Data Type Structures

Structures can be defined in five data types (byte, short, integer, float, double) and in up to 4 dimensions. The FPL structure naming follows the syntax below:

Naming Convention

nc[2d,3d,4d]_[byte,short,int,float,double]_ll[f,d]_t[i,f,d]_l[i,f]

Component	Options	Description
nc[2d,3d,4d]	2d, 3d, 4d	Dimension nomination
_[type]	byte, short, int, float, double	Variable data type
_ll[f,d]	f (float), d (double)	Coordinate (lon/lat) type
_t[i,f,d]	i (integer), f (float), d (double)	Time type (3D and 4D only)
_l[i,f]	i (integer), f (float)	Level type (4D only)

Examples

type(nc2d_double_llf) :: grid2d          ! 2D double, float coordinates
type(nc3d_int_llf_ti) :: grid3d          ! 3D integer, float coords, integer time
type(nc4d_short_lld_ti_li) :: grid4d     ! 4D short, double coords, int time, int level

Structure Fields

All types share common fields. Fields marked with [U] are set by the user; fields marked with [R] are populated when reading data.

Field	Type	Description
varname	character	Variable name [U]
lonname	character	Longitude dimension name [U]
latname	character	Latitude dimension name [U]
timename	character	Time dimension name (3D/4D) [U]
levelname	character	Level dimension name (4D) [U]
long_name	character	Long name / title [U][R]
varunits	character	Variable units [U][R]
lonunits	character	Longitude units [U][R]
latunits	character	Latitude units [U][R]
timeunits	character	Time units (3D/4D) [U][R]
levelunits	character	Level units (4D) [U][R]
dimname	character(:)	Dimension names array [R]
dimunits	character(:)	Dimension units array [R]
nlons	integer	Number of longitudes [U][R]
nlats	integer	Number of latitudes [U][R]
ntimes	integer	Number of time steps (3D/4D) [U][R]
nlevels	integer	Number of levels (4D) [U][R]
ndims	integer	Number of dimensions [R]
vartype	integer	NetCDF type number [R]
FillValue	varies	Fill / missing value [U][R]
longitudes	float/double(:)	Longitude array [R]
latitudes	float/double(:)	Latitude array [R]
times	int/float/double(:)	Time array (3D/4D) [R]
levels	int/float(:)	Level array (4D) [R]
ncdata	varies	Data array [R]
dimid	integer(:)	Dimension IDs [R]
dimsize	integer(:)	Dimension sizes [R]
varids	integer(:)	Variable IDs [R]
dims	integer(2/3/4)	Variable dimension IDs [R]

Code Example — Set Dataset Fields

program main
  use fpl
  implicit none

  type(nc4d_float_llf_ti_li) :: grid4d

  ! Set metadata
  grid4d%long_name  = "My Grid  ~ 1 degree"
  grid4d%varname    = "grid"
  grid4d%lonname    = "lon"
  grid4d%latname    = "lat"
  grid4d%timename   = "time"
  grid4d%levelname  = "level"

  grid4d%varunits   = "dimensionless"
  grid4d%lonunits   = "degrees_east"
  grid4d%latunits   = "degrees_north"
  grid4d%timeunits  = "hour"
  grid4d%levelunits = "m"

  grid4d%ntimes     = 10
  grid4d%nlevels    = 5
  grid4d%FillValue  = -9999

  ! Print metadata
  write(*,*) "Grid 4d info ================================="
  write(*,'(a13,a20)')   "varname:    ", grid4d%varname
  write(*,'(a13,a20)')   "timename:   ", grid4d%timename
  write(*,'(a13,a20)')   "levelname:  ", grid4d%levelname
  write(*,'(a13,i4)')    "ntimes:     ", grid4d%ntimes
  write(*,'(a13,i4)')    "nlevels:    ", grid4d%nlevels
  write(*,'(a13,f10.4)') "FillValue:  ", grid4d%FillValue
end program main

Compile & Run

gfortran -o setfields.out setfields.f90 -I/usr/lib64/gfortran/modules/ -lFPL
./setfields.out