (2016). The optical depth sensor (ODS) for column dust opacity measurements and cloud detection on Martian atmosphere. Experimental Astronomy, 42(1), 61–83. https://doi.org/10.1007/s10686-016-9500-7 Toledo, D., Rannou, P., Pommereau, J.-P., Sarkissian, A., & Foujols, T. (2016). Measurement of aerosol optical depth and sub-visual cloud detection using the optical depth sensor (ODS). Atmospheric Measurement Techniques, 9(2), 455–467. https://doi.org/10.5194/amt-9-455-2016 Wang, H., & Ingersoll, A. P. (2002). Martian clouds observed by Mars global surveyor Mars orbiter camera. Journal of Geophysical Research, 107(E10), 8-1–8-16. https://doi.org/10.1029/2001JE001815 Warren, S. G. (1984). Optical constants of ice from the ultraviolet to the microwave. Applied Optics, 23(8), 1206–1225. https://doi.org/10.1364/ ao.23.001206 West, R., Del Genio, A., Barbara, J., Toledo, D., Lavvas, P., Rannou, P., etal. (2016). Cassini imaging science subsystem observations of Titan's south polar cloud. Icarus, 270, 399–408. https://doi.org/10.1016/j.icarus.2014.11.038 Wolff, M. J., Clancy, R. T., Goguen, J. D., Malin, M. C., & Cantor, B. A. (2010). Ultraviolet dust aerosol properties as observed by MARCI. Icarus, 208(1), 143–155. https://doi.org/10.1016/j.icarus.2010.01.010 Wolff, M. J., Clancy, R. T., Kahre, M. A., Haberle, R. M., Forget, F., Cantor, B. A., & Malin, M. C. (2019). Mapping water ice clouds on Mars with MRO/MARCI. Icarus, 332, 24–49. https://doi.org/10.1016/j.icarus.2019.05.041 Wolff, M. J., Fernando, A., Smith, M. D., Forget, F., Millour, E., Atwood, S. A., etal. (2022). Diurnal variations in the aphelion cloud belt as observed by the emirates exploration imager (EXI). Geophysical Research Letters, 49(18), e2022GL100477. https://doi.org/10.1029/2022gl100477 Wolff, M. J., Smith, M., Clancy, R., Arvidson, R., Kahre, M., Seelos Iv, F., etal. (2009). Wavelength dependence of dust aerosol single scattering albedo as observed by the compact reconnaissance imaging spectrometer. Journal of Geophysical Research, 114(E2), E00D04. https://doi. org/10.1029/2009je003350 21699100, 2023, 7, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JE007785 by Universidad Del Pais Vasco, Wiley Online Library on [22/12/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License