dpres_plevel
Calculates the pressure layer thicknesses of a constant pressure level coordinate system.
Prototype
function dpres_plevel ( plev [*] : numeric, psfc : numeric, ptop [1] : numeric, iopt : integer ) return_val : numeric
Arguments
plevA one dimensional array containing the constant pressure levels. May be in ascending or descending order. Must have the same units as psfc.
psfcA scalar or an array of up to three dimensions containing the surface pressure data in Pa or hPa (mb). The rightmost dimensions must be latitude and longitude. Must have the same units as plev.
ptopA scalar specifying the top of the column. ptop should be .le. min(plev). Must have the same units as plev.
ioptSet to zero. Currently not used.
Return value
If psfc is a scalar the return variable will be a one-dimensional array the same size as plev; if psfc is two-dimensional [e.g. (lat,lon)] or three-dimensional [e.g. (time,lat,lon)] then the return array will have an additional level dimension: (lev,lat,lon) or (time,lev,lat,lon). The size of the lev dimension is the same as the size of plev. The returned type will be double if psfc is double, float otherwise.
Description
Calculates the layer pressure thickness of a constant pressure level system. It is analogous to dpres_hybrid_ccm for hybrid coordinates. At each grid point the sum of the pressure thicknesses equates to [psfc-ptop]. At each grid point, the returned values above ptop and below psfc will be set to psfc@_FillValue. If there is no psfc@_FillValue then the _FillValue will be set to the default for float or double (in V5.2.x and earlier, 1e20 is used). If ptop or psfc is between plev levels then the layer thickness is modifed accordingly. If psfc is set to _FillValue, all layer thicknesses are set to the appropriate _FillValue.
The primary purpose of this function is to return layer thicknesses to be used to weight observations for integrations.
Use the dpres_plevel_Wrap function if metadata retention is desired. The interface is identical.
See Also
dpres_plevel_Wrap, dpres_hybrid_ccm, dz_height
Examples
Example 1 Consider a rawindsonde sounding and it is desired to compute the "internal energy" which is an integrated quantity.
; PRESSURE (hPa)
plev = (/ 1000.,950.,900.,850.,800.,750.,700.,650.,600., \
550.,500.,450.,400.,350.,300.,250.,200., \
175.,150.,125.,100., 80., 70., 60., 50., \
40., 30., 25., 20. /)
; TEMPERATURE (C)
t =(/29.3,28.1,23.5,20.9,18.4,15.9,13.1,10.1, 6.7, 3.1, \
-0.5,-4.5,-9.0,-14.8,-21.5,-29.7,-40.0,-52.4, \
-59.2,-66.5,-74.1,-78.5,-76.0,-71.6,-66.7,-61.3, \
-56.3,-51.7,-50.7,-47.5 /)
plev = plev*100.
plev@units = "Pa" ; Pa = kg/(m s2) (Pascal)
t = t+273.15
t@units = "K"
ptop = min(plev)
psfc = 1018*100.
psfc@units = "Pa"
dp = dpres_plevel(plev, psfc, ptop, 0) ; dp(30)
; Use dpres_plevel_Wrap if metadata retention is desired
; dp = dpres_plevel_Wrap(plev, psfc, ptop, 0) ; dp(30)
print("dp="+dp+" t="+t)
CP = 1004. ; J/(K kg) [ m2/(K s2) ]
GR = 9.81 ; gravity [ m/s2 ]
; integrate vertically
IE = (CP/GR)*sum( t*dp )
IE@long_name = "Internal Energy: Cp*sum( T*dp )/gravity"
IE@units = "kg/s2"
; weighted vertical average
IE_average = IE/sum(dp)
IE_average@long_name = "Internal Energy: Weighted vertical average"
The output from the print statement is:
psfc=101800
(0) plev=100000 dp=4300 t=302.45
(1) plev=95000 dp=5000 t=301.25
(2) plev=90000 dp=5000 t=296.65
(3) plev=85000 dp=5000 t=294.05
(4) plev=80000 dp=5000 t=291.55
(5) plev=75000 dp=5000 t=289.05
(6) plev=70000 dp=5000 t=286.25
(7) plev=65000 dp=5000 t=283.25
(8) plev=60000 dp=5000 t=279.85
(9) plev=55000 dp=5000 t=276.25
(10) plev=50000 dp=5000 t=272.65
(11) plev=45000 dp=5000 t=268.65
(12) plev=40000 dp=5000 t=264.15
(13) plev=35000 dp=5000 t=258.35
(14) plev=30000 dp=5000 t=251.65
(15) plev=25000 dp=5000 t=243.45
(16) plev=20000 dp=3750 t=233.15
(17) plev=17500 dp=2500 t=220.75
(18) plev=15000 dp=2500 t=213.95
(19) plev=12500 dp=2500 t=206.65
(20) plev=10000 dp=2250 t=199.05
(21) plev=8000 dp=1500 t=194.65
(22) plev=7000 dp=1000 t=197.15
(23) plev=6000 dp=1000 t=201.55
(24) plev=5000 dp=1000 t=206.45
(25) plev=4000 dp=1000 t=211.85
(26) plev=3000 dp=750 t=216.85
(27) plev=2500 dp=500 t=221.45
(28) plev=2000 dp=750 t=222.45
(29) plev=1000 dp=500 t=225.65
Variable: IE
Type: float
Total Size: 4 bytes
1 values
Number of Dimensions: 1
Dimensions and sizes: [1]
Coordinates:
Number Of Attributes: 2
units : W/m2
long_name : Internal Energy: Cp*sum( T*dp )/gravity
(0) 2.72987e+09
Example 2 Consider a similar but more common usage. Read data from an ERA-40 (or NCEP) data set. Compute Latent heat energy.
f = addfile ("foo.ERA40.nc" , "r")
lev = f->lev ; (/ 1, 2, 3, 5, 7, 10, 20, 30, \
; 50, 70,100,150, 200,250,300,400, \
; 500,600,700,775, 850,925,1000 /)
Q = f->Q ; kg/kg (time,lev,lat,lon)
psfc= f->PS ; PA (time,lat,lon)
lev = lev*100
lev@units = "Pa" ; to match PS
ptop= 0 ; integrate 0==>psfc at each grid point
; dp(ntim,klev,nlat,mlon)
dp = dpres_plevel(lev, psfc, ptop, 0)
; Use dpres_plevel_Wrap if metadata retention is desired
; dp = dpres_plevel_Wrap(lev, psfc, ptop, 0)
; latent heat of vaporization at 0C
L = 2.5e6 ; J/kg [ m2/s2 ]
g = 9.81 ; m/s
Qdp = Q*dp ; temporary variable
copy_VarCoords(Q, Qdp)
; integrate vertically [level dimension = 1]
LE = dim_sum_n_Wrap( Qdp,1 ) ; LE(ntim,nlat,mlon)
LE = (L/GR)*LE ; (time,lat,lon)
LE@long_name = "Latent Energy: SUM[L*Q*dp]/g"
LE@units = "kg/s2"
printVarSummary( LE )
delete( Qdp ) ; no longer needed