Using surface pressure variations to study atmospheric disturbances generated by the diurnal heating

Harmonic analysis was carried out on the pressure signal from hundreds of ASOS stations in CONUS. The observed diurnal pressure signals are studied and compared with the linear model results. Our interpretation of the data is guided by the assumption that the signal is the sum of three parts 1) Global atmospheric tide driven mostly by solar heating of the stratosphere; 2) Continentally enhanced tide driven by surface heat fluxes; 3) Mesoscale disturbance related to variations in the earth surface (mountain, coast, or land cover). These disturbances are driven locally by surface heat flux gradients, but they can propagate away using various mechanisms: e.g. gravity waves, PV pulses, storm dynamics.

The studies of the observed diurnal pressure signal in different regions show different properties. In Florida, the diurnal pressure signature is weak, so the mesoscale part is hard to determine. Here, we take the atmospheric tide in the literature as a reference. After subtracting the estimated atmospheric tide, we get a fairly smooth pattern. The centers for the phase and the amplitude are displaced from each other. Our linear theory analysis shows that such a displacement can occur when a mean wind is used. In Great Plains and the Mid-West, the region shows a strong diurnal pressure signature. The phases everywhere (relative to Local Solar Time are quite similar indicating a tide-like behaviour. The amplitude is 83 Pascal, 4 times as big as the global tide however. We suppose that this is a “continentally enhanced tide”, driven by surface heat fluxes, distributed upwards through dry and moist convection. When the average diurnal sinusoid is subtracted vectorially from the data, the residual shows a wide variation in phase. This might be noise, but when the residual phase is plotted against longitude, there is a clear slope indicating eastward movement. The speed is about 25m/s. This is probably Carbone's eastward propagating storms. We don't know the mechanism: PV pulses, IGW, or storm propagation with cold pool dynamics. The amplitude decreases eastward, suggesting that the source is the Front Range. In Dry Western Valleys, the pressure oscillation is large and the phase is early compared with that of the diurnal atmospheric tide. This probably represents simple heating of the air trapped in the valleys. No buoyancy adjustment or gravity wave generation can occur because of the valley walls. The tidal correction is almost negligible here as the local signal is so large.