Environmental Engineering Reference
In-Depth Information
the PBL heights estimated by the ACM2 in the WRF model simulations.
Figure 1
shows modeled and observed values for four of these profilers. The lines rep-
resent the median values at each hour of the day for the simulation period
(August 1-October 15) and the bars indicate the 25% and 75% values for each
model and observed distribution. For three out the four sites shown the model
overpredicts the median PBL height by about 100-300 m for most of the day. The
interquartile ranges, however, almost always overlap. It is not clear how much of
these differences are due to model error and how much is due to differences in
technique. For example, the late afternoon collapse of the model PBL heights that
often leads to underprediction is probably an artifact of the model's PBL height
algorithm that is very sensitive to the near surface temperature and surface heat
flux. The profiler PBL height algorithm, on the other hand, is more sensitive to
indicators of ambient turbulent intensities. Thus, residual turbulence can lead to
gradual declines in observed PBL height while the model algorithm senses the
initial stabilization of the lowest layers resulting in very low PBL heights. It is
conceivable that the mid-day overpredictions could have a similar algorithmic
cause.
Fig. 1.
Comparison of modeled and observed PBL height for August 1-October 15, 2006
4. Aircraft Measurements
Aircraft measurements were important components of the TexAQS II experiment.
The NOAA WP-3D Lockheed Orion aircraft (P3) made flights over the east Texas
region during September 11-October 12, 2006. On several occasions the P3 flew a
spiral flight path that extended well above the PBL. These sections of the flight
