OBSERVATIONS OF THE MARS YEAR 35 E (EARLY) LARGE-SCALE
REGIONAL DUST EVENT
D. M. Kass, A. Kleinböhl, Jet Propulsion Laboratory/California Institute of Technology, Pasadena, USA
(David.M.Kass@jpl.nasa.gov), J. Shirley, TORQUEFX, Simi Valley, USA, B. A. Cantor, Malin Space Science
System, San Diego, USA, N. G. Heavens, Space Science Institute, Boulder, USA and London, UK.

Introduction:
In Mars Year 35 a large-scale regional dust event
started at Ls ~ 35° and lasted until Ls ~ 50°. This is
during northern spring when interannual variability
is usually low and generally only local and small
regional dust events occur. This absence of large
dust events makes this season attractive for landing
spacecraft on Mars, so an unusually large dust event
at this season is of interest for forecasting weather
conditions for future landings. We designate the
event as the MY 35 E (for early) large-scale dust
event to distinguish it from the traditional perihelion
season A, B and C events as well as late northern
summer Z events [1, 2]. We use a combination of
MARCI [3] and MCS [4] observations to examine
the context, development and evolution of this unusual dust event.
The Typical Northern Spring Season:
The northern spring (early part of the aphelion
season) is generally a cold, cloudy and relatively
dust free season. The seasonal behavior is partly
driven by Mars approaching aphelion (Ls = 71°) and
thus it is relatively far from the sun which and reduces the amount of heating. Figure 1 shows the
relatively cold temperatures of the northern spring
season. In most years, the weather is limited and the
atmosphere strongly resembles the median behavior.
The atmosphere is dominated by the aphelion water
ice cloud belt [5] that forms in the tropics shortly
after the start of the season. The clouds are particularly prominent over the highest of the volcanoes,
especially Olympus Mons, the Tharsis Montes, Alba
Patera and Elysium Mons (Figure 2).

Figure 1: Median zonal mean temperature at 50 Pa
(~25 km) for northern spring season from MCS. The
median is calculated across MY 29 through MY 35.

During the northern spring, the global atmospheric circulation transitions from an equinoctial two cell
system to a solstitial single cross equatorial cell with
the upwelling branch in the northern hemisphere and
downwelling around the southern polar vortex. This
transition is also associated with the disappearance
of the northern polar vortex.
MARCI Observations:
MARCI daily global maps record the initiation,
development and growth of the MY 35 E event.
Figure 2 shows from May 31, 2019 (Ls = 32.8°) the
initiation of the event as a textured local dust storm
northwest of Olympus Mons (indicated by the red
arrow). This is a common location for textured local
dust storms to form, but not normally at this season
[6]. Otherwise, the weather on this sol is typical for
the season with thick (white) water ice clouds
crowning all of the volcanoes (the green arrow
points to the one over Alba Patera). The general
aphelion cloud belt is also visible as a light bluewhite haze over the tropics (most noticeable over the
darker terrains).
Over the next four days (through June 4th, Ls =
34.6°), the local dust storm expanded into a regional
dust storm. It primarily expanded towards the eastnortheast, basically in the direction of Alba Patera
and then beyond. By June 4th, it was latitudinally
still fairly narrow (no more than 30° wide in latitude), but extended almost to 300° East. By this
point, the dust was clearly interacting with the mountain flow around Alba Patera. The dust still remained primarily to the north of the volcano. The
clouds crowning the volcano had essentially disappeared, although the volcano itself was still visible.
On the following day (June 5th, Ls = 35.1°, not
shown), the dust started to expand southward towards the Tharsis plateau. The largest southward
expansion of the dust was to the south of Alba
Patera. The volcano also became indistinct, indicating that dust was being lofted well above its top—
possibly due to being entrained in the anabatic flows
(up-slope daytime flows). The dust also expanded
another ~15° east in longitude. It is unclear if this
was just dust advected from the lifting centers or if
there was an expansion of the active lifting regions.
Four days later (June 9th, Ls = 36.9°), the dust
reached its maximum visible extent (Figure 3). The
dust (outlined by the red arrows in Figure 3) now
extended well onto the Tharsis plateau as well as

surrounding Olympus Mons and almost reaching
Ascraeus Mons (the northernmost of the Tharsis
Montes). The eastern extension of the dust had contracted back to the west and now did not extend
beyond ~285° East. This day marked the end of the
dust lifting associated with the event. Also, by this
time, the clouds above Olympus Mon disappeared as
they did earlier over Alba Patera (green star) while
those above the Tharsis Montes were very thin.
Over the next several days, the dust abated in the
visible imagery as the thick dust in the boundary
layer sedimented back onto the surface. However,
the atmosphere was still quite hazy, with noticeably
more dust than usual for the season.

MCS Observations:
MCS acquired observations of the MY 35 E
large-scale regional dust event, providing vertical
profiles of temperature, dust and water ice in the
atmosphere throughout its lifetime [7-9]. MCS observations are also used to determine dust and water
ice column opacities [10]. The observations clearly
show the development and evolution of a large-scale
regional dust event with a temperature response
ultimately affecting most of the atmosphere. By the
time the event initiated, the atmosphere had mostly
transitioned into the solstitial single cell circulation.
The circulation at the time was upwelling in the
north and the reverse of what it usually is during

Figure 2: MARCI daily global map for May 31, 2019 (Ls = 32.8°). This is the sol when the large-scale
regional dust event initiated. The red arrow points to the initial local dust storm. The green arrow points to
the water ice clouds above Alba Patera.

Figure 3: MARCI daily global map for June 9, 2019 (Ls = 36.9°), the peak extent of the event in the visible
imagery. The red arrows outline the dust cloud. The green star indicates the location of Alba Patera.

large-scale regional dust events due to the different
season.
At 50 Pa (~25 km i.e., well above the boundary
layer) in the daytime temperature, the initial signature of the event is a very subtle change on June 3rd
(Ls = 34.2°). This is primarily a modest warming of
a region centered around 270° E and 50° N—
basically a northern extension of the mid-latitude
warm region. The first clear signature is not until
two days later (June 5th, Ls = 35.1°). That day shows
a clear warming (~5 K) centered around 270° E and
40° N, or close to Alba Patera. This matches the time
when the MARCI images clearly show signatures of
dust lofting around the volcano. The next day shows
the first hint of a dynamical response in the southern
mid latitudes in the descending branch of the overturning circulation. Over the following days (by
June 9, Ls = 36.9°), the 50 Pa response had significantly strengthened (with temperature increases of
~15 K) and, at least in the north, the warming spread
to all longitudes. The event continued to produce
increasing temperatures at 50 Pa for almost another
week.

Figure 4: Zonal mean temperature at 50 Pa (~25
km) for the MY 35 northern spring season.

Figure 5: Difference in the zonal mean 50 Pa temperatures between MY 35 and the Median (MY 35 –
median), indicating the dust event’s perturbation.
The zonal mean 50 Pa daytime temperatures responded quite strongly to the MY 35 E event (Fig-

ures 4 and 5). Comparing figures 1 and 4 shows the
strong impact that the event had on a normally quiet
and cold season. Figure 5 shows the difference between MY 35 and the median year. The northern (or
primary) warming reached zonal mean temperatures
of almost 20 K above the seasonal values. The
southern warming was not as strong, but still reached
temperatures ~15 K above the seasonally expected
values. Both the north and south temperatures
peaked at Ls ~ 39°.
The MY 35 E event also shows up clearly in the
dust and water ice fields. The dust column shows
strong development and elevated dust in the vicinity
of Alba Patera, extending as far as the equator and
60° N. A relatively thin dust haze was detectable at
all longitudes in the latter part of the event. Initially,
there was only a modest increase in dust in the
southern hemisphere, just outside the vortex—
presumably a small amount transported south by the
overturning circulation. However, dust did start to
appear in the southern tropics at Ls ~ 40° and eventually extended as far as 40° S. As seen in the
MARCI images, the water ice of the aphelion cloud
belt was significantly reduced during the growth
stages of the event, through Ls ~ 40°. The ice did
reappear afterwards and rapidly returned to the seasonally expected opacity and latitudinal coverage.
Comparison to “C” Events:
In some ways, the MY 35 E large-scale regional
dust event is similar to C events that occur after the
solstice in the perihelion season (it also shares some
similarities with the A events), with an expected
north/south swap due to the location of the sub-solar
point and the direction of the circulation. In all these
events, there is a direct heating (and the primary dust
lifting and lofting) in the “sunny” (spring or summer) hemisphere with a strong dynamical response
in the opposite hemisphere in the descending branch
of the overturning circulation. The overall temperature response indicates an enhancement in the thermal tides in both cases. Likewise, the duration of the
E and C events is similar at ~15° of Ls. Another
similarity is the relatively modest fall/winter hemisphere dynamical response with usually a short duration—often shorter than the primary warming in C
events as seen in the E event. One noticeable difference is the peak warming (in the 50 Pa zonal mean
temperature) relative to the “background” conditions. The E event had ~20 K of warming. Some C
events only have 15 K to 20 K of warming, but most
have > 30 K with the strongest exceeding 35 K of
warming. This is not surprising since the sun is
further from Mars (and thus solar irradiance is much
lower) at the season of the E event. Another difference was that in C events, the initiating local/regional dust storm forms in the winter (northern)
hemisphere and travels down one of the crossequatorial dust storm tracks [11]. Overall, it is very
likely that similar dynamical forcing and responses
are triggered by both the E and the C events (and

probably the A events) despite the large seasonal
differences in the atmosphere.
Discussion:
The appearance of a seasonally atypical local
dust storm NW of Olympus Mons might indicate
there was very strong northern hemisphere baroclinic
activity for the season in MY 35. While there are
northern hemisphere local dust storms at this season,
they tend to be polar ones [12, 13] and thus may not
interact with Alba Patera. The ruffled textured structure is indicative of convective rolls in the initial dust
lifting. The orientation of the convective rolls is
indicative of an east-west wind in the region and the
initial transport of the dust to the east indicates it was
a strong westerly flow that triggered the event and
governed its initial growth and spread.
It is clear that a key event in the evolution of the
dust storm was the interaction with Alba Patera,
presumably through mesoscale volcano driven
winds. In particular, the interactions with Alba
Patera are what drove the event from a medium sized
regional dust storm into a full large-scale regional
dust event that affected the thermal structure nearly
globally. It is unclear if without the assistance of
Alba Patera anabatic flows (plus possibly those of
Olympus Mons later in the event) the storm would
have lofted the necessary amount of dust well above
the boundary layer into the lower atmosphere. However, it is not certain that only the interaction with
mountain circulations caused the transition since
dusty deep convection is seen over Olympus Mons
and the Tharsis Montes at seasons without large
scale regional dust events (although this is usually
after then northern solstice) [2, 14]. It is also interesting to consider if Alba Patera modified flows were a
large part of initially driving the dust south and up
onto the Tharsis ridge (perhaps katabatic flows rapidly transported some of the dust from above/on the
southern flanks south overnight on June 4th/5th). It is
probable that interactions with the mesoscale circulations around Tharsis, the Tharsis Montes, and Olympus Mons also played an important role in the later
evolution of the dust event.
It is interesting to note that the MY 35 E event
occurred shortly after the MY 34 global dust event.
However, it must be remembered that there was also
a strong MY 34 C large-scale regional dust event
between the two [15]. Still, one has to ask the question of whether there was an unusual dust distribution that allow the initiation of the dust storm or
allowed it to grow to regional scale. Likewise, one
may ask whether some residual effect from the MY
34 global dust event led to an unseasonably strong
circulation that might be able to initiate such an
unexpected local dust storm.
Acknowledgements:
This work was performed at the Jet Propulsion
Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space

Administration. © 2022, California Institute of
Technology. Government sponsorship acknowledged.
References:
[1] Kass, D. M. et al. (2016) GRL, 43(12), doi:
10.1002/2016GL068978.
[2] Heavens, N. G. et al. (2019) JAS, 76(11), doi:
10.1175/JAS-D-19-0042.1
[3] Bell, J. F. et al. (2009) JGR, 114, E08S92, doi:
10.1029/2008JE003315
[4] McCleese, D. J. et al. (2007) JGR, 112, E05S06.
[5] Clancy et al., ACB
[6] Heavens, N. G. (2017), JAS, 74(4), doi:
10.1175/JAS-D-16-0211.1
[7] Kleinböhl, A. et al. (2009) JGR, 114, E10006.
[8] Kleinböhl, A. et al. (2011) JQSRT, 112, 15681580.
[9] Kleinböhl, A. et al. (2017) JQSRT, 112, 511-522.
[10] Kleinböhl, A. et al. (2022) Seventh Workshop
on Mars Atmosphere Modeling and Observations.
[11] Wang, H. and Richardson, M. I. (2015) Icarus,
251, doi: 10.1016/J.ICARUS.2013.10.033
[12] Cantor, B. A. et al. (2010) Icarus, 208, doi:
10.1016/j.icarus.2010.01.032
[13] Cantor, B. A. et al. (2018), Icarus, 321, doi:
10.1016/j.icarus.2018.10.005
[14] Heavens, N. G. et al. (2015) GRL, 42(10), doi:
10.1002/2015GL064004
[15] Chaffin, M. S. et al. (2021) Nat. Astronomy,
5(10) doi: 10.1038/s41550-021-01425-w