samplam - Samples and laminates variables in multiple datasets.

SYNOPSIS

samplam  [ parameter=value ... ]  [ inputfile ... outputfile ]

Parameters are: point_file, points_are, include_vars, output_coords, laminate_attrs, new_dim_name.

DESCRIPTION

samplam takes one or more input datasets, samples them using the list of points in a point_file dataset, and then laminates the variables along a newly specified dimension (see sample and laminate). The results are written to a single output file. The contents of the point_file and the method of sampling is based on the specification of the parameter points_are as follows:

1) points - requires a one-dimensional dataset of (x, y) or (lat, lon) pairs. The list can be any dimension and the variables must be named (x, y) or (latitude, longitude), respectively. The input data sets are sampled at the locations nearest to the points specified in the list.

2) vectors - requires a one-dimensional dataset of (x, y, delta-x, delta-y) or (lat, lon, delta-lat, delta-lon) values. The list can be any dimension and the variables must be named (x, y, delta-x, delta-y) or (latitude, longitude, delta-lat, delta-lon), respectively. The input data sets are sampled at the locations nearest the points along the specified vectors.

3) tracks - requires a one-dimensional dataset of (x, y) or (lat, lon) pairs. The list can be any dimension and the variables must be named (x, y) or (latitude, longitude), respectively. The list is interpreted to represent one or more sequences of points (one or more tracks). Points in the list made up of bad_values (see datasets) delimit tracks, i.e. the previous track is ended and the next point starts a new track. The input data sets are sampled at the locations nearest to the points along the specified tracks.

4) outlines - requires a one-dimensional dataset of (x, y) or (lat, lon) pairs. The list can be any dimension and the variables must be named (x, y) or (latitude, longitude), respectively. The list is interpreted to represent one or more connected regions (or polygons). Points in the list made up of bad_values (see datasets) delimit polygons, i.e. the previous region being outlined is closed and the next point starts a new polygon. The input data sets are sampled at all locations insided the outlined region.

5) boxes - requires a one-dimensional dataset of (x, y, delta-x, delta-y) or (lat, lon, delta-lat, delta-lon) values. The list can be any dimension and the variables must be named (x, y, delta-x, delta-y) or (latitude, longitude, delta-lat, delta-lon), respectively. Each point in the list is interpreted to represent two opposite corners of a rectangle. The input data sets are sampled at all locations inside the rectangles.

If a dataset has dimensions x, y and z and the sampling is done in the x and y dimensions, the output dataset retains the z dimension of the data.

All input datasets are assumed to have line and sample or latitude and longitude dimensions. They are also assumed to have the same variables names, the data type, bad value, units, and scaling as the first input dataset. An error is returned if the variable names, size of the dimensions or data type do not all match. If latitude and longitude format is used for specifying points, each dataset must have the same earth transform.

PARAMETERS

point_file

This is the name of the TeraScan dataset containing the one-dimensional set of points (see descriptions above).

There is no default.

points_are

Specifies the sampling method (see descriptions above).

Valid responses are [ points, vectors, tracks, outlines, boxes ]. The default is points.

include_vars

Specifies which variables in the input dataset(s) to sample and laminate and write to the output dataset. If this list is prefixed with a minus sign, all variables, except those listed, will be sampled. Wildcards * and ? are allowed.

The default is all variables in the input datasets.

output_coords

If output_coords=yes, samplam writes the x, y or lat,lon coordinates where the samples were taken to the output file. In the case of points_are=points, the information is the same as in the point_file, although for the other cases, the sample locations are generated internally and thus it may be necessary to keep the generated list of sample points.

The default is no.

laminate_attrs

Optionally specifies a list of attributes to laminate and write to the output dataset.

The default implies none.

new_dim_name

Specifies the name of the new dimension along which the data is to be laminated.

There is no default.

EXAMPLES

lapaz% samplam
in/out files   : char(255) ? n12.93201.0221 n12.sample
point_file     : char(255) ? n12.post
points_are     : char(  8) ? [points]
include_vars   : char(255) ? [] avhrr_ch3
output_coords  : char(  3) ? [no] yes
laminate_attrs : char(255) ? []
new_dim_name   : char( 31) ? newdim

SEE ALSO

datasets, xvu, sample, laminate, dimavg, formats.


Last Update: $Date: 1998/05/29 20:42:28 $