[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #WYC-198714]: IDV 4.1u1: Vertical cross section plots conforming to variable surface elevation

> 
> On Dec 2, 2013, at 2:14 PM, Unidata IDV Support 
> <address@hidden<mailto:address@hidden>>
> wrote:
> 
> IDV support,
> 
> I’m using IDV v4.1u1 (latest nightly build) to plot vertical cross sections 
> of WRF-ARW model output. This output is on sigma coordinate surface (and 
> hence terrain following at the surface), but when I plot data (such as 
> vertical velocity, which is defined on grid surfaces that include the terrain 
> surface itself), the bottom edge of the plot does not conform to the 
> superimposed terrain elevation plot—rather, it seems to cut off at a fixed 
> elevation, around 190 meters (which intersects the terrain profile).
> 
> Dave,
> I discussed your model result with one WRF support people at MMM, she said 
> the gap between data and the terrain is normal. If you want to make the gap 
> smaller, you can change the sigma level to 1, 0.997, .... instead of 1, 
> 0.993..., some software does interpolate the data to make it looks better. I 
> am not sure if this is the case for the ncl.
> 
> Yuan,
> 
> Since the plot I'm talking about is a vertical cross section, it presumably 
> uses data from "All levels", not any particular sigma level, and that's what 
> I specify when I create the plot. Specifying a particular level doesn't 
> change anything, as far as I can tell. (The IDV is smart enough to ignore the 
> inconsistency between specifying a vertical cross section plot and a 
> particular level at the same time.)
> 
> Using data from all sigma levels should include data from the lowest sigma 
> level (in the case of vertical velocity, that's sigma = 1, which is the 
> terrain surface itself). Hence, ideally the bottom edge of the vertical cross 
> section plot should conform to the terrain profile and not intersect it, 
> assuming that the terrain profile and the vertical cross section are plotted 
> on the same vertical scale.
> 
> However, although the bottom of my vertical cross section plot comes close to 
> conforming to the terrain profile at the highest terrain elevations, it 
> doesn't come very close where the terrain elevation is lower. (The bottom 
> edge of the plot is not smooth but seems to outline the model's rectangular 
> grid cells, at least where the plot skirts the highest terrain elevations.)
> 
> One question is, what does the IDV use to determine the vertical coordinate 
> of each grid point in the field that it is contouring? If it is the value of 
> sigma (between 0 and 1, with 1 at the bottom and 0 at the top of the domain), 
> then where does sigma = 1 begin on the plot's vertical scale? (Should it be 
> same as z = 0 for the topographic profile?) And why wouldn't the bottom edge 
> of the plot be a straight, horizontal line (corresponding to the same value 
> of sigma)? (In my plot, the bottom of the vertical cross section is mostly a 
> straight, horizontal line, but it indents upward in distinct, rectangular 
> steps to stay above the highest terrain.)
> 

Dave,
     In the IDV, the sigma levels are transferred to pressure coordinate. I 
have the feeling that only base state pressure being counted in the transfer 
process. This might be the reason for the gap.
     By the way, I will be in the AGU next week. Likely, I will spend a lot of 
time in the UCAR booth and I will like to meet you if you are around. Send me 
email offline.

Yuan
> On the other hand, if the IDV uses height above sea level (rather than sigma) 
> of each grid point in the cross section to plot contours, then why doesn't 
> the bottom of the contour plot more closely conform to the terrain profile, 
> which is also based on height above sea level? (To use height above sea level 
> of each grid point in the cross section, the IDV would have to be smart 
> enough to construct that height by adding the perturbation geopotential to 
> the base-state geopotential, which are the only two pieces of information the 
> height of 3-D grid points in the WRF-ARW model output file, but I'd be 
> surprised if it would do that. The same issue would arise if it used pressure 
> as its "height" information.)
> 
> I've attached a screen shot to show the discrepancy--I don't think I did that 
> last time. The white line is a terrain profile aligned with the vertical 
> cross section of vertical velocity (a color-fill contour plot). The lowest 
> grid level for vertical velocity is the terrain surface.
> 
> -- Dave
> 
> [cid:663FABB8-AFB3-4A31-B9F1-420D9A0BC092]
> 
> 
> Yuan
> 
> Is there a way to get the bottom of the plot to conform to the underlying 
> terrain? (I know this is possible using NCAR graphics, for example). And what 
> determines the apparent elevation of the bottom of the plot?
> 
> I’ve included an IDV bundle to illustrate the point. It shows (among other 
> things) a vertical cross section plot of vertical velocity with horizontal 
> wind speed contours superimposed, along with a terrain profile. The data come 
> from a NetCDF file located at 
> http:://virga.sfsu.edu/data/wrf/<http://virga.sfsu.edu/data/wrf/><http://virga.sfsu.edu/data/wrf/>wrfout_2013-11-21_12_GFS_CenCal_BayArea_EastBay.1.nc
> 
> — Dave
> 
> 
> ****************************************************************
> *                                      |         __ __    \|/  *
> *   Dr. Dave Dempsey                   | )   ^  /|| ||\  --0-- *
> *   Dept. of Earth & Climate Sciences  |) ) ^  / ||_|| \  /|\  *
> *   San Francisco State University     | ) )  /  | _ |  \      *
> *   1600 Holloway Ave.                 |)   )/   || ||   \     *
> *   San Francisco, CA   94132          | )   )   ||_||    \    *
> *                                      |)  ) )   | _ |     \   *
> *   Phone:  (415) 338-7716             |  )   )  || ||      \  *
> *   FAX:    (415) 338-7705             |)   )  )~~~~~~~~~~~~~~~*
> *   Email:  address@hidden<mailto:address@hidden><mailto:address@hidden>      
>      |  )  )  )  ~  ~   ~ ~  *
> *                                      |)  )   ) )  ~   ~  ~   *
> ****************************************************************
> 
> 
> 
> 
> 
> Ticket Details
> ===================
> Ticket ID: WYC-198714
> Department: Support IDV
> Priority: High
> Status: Closed
> 
> 
> 
> ***************************************************************
> *   Dr. Dave Dempsey                             |       ^   ___       \|/    
>  *
> *   Dept. of Earth & Climate Sciences   |  ) ^     /||_||\    --0--  *
> *   San Francisco State University        | )  )    /  ||_||  \    /|\     *
> *   1600 Holloway Ave.                           |  )  )  /   ||_||    \      
>     *
> *   San Francisco, CA   94132               |  )  ) /    ||_||      \   ^   *
> *                                                                  | )  )  )  
>   ||_||        \      *
> *   Phone:  (415) 338-7716                    |  )  )  )~||~||~~~~ \~~*
> *   FAX:      (415) 338-7705                    | )  )  )  ) ~  ~  ~ ~ ~ ~*
> *   Email:   address@hidden<mailto:address@hidden>             |  )  )   )  ) 
> ) ~  ~   ~ ~ *
> ***************************************************************
> 
> 
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: WYC-198714
Department: Support IDV
Priority: High
Status: Open