Relative humidity and vapor amount at Jezero Crater
L. K. Tamppari, Jet Propulsion Laboratory/California Institute of Technology, Pasadena, CA, USA
(leslie.tamppari@jpl.nasa.gov), G. Martínez, Lunar and Planetary Institute, Houston, TX, USA, J. A.
Rodríguez-Manfredi, Centro de Astrobiologia (INTA-CSIC), Madrid, Spain, M. de la Torre-Juárez, Jet Propulsion Laboratory/California Institute of Technology, Pasadena, CA, USA, M. Hieta, J. Polkko, I.
Jaakonaho, M. Genzer, A.-M. Harri, Finnish Meteorological Institute, Helsinki, Finland, T. McConnochie,
Space Science Institute, Boulder, CO, USA, F. Montmessin, LATMOS, Guyancourt, France, M. Smith, NASA
Goddard Space Flight Center, Greenbelt, MD, USA, M. Wolff, Space Science Institute, Boulder, CO, USA, E.
Fischer, University of Michigan, Ann Arbor, MI, USA, P. Conrad, Carnegie Institution of Science, Washington,
D.C., USA, F. Gómez, M.-P. Zorzano, and D. Viudez-Moreiras, Centro de Astrobiologia (INTA-CSIC), Madrid, Spain, P. Patel, Jet Propulsion Laboratory/California Institute of Technology, Pasadena, CA, USA, M.
Lemmon, Space Science Institute, Boulder, CO, USA, and the MEDA Team.
Introduction:
Near-surface water content and relative humidity
are important for understanding the overall column
distribution of water, the exchange with the surface
and near-subsurface, the potential for brine or frost
formation, and general modern habitability potential
of the local environment (see e.g., [1-3]). Humidity
sensors have been used on Phoenix, MSL, and now
Perseverance to gain this understanding.
The Perseverance MEDA relative humidity
measurements at Jezero crater add additional time
coverage and an additional location to help discern
the regional similarities and differences across Mars.
In combination with independent measurements by
the Perseverance SuperCam or TIRS instruments,
measurements of relative humidity can be used to
detect frost formation [4], to check whether the environmental conditions at the surface and in the near
surface are compatible with the formation of liquid
brines [5-6], and to assess diurnal changes in the
near-surface water content, possibly due to subsurface exchange (e.g., [7-8, 3]) or consistent with surface frost [4].
The relative humidity sensor supports the science
and human exploration goals of the rover, the sample
collection via documenting the atmospheric field
context in which the sample was acquired, and the
Ingenuity helicopter flights.
Methods:
The Mars Environmental Dynamics Analyzer
(MEDA; [1]) aboard the Mars Perseverance Rover is
a suite of environmental sensors to measure relative
humidity (HS), wind (WS), atmospheric temperature
(ATS), atmospheric pressure (PS), brightness temperature of the surface and lower atmosphere (TIRS),
atmospheric optical depth and particle properties
(RDS). Here we focus on the relative humidity sensor (HS), which is on the rover mast at about 1.6m
above the surface (Fig. 1, 2). The HS contains
two Humicap® by Vaisala, Inc. which have an active polymer that changes the sensor head capacitance as a function of atmospheric relative humidity

and temperature, measuring 0 to 100% RH. The
humicaps have PT1000 platinum resistance temperature sensors to provide accurate temperature of the
sensor chip. These two measurements, along with
the PS, can then be used to derive the volume mixing
ratio (VMR) of water vapor in the atmosphere.
Improvements in the HS from the Curiosity rover
version have been made to enable a more rapid response to the environment and to have a wider dynamic range, allowing for better resolution in the
measurements.
The MEDA RH sensor is healthy and operating nominally. Data are acquired along with other
MEDA sensors in ~1 hour Observation Table (OT)
blocks that are scheduled every other hour throughout the diurnal sol and alternate every sol to cover
even hours and then odd hours. There are two RH
instrument modes that are used regularly: Continuous
(CM) and High-Resolution Interval (HRIM). CM
will measure continuously throughout the hour block
and gives more insight into short-timescale variability, but is less accurate after the first few seconds. HRIM takes a few s of data every 5 min
throughout the hour block. The first few s are most
accurate owing to self-heating over time. This mode
is used in 24-06h period.
For more information on the MEDA humidity
sensor, see Hieta et al., this conference.

Figure 1. Artist’s concept of the Perseverance Rover on
the surface of Mars. MEDA has sensors on the mast and
on the rover itself (see also Fig. 2). The Humidity Sensor
(HS) is near the one of the Wind Sensor (WS) booms on
the mast and also near some of the Atmospheric Temperature Sensors (ATS). Image credit: NASA

Figure 2. (Left) HS, WS, and ATS locations on the Remote Sensing Mast. (Right) HS without the cover; Humicaps®, which measure the relative humidity, are on right
(white rectangles), Thermocap sensor heads next to them
[1].

Fig.4. TES zonally averaged (18-20N) column water
vapor abundances over 3 MY. (Note that Ls (x-axis) continually increments, so MY25 is 360-720, for example.)

Results:
Comparison of seasonal RH/VMR to column
abundances in Jezero. Fig. 2 shows the seasonal
behavior of nighttime (21-06h) relative humidity and
the derived volume mixing ratio (VMR). The variability of the nighttime relative humidity has decreased as the season progressed. Fig. 3 shows the
VMR, which has a seasonal peak near sol 300-320
(Ls=146.3-156.8). There is a clear VMR trend of decreasing vapor amount as spring progressed, followed by an increase later in the summer, as is seasonally expected (Fig. 4, 5).
Fig. 5. TES water vapor abundances over 3 MY located over Jezero (5º lat, 7.5 deg lon, 5 deg Ls averaging).

Fig. 2. Diurnal atmospheric relative humidity (%) for sols
65-275 (Ls ~ 36-134).

Fig. 3. MEDA nighttime “best” VMR through sol 368.

Diurnal measurements. The diurnal behavior of
relative humidity is also shown in Fig. 2. As expected, the highest relative humidity values are in the
midnight-pre-dawn timeframe. Because there is so
little water on Mars, the daytime relative humidity
values approach zero and are smaller than the uncertainty for most of the ~07-18h period.
Comparing the measurements to models can both
help interpret the measurements and validate or improve the models. In Fig. 6, we show output from a
single column model (SCM) which takes into account sub-surface exchange and which is driven by
precipitable water content as taken from either the
Laboratoire de Meteorologie Dynamique (LMD)
Mars Climate Database (MCD) in Jezero crater or
from the Thermal Emission Spectrometer (TES) zonal mean measurements [9].

and/or efficiency of surface-atmosphere exchange,
combining these HS data with Supercam column
water values to better understand the vertical distribution of water, and initializing the models' column
water amounts for just the rover location and season
(e.g., non-zonal mean TES data). All of these are the
subject of future work.
Fig. 6. The daily maximum (14h) and minimum (06h) VMR
at 1.6 m for Jezero as predicted from a single-column
model (SCM; [1]). The SCM is initialized with the MCD
local precipitable water content (PWC) and also by TES
MY26 zonal mean PWC near 18oN. The nighttime VMRs
predicted are higher than observed by MEDA RH (Fig. 5,
6), but the seasonal trend appears to be similar.
Comparing the minimum values (06h) to the
nighttime values in Fig. 2, 3, it appears that the model is over-predicting by about a factor of 2 the
amount of vapor compared to what is measured. A
direct comparison of the HS measured vs. SCM
modeling minimum VMR values is shown in Fig.
7. The model does show a seasonal trend with increasing minimum vapor, peaking near Ls=150 (~Sol
3000; Fig. 6).

Figure 7. Comparison of the minimum diurnal VMR (0507h) determined from the MEDA HS data to the VMR at
06h produced using the single-column model (SCM; PlaGarcía et al., 2020). The SCM output is initialized by the
MCD local PWC (green) or via the TES MY26 zonal mean
PWC for the Jezero latitude of 18oN (blue).
The differences between the modeled nearsurface VMR and the measurements could be the
result of a number of factors. The LMD-MCD does
not include sub-surface exchange and is based on
monthly quantities. The SCM does not include horizontal advection, so it is possible that drier air flows
into Jezero crater every night that wouldn't be reflected in the model. It could also be that the nearsurface atmosphere is losing more water to the subsurface at night than predicted.
In order to obtain a better understanding of
these differences model improvements could be
made, such as investigating the surface porosity

Comparison of Jezero to Gale and Phoenix landing site. Previous spacecraft that flew relative humidity sensors were the Phoenix lander [10-13] and
the Mars Science Laboratory/Curiosity rover [14-15,
7].
The MEDA HS nighttime VMR values more
closely match those measured at Curiosity by the
REMS relative humidity sensor (Fig. 8) that what
would be expected from TES column abundances,
but the MEDA HS values are even drier than REMS
early in the season. This is surprising since the column abundances during this season are higher for the
Jezero location than for the Gale location. Later in
the season, the two are more alike.
The MEDA HS measures somewhat lower
nighttime VMR values during a similar seasonal period compared to Phoenix (Fig. 9), which is not surprising given that Phoenix was close to the northern
polar cap during northern summer when sublimation
of water ice is expected.

Figure 8. Curiosity REMS nighttime VMR values by season. Magenta values are for MY36 through Ls=90, concurrent with the MEDA HS measurements.

Figure 9. Phoenix nighttime TECP VMR values. Phoenix
operated for 142 sols between Ls=76-150.
Summary:
The Perseverance MEDA RH sensor is working
well and producing a rich dataset that is sure to expand our knowledge of water on Mars. The presentation will provide an overview of the dataset in more
depth.
Bibliography: [1] Rodríguez-Manfredi et al.,
2014, [2] Savijärvi et al., 2019a, [3] Tamppari and
Lemmon, 2020, [4] Martínez et al., 2016, [5] MartínTorres et al., 2015, [6] Rivera Valentín et al. 2018,
[7] Martínez et al., 2017; [8] Savijärvi et al., 2019b;
Pla-García et al., 2021, [10-12] Zent et al., 2009;
2010; 2016, [13] Fischer et al., 2019, [14] GómezElvira et al., 2014; [15] Harri et al., 2014.