pslhyp, pslec, pslhor

Functions for computing sea level pressures.

Synopsis

    function pslhyp(
        pres : float,
        z    : float,
        tv   : float
    )

    function pslec(
        t    : float,
        phis : float,
        ps   : float,
        pres : float
    )

    function pslhor(
        z    : float,
        t    : float,
        phis : float,
        ps   : float,
        pres : float,
        lats : float
    )

Arguments

phis

surface geopotential (m^2/s^2). A 2-dimensional array dimensioned nlat x mlon where nlat is the number of latitudes and mlon is the number of longitudes. Note: if the utility "ccm2nc" was used to convert the original CCM History Tape file to netCDF, this field may have a time dimension of length one. In this case use "phis(0,:,:)" as the argument to the functions.

pres

pressure (Pascals). For pslhyp and pslec, an array (corresponding to the lowest vertical level) with at least 2 dimensions. The last 2 dimensions being nlat x mlon where nlat is the number of latitudes and mlon is the number of longitudes. It must be at least 3 dimensions for pslhor because it requires multilevel fields which start at the bottom of the atmosphere (ie, the lowest vertical level). The last two dimensions must be nlat x mlon. For pslhyp, missing values should be indicated by pres@_FillValue. If pres@_FillValue is not set, then the NCL default (-999) will be assumed.

ps

surface pressure (Pascals) (last two dimensions must be the same dimensions as phis). It must be the same dimensions as pres except for in pslhor it won't have a multilevel field.

t

temperature (K) (same dimensions as pres)

tv

virtual temperature (K) (same dimensions as pres)

z

geopotential height (same dimensions as pres)

lats

latitudes (degrees). A vector of length nlat.

Description

pslec computes the sea level pressure using the ECMWF formulation and returns the results as a float array with the same dimensions as its input arrays.

pslhor computes the sea level pressure using the ECMWF formulation and Trenberth's horizontal correction and returns the results as a float array with the same dimensions as its input arrays ps and phis.

All three of these routines return the attributes long_name, short_name, and units.

Note: For the arrays whose last two dimensions are nlat x nlon, the rest of the dimensions (if any) are collectively referred to as nt. If the input/output arrays are just two dimensions, then nt can either be considered equal to 1 or nothing at all.

Examples

Example 1

t(time, lev, lat, lon), etc. for 4D variables
phis(time,lat,lon), etc. for 3D variables

PSL = pslhyp (pres(:,0,:,:), z(:,0,:,:), tv(:,0,:,:)

PSL = pslec (t(:,0,:,:), phis, ps, pres(:,0,:,:) )

PSL = pslhor (z,t,phis,ps,pres,lat)

Example 2

begin
  nlat  =  64              ; T42 gaussian grid
  mlon  = 128
 
  fname  = "pslecT42.hsl"              ; sequential fortran file
  nrec   = fbinnumrec (fname)          ; number of records

  t      = fbinrecread(fname, 0 ,(/nlat,mlon/), "float" )
  phis   = fbinrecread(fname, 1 ,(/nlat,mlon/), "float" )
  ps     = fbinrecread(fname, 2 ,(/nlat,mlon/), "float" )
  pres   = fbinrecread(fname, 3 ,(/nlat,mlon/), "float" )  
  psl    = pslec (t,phis,ps,pres)
  print(psl)
  print(psl@long_name)
  print(psl@short_name)
  print(psl@units)
end

Reference Manual Control Panel

NG4.1 Home, Index, Examples, Glossary, Feedback, Ref Contents, Ref WhereAmI?