Constraints on Mars Aphelion Cloud Belt phase function and ice crystal geometries

Highlights

• Cloud phase function was observationally constrained using Mars Science Laboratory.

• Aggregates, columns, plates, and bullet rosettes are more probable ice crystal habits.

• Droxtals and spheres were the habits found less likely to dominate water ice clouds.

Abstract

This study constrains the lower bound of the scattering phase function of Martian water ice clouds (WICs) through the implementation of a new observation aboard the Mars Science Laboratory (MSL). The Phase Function Sky Survey (PFSS) was a multiple pointing all-sky observation taken with the navigation cameras (Navcam) aboard MSL. The PFSS was executed 35 times during the Aphelion Cloud Belt (ACB) season of Mars Year 34 over a solar longitude range of $L_s=61.4°-156.5°$. Twenty observations occurred in the morning hours between 06:00 and 09:30 LTST, and 15 runs occurred in the evening hours between 14:30 and 18:00 LTST, with an operationally required 2.5 h gap on either side of local noon due the sun being located near zenith. The resultant WIC phase function was derived over an observed scattering angle range of 18.3°–152.61°, normalized, and compared with 9 modeled phase functions: seven ice crystal habits and two Martian WIC phase functions currently being implemented in models. Through statistical chi-squared probability tests, the five most probable ice crystal geometries observed in the ACB WICs were aggregates, hexagonal solid columns, hollow columns, plates, and bullet rosettes with p-values greater than or equal to 0.60, 0.57,0.56,0.56, and 0.55, respectively. Droxtals and spheres had p-values of 0.35, and 0.2, making them less probable components of Martian WICs, but still statistically possible ones. Having a better understanding of the ice crystal habit and phase function of Martian water ice clouds directly benefits Martian climate models which currently assume spherical and cylindrical particles.

Introduction

Water ice clouds (WICs) were not considered a significant aspect of Mars' climate following the warm and dusty Viking era (Tamppari, 2000), and their role in the atmosphere remained greatly unappreciated until the 1990's when the Aphelion Cloud Belt (ACB) was discovered (Clancy and Lee, 1991). The scale, duration, and year-to-year repeatability of the ACB led to new speculation and investigations into the range of physical and thermal impacts of Martian water ice clouds on the atmosphere and climate.

The ACB typically extends from 10°S to 30°N latitude, with a range of optical depths $\left(\tau \right)$ from 0.05 to 0.5 within solar longitude $\left(L_s\right)=70°-100°$ (Wolff et al., 1999). With such a broad geographical and temporal extent, the ACB offers an annually re-occurring opportunity to study a variety of Martian WICs, both globally and locally. As a result, there has been great interest in the last two decades to map and characterize Martian WICs, as well as to conduct retrievals of cloud physical properties. Globally, these investigations have utilized the Mars Colour Imager (MARCI), Mars Climate Sounder (MCS) (Kleinböhl et al., 2009), and Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) (Guzewich et al., 2014) aboard the Mars Reconnaissance Orbiter (MRO), the IR Mapping spectrometer OMEGA aboard Mars Express (Madeline et al., 2012), and Mars Orbiter Camera (MOC) and Thermal Emission Spectrometer (TES) aboard Mars Global Surveyor (MGS) (Clancy and Wolff, 2003). These analyses have returned cloud and haze optical depths, ice crystal particle sizes, cloud morphologies and qualitative classifications.

From the surface, Mars Pathfinder (Smith and Lemmon, 1999), the Mars Exploration Rovers (Lemmon, 2004), the Phoenix Lander (Whiteway et al., 2009; Moores et al., 2010) and Mars Science Laboratory (Moores et al., 2015; Kloos et al., 2016) have also completed extensive high-resolution local observations. These surface studies have led to the discovery of precipitation and near-surface fog on Mars (Whiteway et al., 2009; Moores et al., 2011), as well as an understanding of how cloud optical depths vary seasonally, and diurnally (Wilson et al., 2007; Kloos et al., 2018).

One aspect of Martian WICs that has yet to be thoroughly investigated via direct observation from the surface is the scattering phase function, and by extension, the ice crystal habit. Publications that span from the Viking era to present day have largely utilized a method of fitting RT models to emission phase function (EPF) data taken over a finite range and resolution of emission angles (Pollack et al., 1979; Clancy and Lee, 1991; Clancy and Wolff, 2003; Wolff et al., 2009). When the resultant phase functions are plotted against their respective scattering or phase angles, they typically produce flat curves lacking many of the peaks expected across all scattering angles.

In 2016, Kloos et al. used a technique that was similar in nature to the terrestrial retrievals outlined in Chepfer et al. (2002) to constrain the general phase function of Martian WICs. That study resulted in a low resolution constraint on the lower bound of the Martian WIC phase function over a scattering angle range of 70°–115°. In order to achieve this, Kloos et al. utilized a number of single pointing cloud movies taken by the MSL navigation cameras (Navcam) at various observation times, for a little over a Martian year.

This work picked up where Kloos et al. (2016) left off by determining a higher resolution constraint for the scattering phase function, with observational data that spanned a greater range of scattering angles. The increased resolution and scattering angle range were achieved through the development of a new Navcam activity onboard MSL specifically designed for this purpose. This new observation labeled the ‘Phase Function Sky Survey’ (PFSS) was implemented regularly during the ACB season to document Martian WICs at different times of day, over a wide range of scattering angles. The resultant data was compared with previously developed composite EPF and RT modeled phase functions from Clancy and Lee (1991) and Clancy and Wolff (2003) along with the modeled phase functions of 7 randomly oriented ice crystal habits from Yang et al. (2010), in order to constrain dominant ice crystal geometries and the observed phase function curve for Martian WICs. The shape of the normalized phase function was more relevant than the absolute magnitude of the phase function, for the comparison of our results to the modeled curves. This is because phase functions in the literature are typically normalized to unity over scattering angles 0°–180°. After derivation, our phase function was normalized and then a statistical approach was used to find which modeled phase functions most closely correlated with our observed results.