Week of September 12, 2005 research report

WRF PV profiles


Greg, I kept what I showed you yesterday and incorporated the new plots for today into it with this table. The new plots are in the column highlighted in red with "New." In these calculations, I used both a forward and centered difference, as well as following the gridpoint at the center and for averaging within the 280 K tropopause theta contour. The difference with these is as discussed today, where I took the EPV differences following the vortex instead of the EPV differences at one gridpoint. I don't know what went wrong here, espeically for in the upper levels. I'll sit on this and think more about what could be wrong tonight, just wanted to show you...it's a pretty simple calculation so I'm not sure what could be wrong. Any thoughts?

Ertel PV vertical profiles
Vertical profile is of potential vorticity and potential vorticity changes at the potential temperature core (gridpoint) of the vortex (6-hour change) Vertical profile is of potential vorticity and potential vorticity changes at the potential temperature core (gridpoint) of the vortex (12-hour change)
Vertical profile is of potential vorticity and potential vorticity changes averaged within the 280K tropopause potential temperature contour (6-hour change)
Plots without advection calculated
Fixed gridpoint ***NEW*** Following the flow
Change in Ertel PV vs. change from diabatic tendencies (full vertical profile) Change in Ertel PV (centered difference) following vortex core vs. change from diabatic tendencies (full vertical profile)
Change in Ertel PV vs. change from diabatic tendencies (lower levels vertical profile) Change in Ertel PV (centered difference) following trop. theta < 280 K vs. change from diabatic tendencies (lower levels vertical profile)
Change in Ertel PV (forward difference) following vortex core vs. change from diabatic tendencies (full vertical profile)
Change in Ertel PV (forward difference) following trop. theta < 280 K vs. change from diabatic tendencies (lower levels vertical profile)
Profiles with advection. Uses centered differencing in time and space to estimate and instantaneous winds to calculate advection.
Change in Ertel PV vs. change from diabatic tendencies (full vertical profile) Change in Ertel PV vs. change from diabatic tendencies (lower levels vertical profile)



Old plots. First set shows the scaling of the diabatic contributions compared to full epv tendency. Second set shows diabatic + advection contributions except with average winds at time t used in advection term. Third set shows diabatic + advection contributions except using adv(pvor) in GEMPAK as advection term.


Please download a PDF that explains these plots here. They have to do with verifying the tendencies as we discussed last week. The first section is a bit detailed and really just goes through how I made the plots in gory detail, and the second section gives the bottom line results of what is plotted.
Profiles without advection. These give the scales of the changes in EPV compared with the changes due to the diabatic tendencies.
Change in Ertel PV vs. change from diabatic tendencies (full vertical profile)
Change in Ertel PV vs. change from diabatic tendencies (lower levels vertical profile)
Profiles with advection I: Calculating PV advection as discussed in the .pdf. The green line should at least be close to the red line, however it isn't really, especially in the upper levels.
Change in Ertel PV vs. change from diabatic tendencies and change from diabatic tendencies + advection (full vertical profile)
Change in Ertel PV vs. change from diabatic tendencies and change from diabatic tendencies + advection (lower levels vertical profile)
Profiles with advection II: Calculating PV advection using adv(pvor) in Gempak. This looks a little closer to me then in the above plots.
Change in Ertel PV vs. change from diabatic tendencies and change from diabatic tendencies + advection (full vertical profile)
Change in Ertel PV vs. change from diabatic tendencies and change from diabatic tendencies + advection (lower levels vertical profile)