Why the Diurnal Pressure Variation at Curiosity is so Large

The diurnal range of pressure observed by the Curiosity Rover Environmental Monitoring Station (REMS) is larger than that seen at any prior Martian landing site [Harri et al., JGR 2013]. Shortly after landing, the percentage diurnal variation was roughly 11%. This range compares with general circulation model (GCM) predictions at roughly 2-5 degrees resolution of 6-7%. In the GCM, the range is primarily due to the large scale thermal tides and equatorial Kelvin waves. However, the question of why the observed range was so much larger than the GCM-predicted range was the biggest early puzzle of the REMS observation campaign. A significant clue to the puzzle is provided by the fact that high resolution models (~<10 km resolution) can adequately capture the observed range, but that if these models are averaged over an area equivalent to the GCM grid cell size, they also simultaneously match the GCM-predicted ranges. The augmented range in pressure seen by REMS and also by the mesoscale models is thus a process operating on scales smaller than captured by the GCM. Indeed, the Gale Crater site is unique amongst Mars landing sites in the large degree of topographic relief on scales of ~10km. Using basic physical arguments and mesoscale modeling, we show that the augmented range of daily pressure variation measured by REMS is due to a process of hydrostatic adjustment in response to the daily cycle of air temperature. In the presence of a slope, a change in scale height (which depends only on the air temperature) demands a change in the horizontal gradient of surface pressure. Thus, during the warmer portion of the day, the pressure difference between two fixed points at different elevations along a slope will be smaller than that during the cooler portion. If the lower elevation point is held at fixed pressure, the upper point experiences a daily pressure cycle with higher pressure during the warmer daytime and lower during the cooler night. If instead, the domain-average pressure is held fixed, then a point at below average elevation experiences a daily cycle out of phase with that above average elevation. Indeed, we can show that the very low elevation of the Curiosity landing site results in such a "hydrostatic adjustment" pressure cycle, and that this component augments the thermal tide (adequately resolved by the GCM) so as to explain the large diurnal surface pressure range observed by REMS. Since surface pressure is solely a function of the overlaying atmospheric mass, the hydrostatic adjustment requires lateral mass transport on a daily basis. It is important to note, however, that this hydrostatic adjustment flow is distinct from buoyancy-driven slope circulations. Indeed, the pressure distribution needed to maintain buoyancy slope circulations against friction and the adiabatic influence of such circulations on the air temperature both work to oppose the hydrostatic adjustment and the former contributes to slight deviations of the pressure distribution from hydrostatic balance. In summary a diurnal range due to tides and large-scale waves yields a pressure cycle at GCM-mean elevation of about 65% of the observed range. We show that the hydrostatic adjustment process can explain essentially all of the remaining observed range, with the signatures of buoyancy slope flows and other propagating systems representing (non-negligible) perturbations.