Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Planetary Science. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 20 January 2021

The Planetary Boundary Layer of Marsfree

  • Aymeric SpigaAymeric SpigaSorbonne University, Laboratoire de Météorologie Dynamique(LMD/IPSL), Paris, Franec; Institut de Universitaire de France, Paris, France


The planetary boundary layer of Mars is a crucial component of the Martian climate and meteorology, as well as a key driver of the surface-atmosphere exchanges on Mars. As such, it is explored by several landers and orbiters; high-resolution atmospheric modeling is used to interpret the measurements by those spacecrafts. The planetary boundary layer of Mars is particularly influenced by the strong radiative control of the Martian surface and, as a result, features a more extreme version of planetary boundary layer phenomena occurring on Earth. In daytime, the Martian planetary boundary layer is highly turbulent, mixing heat and momentum in the atmosphere up to about 10 kilometers from the surface. Daytime convective turbulence is organized as convective cells and vortices, the latter giving rise to numerous dust devils when dust is lifted and transported in the vortex. The nighttime planetary boundary layer is dominated by stable-layer turbulence, which is much less intense than in the daytime, and slope winds in regions characterized by uneven topography. Clouds and fogs are associated with the planetary boundary layer activity on Mars.

Introduction: How and Why the Martian Planetary Boundary Layer Was Explored

The planetary boundary layer (PBL) is the lowermost part of the atmosphere, directly influenced by the surface of the planet (Stull, 1988, Garratt, 1992). The PBL is a key component of both the weather and climate of a planet. This is particularly true on Mars (Petrosyan et al., 2011). Furthermore, the Martian PBL is an original extraterrestrial laboratory to test the existing PBL theory based primarily on terrestrial measurements.

The PBL is a particularly active and dynamic part of the Martian atmosphere, characterized by rapid changes in temperature, pressure, and wind. Landers and rovers at the surface of Mars experience those intense fluctuations every day (Martínez et al., 2017). Spacecraft orbiting Mars also witness everyday phenomena associated with dynamics in the Martian PBL (Fenton et al., 2016). This intense PBL activity has to be considered to ensure the safe entry, descent, and landing of any spacecraft at the surface of Mars (Kass et al., 2003; Tyler, Barnes, & Skyllingstad, 2008).

The Martian PBL is a crucial part of the global climate system (Read & Lewis, 2004), as it is the atmospheric layer in which surface-atmosphere exchanges of heat, momentum, dust particles (a key driver of the Martian climate), and molecular species (e.g., water [H2O], methane [CH4]) take place.

The exploration of the Martian PBL began in the late 1970s/early 1980s with the Viking missions. The Viking landers’ in situ pressure, temperature, and wind measurements during three Martian years helped to establish the large atmospheric variability within the Martian PBL (Sutton, Leovy, & Tillman, 1978). The Viking landers and orbiters also helped to characterize typical PBL phenomena such as dust devils (Ryan & Lucich, 1983; Thomas & Gierasch, 1985). The temperature profile obtained during the entry of the Viking landers in the Martian atmosphere revealed a 6-km-deep daytime convective PBL (Seiff & Kirk, 1977).

Measurements on board following missions extended the available data sets further, although the quality and consistency of the Viking data sets remain unmatched. The Pathfinder mission landed in the same region as the Viking Lander 1 and evidenced further the Martian PBL variability, including pressure, wind and temperature signatures of turbulent gusts, and dust devils (Ferri, Smith, Lemmon, & Rennó, 2003; Schofield et al., 1997). The Mars Exploration rovers Spirit and Opportunity, albeit lacking dedicated in situ meteorological measurements, characterized the near-surface vertical gradients of temperature by thermal infrared spectrometry (Smith et al., 2006, Spanovich et al., 2006). The Mars Phoenix lander featured both a meteorological station and laser-induced sounding (LIDAR), which permitted the unveiling of exotic PBL phenomena specific to the Martian polar regions (Moores, Komguem et al., 2010; Whiteway et al., 2009). Mars Science Laboratory, or the Curiosity rover, landed within a 3-km-deep impact crater named Gale Crater, in which topography-induced fluctuations of wind and temperature were witnessed (Haberle et al., 2014, Newman et al., 2017, Ullán et al., 2017).

From orbit, (spectro-)imaging on board Mars Global Surveyor, Mars Odyssey, Mars Express, and Mars Reconnaissance Orbiter provided information on cloud and aerosol formations resulting from Martian PBL dynamics, such as cloud streets (Madeleine et al., 2012; Malin & Edgett, 2001) and dust devils (Reiss, Spiga, & Erkeling, 2014; Stanzel et al., 2008). Radio-occultations from orbit were also instrumental in the exploration of the Martian PBL given the extent of their measurements toward the surface and their vertical resolution, which led to accurate estimates of the depth of the daytime convective boundary layer (Hinson, Flasar, Simpson, Twicken, & Tyler, 1999) and its regional and seasonal variability (Hinson, Pätzold, Tellmann, Häusler, & Tyler, 2008).

These observations of the Martian PBL have been supported and interpreted through various numerical modeling techniques. A simple yet informative method is to design a single-column representation of the atmosphere above a given site on Mars (Haberle, Houben, Hertenstein, & Herdtle, 1993; Määttänen & Savijärvi, 2004; Sävijarvi, 1999). In this type of unidimensional model, parametric methods are designed to represent turbulence in the Martian PBL: the impact of unresolved dynamical phenomena (e.g., turbulent mixing) is represented by a set of equations and adjustable parameters (Hourdin et al., 2017). Similar parametric techniques for the PBL are used in global climate models in which mesh spacing is too coarse to resolve turbulent circulations.

A key modeling strategy to understand the PBL consists in resolving the larger turbulent plumes, responsible for most of the energy transport within the PBL by three-dimensional numerical simulations at horizontal resolutions of a few tens of meters (Lilly, 1962; Sullivan & Patton, 2011). This technique, called large-eddy simulations, has provided particularly rich insights into the Martian daytime convective PBL (Kanak, 2006; Michaels & Rafkin, 2004; Spiga, Forget, Lewis, & Hinson, 2010; Toigo & Richardson, 2003; Tyler et al., 2008), from its horizontal organization to the occurrence of convective vortices with the potential to form dust devils (Nishizawa et al., 2016; Spiga et al., 2016). Large-eddy simulations have also become powerful tools in preparing for the robotic exploration of Mars and have been used to characterize atmospheric hazards in the entry, descent, and landing phase of a spacecraft (Kass et al., 2003; Rafkin & Michaels, 2003; Tyler et al., 2008).

Observations and models indicate that the intensity and nature of PBL dynamics on Mars during day and night are in strong contrast. The daytime PBL depth can reach about 10 km above the surface in the daytime (Hinson et al., 2008), when it is prone to intense turbulent convection associated with strong surface heating (Gierasch & Goody, 1968; Savijärvi & Kauhanen, 2008). Conversely, at night, convective motions are inhibited by surface radiative cooling, which creates a near-surface stable layer quickly formed at sunset and removed at sunrise. Yet the Martian atmosphere can still be very active in the nighttime: for instance, close to uneven topography (volcanoes, craters, polar caps), powerful downslope winds can develop (Spiga et al., 2011).

The goal of this article is to provide an overview of the typical phenomena taking place in the Martian PBL and how those impact the weather and climate of Mars. This article addresses the following questions: What are the energetics of the Martian PBL? What are the main characteristics of the daytime Martian PBL? What is going on in the nighttime Martian PBL? How dust is lifted from the surface in the Martian PBL? What is the impact of the Martian PBL on the formation of clouds and the transport of chemical species? For exhaustive and more detailed discussions, see articles and review papers cited herein.

Energy Balance at the Surface and in the PBL on Mars

Since the PBL is strongly influenced by the surface, it is essential to describe first the energy balance of the Martian surface. Contrary to the situation at most places on the Earth, the Martian surface is very close to radiative equilibrium between the heat conduction in the soil and the net radiation at the surface (Kieffer, Christensen, Martin, Miner, & Palluconi, 1976; Larsen, Jørgensen, Landberg, & Tillman, 2002; Savijärvi & Kauhanen, 2008), accounting for the contributions of the incoming sunlight in visible wavelengths and the net flux of infrared radiation, outgoing from the surface (thermal emission) and incoming at the surface (moderate greenhouse effect on Mars). The main difference with the Earth PBL is the fact that on Mars the sensible heat flux is small compared to the radiative contributions (Haberle et al., 1993; Sutton et al., 1978). On Mars, the exchange of heat between the surface and the near-surface atmosphere is strongly limited by the thin Martian atmosphere, where density and pressure are two orders of magnitude less than what is the case at the surface of the Earth. Another key difference between the terrestrial PBL and the Martian PBL is that the absolute quantity of water vapor and ice is very small on Mars, which means that latent heat exchanges between the atmosphere and the surface are negligible (Sävijarvi, 1999).

The fact that the Martian surface is close to radiative equilibrium has key consequences on the dynamics of the Martian PBL. Given the low thermal inertia of the Martian surface (Putzig & Mellon, 2007), typical of arid desert surfaces, the diurnal cycle of surface temperature is very large. The difference of temperature between day and night on Mars is often of the order of 100 K. Because the radiative timescale of the Martian atmosphere is also very short (Goody & Belton, 1967), this gives rise to very contrasted vertical gradients of temperature in the Martian PBL between the day and the night (Schofield et al., 1997; Smith et al., 2006).

In the daytime, the temperature strongly decreases with height near the surface. Those near-surface atmospheric conditions were best characterized by the Mars Pathfinder lander, which featured a mast with three temperature sensors at three different heights from the surface (Figure 1). Observations by Pathfinder showed that the vertical gradient of temperature at the base of the Martian daytime PBL is such that there will be a 15 K warmer atmosphere at the feet of a human standing at the surface of Mars compared to the head (Schofield et al., 1997). Vertical motions are favored by those convectively unstable conditions: the daytime Martian PBL is thus prone to an intense convective turbulence.

Conversely, in the nighttime, the radiative cooling of the surface and the atmosphere causes the temperature to increase with height near the surface. Vertical motions are strongly inhibited by those stably stratified conditions. Those nighttime conditions in the Martian PBL were also characterized by the Mars Pathfinder’s meteorological mast (Figure 1), as well as via thermal infrared spectrometry on board the Mars Exploration rovers (Smith et al., 2006).

Another consequence of the strong radiative control of the Martian surface is that its temperature is not dependent on the topographical level. The surface is as hot during the day and as cold during the night on the top of the Martian volcanoes (e.g., Olympus Mons) as it is on the surrounding plains. This creates large near-surface gradients of temperature close to topographical slopes on Mars. As a result, strong upslope daytime winds and downslope nighttime winds develop in the Martian PBL (Rafkin, Haberle, & Michaels, 2001; Savijärvi & Siili, 1993; Spiga & Forget, 2009; Tyler, Barnes, & Haberle, 2002), respectively named anabatic and katabatic winds.

The Martian PBL dynamics is strongly coupled with the surface’s large diurnal cycle. The key element driving the activity of the PBL is the near-surface gradient of temperature, not so much the surface temperature itself. This explains why the Martian PBL is very active, despite the incoming flux from the Sun on Mars being less than half the flux received on Earth. There is an additional explanation for the stronger PBL activity on Mars compared to the Earth’s. In the Martian PBL, especially within several hundreds meters above the surface, the vertical flux of net infrared radiation will often dominate over the turbulent heat flux (Haberle et al., 1993; Larsen et al., 2002; Sävijarvi, 1999, Spiga et al., 2010). The outgoing infrared radiation from the surface is absorbed in the near-surface colder Martian atmosphere by CO2 (the main component of the Martian atmosphere and a greenhouse gas) and dust, as well as H2O, albeit to a lesser extent. This strong radiative control of the Martian PBL is not encountered on Earth except in rare settings. In other words, during the daytime, the Martian PBL is not only warmed from below by the surface heated by incoming sunlight but also from inside by CO2 (and dust) absorption of infrared radiation outgoing from the surface.

Figure 1. On the left, an image of Mars Pathfinder’s meteorological mast, which measured atmospheric temperatures shown on the right for one diurnal cycle on Mars, with colors indicating the considered mast thermocouple (red: top, middle: black, bottom: blue). The Pathfinder thermocouples are located 100, 50, and 25 cm above the plane of the lander solar panels (lying on the Martian surface). The top plot on the right shows measured temperatures sampled at 4-sec intervals, to which a 2-min running mean is applied for clarity; the bottom plot on the right shows temperature deviations from the average temperature obtained by the three thermocouples.

Adapted from Schofield et al. (1997) with permission.

A last point that has to be discussed here is the surface layer, or the lowermost part of the PBL—the few tens of meters above the surface. In this layer, the smaller-scale turbulence plays a prominent role in heat and momentum transfers from the surface to the atmosphere (Foken, 2006; Monin & Obukhov, 1954). The sensible heat flux is directly determined by heat exchanges in the surface layer (Martínez, Valero, & Vázquez, 2009). The low density of the Martian atmosphere causes the kinematic viscosity to be larger (Larsen et al., 2002), which entails a Kolmogorov scale an order of magnitude larger than on Earth. This means that the transition between turbulent and laminar flow is more accessible on Mars than it is on Earth: the inertial subrange of turbulence in the near-surface atmosphere is much less important on Mars than on Earth. Close to the surface, the larger value of kinematic viscosity for Mars generally makes the near-surface atmosphere closer to laminar flow than over similar surfaces on Earth. This was confirmed by temperature and wind spectra observed by Viking and Pathfinder (Larsen et al., 2002; Tillman, Landberg, & Larsen, 1994) and would be an additional explanation for the negligible role of the sensible heat flux in the surface energy budget. Interestingly, the fact that the Martian near-surface atmosphere is prone to strong radiative heating/cooling puts in jeopardy one of the key assumptions of the Monin-Obukhov theory to compute the temperature and wind profiles in the surface layer (Monin & Obukhov, 1954; Petrosyan et al., 2011). Although this theory was found to yield reasonable results to first order to interpret in situ measurements in the Martian surface layer (Davy et al., 2010; Larsen et al., 2002; Määttänen & Savijärvi, 2004; Martínez et al., 2009; Sutton et al., 1978; Tillman et al., 1994), some refinements of the Monin-Obukhov theory, inspired by the radiatively controlled Martian environment, may be necessary in the future (Gentine, Steeneveld, Heusinkveld, & Holtslag, 2018; Spiga et al., 2018).

Characteristics of the Daytime PBL on Mars

The strong unstable conditions in the Martian daytime PBL are conducive to powerful convective turbulent motions. A so-called mixing layer develops, to remove heat from the near-surface environment and warm the PBL, causing the depth of the Martian PBL to deepen significantly from the morning to the afternoon. Large-eddy simulations show that the Martian PBL is already a couple of kilometers deep in the middle of the morning (Michaels, Colaprete, & Rafkin, 2006; Spiga et al., 2010; Tyler et al., 2008), before it reaches a depth of 5 to 10 km in early afternoon, consistent with radio-occultation measurements (Hinson et al., 2008). This is significantly deeper than the PBL depth in most terrestrial regions (Spiga, 2011). The convective motions in the Martian daytime PBL cause the near-surface air temperature, winds, and pressure to undergo strong temporal variations over tens of seconds, which were recorded by meteorological instruments on board landing spacecraft at the surface of Mars (Schofield et al., 1997; Smith et al., 2006). Thermal infrared spectrometry on board the Mars Exploration rovers recorded, for instance, temperature fluctuations of amplitude ± 2.5 K up to 100 m above the surface (Smith et al., 2006).

Modeling by turbulence-resolving large-eddy simulations reveals the Martian PBL dynamics (Figure 2) underlying the daytime mixing layer observed from orbit in radio-occultations. The daytime turbulent motions on Mars are organized similarly to other natural manifestations of convection: vertical upward and downward plumes (updrafts and downdrafts) organized horizontally as polygonal convective cells delimited by narrow updrafts with broad downdrafts in the middle (Michaels & Rafkin, 2004; Rafkin et al., 2001; Spiga & Forget, 2009). At the intersection of convective cells, horizontal vorticity can be tilted by converging upward motions to give rise to convective vortices (Gheynani & Taylor, 2011; Kanak, 2006; Nishizawa et al., 2016; Toigo & Richardson, 2003). These convective vortices in the Martian PBL may develop into dust devils, when dust particles are lifted from the surface and transported in the swirling vortices. Models indicate that convective vortices cause pressure drops from 1 to 6.5 Pa and tangential velocities from 10 to 20 m s−1 (see Table 2 in Spiga et al., 2016). This accounts for the typical values observed in situ by, for example, the Pathfinder and Phoenix lander (Ellehoj et al., 2010; Schofield et al., 1997); every lander has been able to detect pressure drops associated with convective vortices (Murphy et al., 2016). Tangential velocities within dust devils were measured from orbit by high-resolution high-cadence imaging on board Mars Reconnaissance Orbiter (Choi & Dundas, 2011), and large-eddy simulations appear consistent with those measurements.

The regional and seasonal variability of daytime PBL activity is correlated with surface temperature, controlled by the thermophysical properties of the surface (albedo, thermal inertia) and the incoming sunlight (depending on latitude and season). However, as is described in later, the Martian daytime PBL is also controlled by infrared radiation. Modeling shows that, close to the surface in the Martian PBL, the resulting strong radiative heating forces the convective processes to cool the atmosphere rather than warm it (Haberle et al., 1993; Sävijarvi, 1999; Spiga et al., 2010), as is the case on Earth and in the upper part of the Martian PBL where the turbulent heat flux dominates.

This radiative forcing influences the daytime growth of the unstable Martian PBL (Sorbjan, 2007). It also impacts the regional and seasonal variability of the depth of the daytime PBL. All other things being equal, in particular similar surface temperature conditions, which is often the case given the relative independence of surface temperature with topography, observations by radio-occultations on board Mars Global Surveyor and Mars Express showed that the daytime PBL mixing is significantly deeper over higher plateaus than over lower surrounding plains (Hinson et al., 2008). This behavior of the daytime Martian PBL is contrary to the typical behaviour in the Earth’s PBL. Large-eddy simulations showed that this results from the radiative forcing of the Martian PBL being prominent over the forcing by sensible heat flux (Spiga et al., 2010). As a result, the Martian high plateaus are particularly prone to powerful daytime convection, which explains the occurrence of dust devils over the Martian volcanoes (Reiss et al., 2009). This deep daytime PBL mixing above high topographical obstacles might account for the injection of dust particles at high altitudes in the vicinity of Martian volcanoes, although the transport by anabatic upslope circulations is likely to be dominant (Rafkin, Sta. Maria, & Michaels, 2002).

Another source of regional variability in the daytime convection in the Martian PBL is actually associated with anabatic winds. Over the slopes of Martian craters and volcanoes, as a result of near-surface gradients of air temperature, anabatic upslope winds can reach several tens of meters per second, according to regional climate modeling (Rafkin & Michaels, 2003; Tyler & Barnes, 2015; Tyler et al., 2002). Those winds are responsible for upward transport of water vapor over Olympus Mons and the Tharsis volcanoes, hence the formation of characteristic clouds at their summits (Michaels et al., 2006; Spiga & Forget, 2009; see also later discussion and Figure 4). The observations by the Mars Science Laboratory Curiosity within Gale Crater showed that the PBL activity is weaker than in surrounding plains (Moores, Lemmon, Kahanpää, et al., 2015). Regional climate modeling, using a parametric model for PBL mixing, suggested that the depth of the daytime PBL within Gale Crater is particularly low according to Martian standards—about 1 km deep in the afternoon (Tyler & Barnes, 2015). This low PBL depth is caused by the compensating subsidence higher up in the PBL, associated with the near-surface upslope flow according to the conservation of mass. As a result, turbulent motions and convective vortices are less strong in Gale Crater than in neighboring plains. More generally, large-eddy simulations tend to show that the growth of the daytime Martian PBL is influenced by regional to large-scale winds (Tyler et al., 2008).

Figure 2. Results of high-resolution large-eddy simulation of the turbulent convection in the Martian daytime PBL at an horizontal resolution of 5 m. The figure shows a cross-section at an altitude of 60 m above the surface of vertical velocity (a, c) and vertical vorticity (d), as well as a three-dimensional view on vertical vorticity (b). Figure (a) highlights the horizontal organization of the convection in polygonal cells, while Figure (b) shows the resolution of convective vortices (a particular structure is shown in Figures [c] and [d], which offer a magnified view of the squared area in Figure [a]).

Figure used from Nishizawa et al. (2016) with permission.

A final source of both regional and seasonal variability of daytime turbulence in the Martian PBL is the occurrence of dust storms in the Martian atmosphere, at spatial scales ranging from local to planet- encircling (Cantor, 2007; Malin et al., 2008). In the regions and seasons impacted by a dust storm, the atmospheric loading of dust particles (dust opacity) is very high, which tends to severely reduce the incoming sunlight reaching the surface and so reduce the daytime maximum surface temperature. As a result, the daytime PBL convection on Mars is weak in dust storm conditions.

Nighttime Circulations in the Martian PBL

At sunset, due to radiative cooling, stably stratified conditions build up quickly in the Martian PBL, leading to the collapse of the Martian convective PBL described previously. Yet the stably stratified free atmosphere above the convective PBL is perturbed in late afternoon conditions by daytime updrafts, which gives birth to gravity waves in which buoyancy is the restoring force. The propagation of those gravity waves keeps the PBL active a few hours after the collapse of the daytime PBL (Spiga & Forget, 2009). Furthermore, after the Martian PBL collapses in the evening, a nighttime low-level jet arises above the nighttime PBL (Savijärvi & Siili, 1993). A similar phenomenon occurs when the Martian PBL depth weakens in dust storm conditions (Joshi, Haberle, Barnes, Murphy, & Schaeffer, 1997).

Nighttime turbulence in the Martian PBL is much less intense than in the daytime PBL, which is prone to unstable conditions. The rapid radiative cooling of the surface leads to ultrastable vertical temperature gradients in the nighttime Martian PBL, where vertical motions are being inhibited by those negative buoyancy conditions. As a result, the nighttime Martian PBL is very shallow (a couple hundred meters deep) and mostly prone to small-scale turbulence caused by horizontal wind shear close to the surface. The strong contrast in Martian PBL activity between the strong buoyancy-driven deep turbulence in the daytime and the weak shear-driven shallow turbulence in the nighttime was actually predicted by calculating the Richardson number (Gierasch & Goody, 1968) a decade before the Viking observations would confirm this behavior. Interestingly, the Curiosity rover observed in the night pressure drops characteristic of daytime convection in the Martian PBL (Ordonez-Etxeberria, Hueso, & Sánchez-Lavega, 2018). This observation is unexpected given the low turbulent activity in the nighttime Martian PBL and might be related to nighttime slope circulations.

The nighttime circulations in the Martian PBL are much more intense close to topographic highs and lows, such as mountains, craters, and canyons. Downslope katabatic flows develop as the near-surface atmosphere along the slope is being radiatively cooled at night (Rafkin & Michaels, 2003; Savijärvi & Siili, 1993; Spiga & Forget, 2009; Toigo & Richardson, 2003; Tyler et al., 2002). Despite the acceleration of gravity being three times weaker on Mars than on Earth, the katabatic acceleration is stronger on Mars than it is on Earth (Blumsack, Gierasch, & Wessel, 1973; Spiga, 2011; Ye, Segal, & Pielke, 1990) because the near-surface gradients of temperature are larger above the radiatively controlled surface of Mars, where surface temperatures are nearly independent of surface elevation (see earlier discussion). While the occurrence of katabatic events on Earth is strongly dependent on synoptic conditions, on Mars katabatic winds develop every night over craters and volcanoes, especially where steep slopes are encountered. Mesoscale modeling shows that katabatic winds could reach 30 to 40 m s−1 about 50 to 100 m above the surface over the Valles Marineris canyon (Rafkin & Michaels, 2003; Tyler et al., 2002) and over Olympus Mons and the Tharsis volcanoes (Michaels et al., 2006; Rafkin et al., 2002; Spiga et al., 2011), that is, the most prominent topographical obstacles. Katabatic winds in the less prominent craters and mountains can be modulated by the global and regional circulation, as is the case in Gale Crater (Pla-Garcia et al., 2016; Rafkin et al., 2016; Steele, Balme, Lewis, & Spiga, 2017) where the Mars Science Laboratory rover is operating. Wind measurements within Gale Crater by this rover (Haberle et al., 2014; Newman et al., 2017; Viu´dez-Moreiras et al., 2019) are key to understanding slope winds.

The indirect effects of katabatic winds in the Martian PBL can be observed through a variety of phenomena. Erosion features such as wind streaks on slopes of craters and volcanoes or frost streaks on permanent polar caps are compliant with the wind direction predicted by mesoscale models (Fenton, Toigo, & Richardson, 2005; Greeley, Kuzmin, Rafkin, Michaels, & Haberle, 2003; Kuzmin, Greeley, Rafkin, & Haberle, 2001; Massé et al., 2012; Toyota, Kurita, & Spiga, 2011). The presence of a thick “haze” of dust particles within the Valles Marineris canyon in early morning, detected from orbit (Inada et al., 2008; Mishra, Chauhan, Singh, Moorthi, & Sarkar, 2016), possibly results from the transport exerted by the katabatic winds converging at the bottom of the crater (Tyler et al., 2002). As is the case on Earth (Nylen, Fountain, & Doran, 2004), katabatic winds may also cause the atmospheric temperature to rise in the Martian PBL along slopes, as a result of adiabatic warming. Since those near-surface katabatic winds also increase the sensible heat flux (i.e., surface-atmosphere transfer of heat), this causes the surface to be warmer over Martian steepest slopes at night, which was observed by thermal infrared spectrometry by orbiting spacecraft (Spiga et al., 2011). This might be the only situation on Mars where sensible heat flux dominates over radiative contributions in the surface energy budget. Peculiar elongated clouds are also associated with nighttime katabatic winds.

Dust Lifting and Dust Devils in the Martian PBL

The dust cycle is a key component of the Martian climate system, given that airborne dust particles both absorb and scatter incoming sunlight and absorb and emit infrared radiation, activities essential for the radiative forcing of the thin Martian atmosphere (Madeleine, Forget, Millour, Montabone, & Wolff, 2011). The Martian PBL is the place where the lifting of dust particles occurs, hence it plays a central role in the Martian dust cycle.

There is an interplay of complex mechanisms governing the injection of Martian dust particles beyond the lowest couple of meters in the Martian PBL (Greeley & Iversen, 1987; Kok, Parteli, Michaels, & Karam, 2012). Herein only simplified discussions are provided. Both models and wind-tunnel experiments show that near-surface Martian winds tend to catch large dust particles (whose sizes are tens to hundreds of microns) rather than the micron-sized dust particles encountered in the Martian atmosphere (Newman, Lewis, Read, & Forget, 2002; White, 1979). This is why saltation and sand-blasting of large dust particles have been proposed as possible mechanisms accounting for the lifting of those micron-sized dust particles from the Martian surface to the PBL (Greeley, 2002). Even in a situation where the large-scale and regional winds are weak, turbulent motions in the daytime PBL can raise the local wind to the threshold necessary for lifting of those large dust particles to occur (Fenton & Michaels, 2010; Mulholland, Spiga, Listowski, & Read, 2015; Newman et al., 2002).

A particularly ubiquitous phenomenon in the Martian daytime PBL is the occurrence of dust devils (Figure 3; Balme & Greeley, 2006; Lorenz et al., 2016; Renno, Nash, Lunine, & Murphy, 2000). In the distribution of wind perturbations caused by daytime turbulence, the convective vortices are responsible for most of the tail of the distribution. Should dust be available for lifting, convective vortices in the Martian PBL cause dust devils when dust particles are being lifted from the surface and advected in the swirling structure of the convective vortex, often leaving dust devil tracks on the surface (Michaels, 2006; Reiss et al., 2016; Statella, Pina, & da Silva, 2012). By combining imaging on board orbiting spacecraft (Fenton et al., 2016; Stanzel et al., 2008) and in-situ monitoring at the surface of Mars (Ferri et al., 2003; Murphy et al., 2016), it is known that in the Martian PBL dust devils’ size could range from a couple meters to hundreds of meters (Lorenz, 2011) with durations up to several hundred seconds; their associated pressure drops and wind perturbations (Greeley et al., 2010; Kahanpää et al., 2016; Murphy & Nelli, 2002) correspond to those of the associated convective vortices discussed in earlier.

Figure 3. On the left, an image of a dust devil captured in 2004 in Gusev crater by the Mars Exploration Rover Spirit (Product ID: 2N 177950967 RAD AD ND P0645 L0 C1). On the right, a dust devil caught in a 2012 image by the High Resolution Imaging Science Experiment camera on NASA’s Mars Reconnaissance Orbiter. This is late-spring afternoon occurrence in the Amazonis Planitia region. The dust devil is 800 m high, its core is about 30 m diameter, and the whole scene is about 600 m across.

Image credits: NASA/JPL-Caltech/University of Arizona (PIA15116).

There is strong seasonal variability in dust devil activity (Greeley et al., 2010; Verba, Geissler, Titus, & Waller, 2010), which appears to follow variations in incoming sunlight. Regions on Mars propitious to dust devils are evenly distributed (Fisher et al., 2005), potentially reflecting the regions where dust is available for lifting (Fenton et al., 2016). For instance, while the Spirit rover observed many dust devils (Greeley et al., 2006), the Opportunity rover did not. The regional contrast in dust devil activity may also be caused by the contrast in Martian PBL activity. The Curiosity rover did not observe many dust devils (Moores, Lemmon, Kahanpaää, et al., 2015). This may be related to the growth of the daytime PBL being limited by slope circulations (Tyler & Barnes, 2015), which could explain why the many convective vortices observed in Gale Crater (Kahanpää et al., 2016; Ordonez-Etxeberria et al., 2018) would often not be intense enough to lift dust particles.

The relative contribution of the injection of particles by dust devils in the Martian PBL to the permanent veil of dust particles in the atmosphere of Mars is uncertain (Fenton et al., 2016; Klose et al., 2016). Both global climate modeling using a parameterized module for dust devil lifting (Kahre, Murphy, & Haberle, 2006; Newman et al., 2002) and estimates of the annual dust devil flux from lander and orbiter observations (Cantor, Kanak, & Edgett, 2006; Whelley & Greeley, 2008) indicate that Martian dust devils could contribute as much as 30-% to 50% of the annual atmospheric dust loading in the atmosphere of Mars in a year without global dust storms.

Clouds and Chemical Species

The absolute quantity of water vapor (absolute humidity) in the atmosphere of Mars is low, but water-ice clouds are nevertheless quite common on Mars (Clancy et al., 1996; Kahn, 1984) because the partial pressure of water vapor is close enough to the saturation vapor pressure for the ice-vapor equilibrium given by the Clausius-Clapeyron equation in conditions of Martian atmospheric pressures and temperatures (Montmessin, Forget, Rannou, Cabane, & Haberle, 2004; Richardson, Wilson, & Rodin, 2002). As a result, water-ice clouds form in the Martian PBL and trace the PBL dynamics (Figure 4).

During the day, water-ice clouds are found at the summit of the Martian volcanoes, resulting from water vapor transport by upslope anabatic winds (Benson et al., 2003; Michaels et al., 2006; Spiga & Forget, 2009). Over Martian plains, the horizontal organization of the PBL convection as polygonal cells gives rise to “streets” of cumulus clouds detected from orbit at the summit of the daytime Martian PBL (Madeleine et al., 2012; Malin & Edgett, 2001). Thin afternoon water-ice clouds were observed in the Martian PBL from the surface of Gale Crater by the Mars Science Laboratory and appeared to be influenced by propagating gravity waves (Moores, Lemmon, Rafkin, et al., 2015).

In the night, water-ice clouds develop in specific regions prone to strong katabatic winds in the Martian PBL. Elongated water-ice clouds within the troughs of the Martian polar caps (Smith, Holt, Spiga, Howard, & Parker, 2013; Smith, Spiga, & Holt, 2015) denote the occurrence of katabatic jumps, the PBL analogs of hydraulic jumps (Spiga & Smith, 2018). Early-morning water-ice clouds with a similar elongated morphology have been observed in the Tharsis region since the Viking Orbiters (Briggs, Klaasen, Thorpe, & Wellman, 1977; Hunt, Pickersgill, James, & Evans, 1981) and have been interpreted as resulting from an atmospheric bore wave generated by a katabatic front (St. Maria, Rafkin, & Michaels, 2006). Outside the regions prone to strong katabatic winds, nighttime fog can form in the shallow Martian PBL due to radiative cooling (Savijärvi, 1995). Such late night/early morning fogs were detected directly by the Phoenix lander close to polar regions (Moores, Lemmon et al., 2010) and indirectly via measurements of sky optical depth by the two Viking landers (Pollack et al., 1977; Savijärvi, Paton, & Harri, 2018).

Figure 4. This figure shows water-ice clouds associated with Martian PBL phenomena. On the top left, imaging on board Mars Global Surveyor evidenced cumulus cloud streets caused by convective cells in the daytime PBL (Malin & Edgett, 2001; credit Calvin J. Hamilton). On the bottom left, this image from the Indian Mars Orbiter Mission Mangalyaan shows afternoon clouds at the top of Olympus Mons caused by anabatic winds (credit: ISRO). The figures on the right are from a Mars Odyssey THEMIS image and show katabatic jumps caused by the downslope currents above the Martian northern polar caps (Spiga & Smith, 2018).

The water-ice clouds detected at night by the LIDAR on board the Phoenix lander bridge the gap between the daytime and nighttime Martian PBL. The detection with the same instrument of well-mixed dust in the daytime PBL confirmed the efficiency of daytime turbulent convection in the PBL. Water vapor is similarly well mixed in the daytime PBL. This gives birth during the cold night to water-ice clouds at the top of the daytime PBL a long time after the daytime convection is suppressed (Savijärvi & Määttänen, 2010; Whiteway et al., 2009). Interestingly, water-ice crystals precipitating from this nighttime cloud toward the ground were detected by the LIDAR on board the Phoenix lander (Daerden et al., 2010; Dickinson, Whiteway, Komguem, Moores, & Lemmon, 2010). Convective motions driven by the radiative cooling within the water-ice cloud could account for those precipitating structures (Spiga et al., 2017), showing that turbulent convective motions could also be found on Mars above the shallow PBL at night.

The strong mixing in the daytime PBL, with a typical convective overturning of a couple hundred seconds, impact not only the distribution of dust particles and water vapor close to the surface but also any chemical species with a lifetime longer than the convective overturning. This means a tracer like methane with a very long photochemical lifetime (Lefèvre & Forget, 2009) is very likely to be efficiently mixed within the daytime PBL once emitted close to the surface (Webster et al., 2018).


The Martian PBL is both active dynamically and plays a key role in surface-atmosphere exchanges. Landers operating at the surface of Mars, such as the InSight mission (Spiga et al., 2018), the Mars 2020 mission (with a featured Mars Helicopter which will fly in the Martian PBL), the European-Russian ExoMars rover, as well as planned surface exploration by China, India, and Japan are needed to better understand the Martian PBL. As was discussed throughout this article, orbital observations are also of a great help to complement the in-situ observations. Studying the Martian PBL is all the more important as it is the environment in which spacecraft, and eventually humans, operate on reaching the surface of Mars.