Fwd: Surface terrain problem

Hi Eliot,

Your email is being forwarded to ' wrfhelp@ucar.edu '

The wrf_* functions were provided to NCL by WRF.

Good luck

-------- Original Message --------
Subject: [ncl-talk] Surface terrain problem
Date: Mon, 5 Jul 2010 17:24:19 -0400
From: Tardif, Elliot <etardit@uncc.edu>
To: <ncl-talk@ucar.edu>

Hello,

I am using NCL to produce surface-level WRF model output for a given
lat/lon point. I am using NCL v. 5.2.0, and have installed NCL v. 5.2.0
on a Mac (OS X 10.5.4) and on a PC (Windows Vista 64-bit). I have
noticed, however, while running NCL that it is not treating the surface
terrain correctly. As an example, in producing surface point output for
an ASOS station at Danville VA (KDAN), I have noticed that the terrain
height produced from NCL is 341 meters, while the observed station
elevation is around 170 meters. This consequently has negative
consequences on the accuracy of the surface weather data I am getting
from my WRF files. I have checked the WRF files themselves using IDV to
ensure that the terrain is correct within the wrfout* files, and they
are. I have run NCL, using the same file, on both the Mac and PC and
have come up with the same result. I am using Lambert conformal map
projection when producing my WRF files, but I don't believe that is a
problem in this case. I have attached the NCL script file I am using to
produce my point output, as well as a sample of the NCL output itself,
below. What is causing this problem, and how can I fix it?

Thanks in advance for your time,

Elliot Tardif.

NCL script I'm using is given as follows:

begin

;DEFINE OPENING FILES

a = addfile("wrfout_sample.nc","r") ;defining the file to be added in
;-------------------------------
;DEFINE TIMES
res = True
res@returnInt = True
times = wrf_user_list_times(a)
ntimes = dimsizes(times)
;----------------------
;DEFINE LOCATION
loc = wrf_user_ll_to_ij(a, -79.336, 36.573,res)
locX = loc(0) - 1
locY = loc(1) - 1
print(locX)
print(locY)

;----------------------
;DEFINE ATMOSPHERIC VARIABLES
td2 = wrf_user_getvar(a,"td2",-1)
tc2 = wrf_user_getvar(a,"T2",-1) ; T2 in Kelvin
tc2 = tc2 - 273.15 ; T2 in C
u10 = wrf_user_getvar(a,"U10",-1) ; u at 10 m, mass point
u10 = u10 * 1.94386 ; convert to knots
v10 = wrf_user_getvar(a,"V10",-1) ; v at 10 m, mass point
v10 = v10 * 1.94386 ; convert to knots
psfc = wrf_user_getvar(a,"PSFC",-1)
psfc = psfc / 100
precip_exp = wrf_user_getvar(a,"RAINNC",-1)
precip_con = wrf_user_getvar(a,"RAINC", -1)
precip_tot = precip_exp + precip_con
hgt = wrf_user_getvar(a,"ter",-1)
;---------------------------
;DEFINE ARRAY AND PRINT THE DATA

dimtd2= dimsizes(td2)
gtd2 = dimtd2(0)
data = new(gtd2,"string")

n = -1
do nt=0, ntimes-1, 1
n = n + 1
data(n) = sprinti("%0.3i", (n))
data(n) = data(n) + sprintf("%7.3f", tc2(n,locX,locY))
data(n) = data(n) + sprintf("%7.3f", td2(n,locX,locY))
data(n) = data(n) + sprintf("%10.5f", u10(n,locX,locY))
data(n) = data(n) + sprintf("%10.5f", v10(n,locX,locY))
data(n) = data(n) + sprintf("%16.3f", psfc(n,locX,locY))
data(n) = data(n) + sprintf("%10.3f", precip_tot(n,locX,locY))
data(n) = data(n) + sprintf("%10.3f", hgt(n,locX,locY))
end do

asciiwrite("wrfout_KDAN.txt", data)
end

Here is a sample of the output from the above program:

000 2.711 2.726 -0.69524 -0.12440 981.227 0.000 341.977
001 3.665 2.643 -1.50709 0.84538 980.236 0.000 341.977
002 3.702 2.687 -1.55680 0.73297 979.061 0.000 341.977
003 3.741 2.780 -1.74506 0.74930 977.958 0.000 341.977
004 3.945 2.957 -1.36460 0.93940 977.420 0.000 341.977
005 4.216 3.225 -1.22440 0.97439 976.941 0.000 341.977
006 4.526 3.667 -1.05991 1.47378 976.562 0.015 341.977
007 5.458 4.591 -1.14547 2.24159 975.954 0.107 341.977
008 6.597 5.614 -1.50793 2.70870 975.397 0.573 341.977

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk