Geospatial Modelling Environment

genregionsampleplots (Generate Sample Plots)

A feature-rich tool for generating sampling plots (and zones) within a specified area


This tool generates a sampling grid: a series of regularly sized and spaced square or rectangular polygons. Unlike the simpler ‘genvecgrid’ command, however, this command provides options for generating both plots and zones (plots are the smallest sampling unit, and zones are groups of plots), for characterizing the plots and zones based on a suitability raster, and for taking into account barriers in the landscape. For instance, this tool can be used to generate sampling plots that do not cross rivers.

The extent of the sampling grid can be specified by referencing a feature data source, or by directly specifying the minimum and maximum x and y coordinates. An ideal choice of reference layer would be a polygon dataset containing a study area boundary. Sampling plots are only generated in areas covered by the features in this layer. Alternatively, if you specify the extent of a rectangle then plots are generate throughout that area.

The primary output of this tool is a plot polygon layer. You may optionally also generate zones (see the ‘outzones’ and ‘dimzones’ command options). Zones would be useful if you wish to perform stratified random sampling in order to cluster sample plots in zones. For instance, if the extent of the area you are sampling is very large then pure random sampling of plots may create a distribution of sample plots that is impractical to visit in reality. A more detailed discussion of sampling designs and the issues that motivate different designs is included in the sampling chapter.

The dimensions of the plots (and zones) are specified in coordinate system units (e.g. meters for the UTM projection). It is recommended that you do not use this tool with unprojected data (geographic coordinate systems). The dimension of the zones should be exactly divisible by the dimension of the plots. For instance, plots of 100m and zones of 1km is suitable, but plots of 30m and zones of 1km is not.

The snap option forces the spatial position of the grid to be aligned with the underlying coordinate system. For instance, specifying snap=1000 would force the vector grid lines to be aligned with the exact 1km positions in a UTM grid. For most sampling purposes this is not an important consideration.

The ‘raster’ option refers to a suitability raster that classifies the landscape into areas that can be sampled (1) and those that cannot (0). For instance, if you are sampling forest habitat in a landscape that is a mixture of forest and open areas, then the raster would depict all forested areas as 1, and all open areas as 0. This raster is used to calculate the proportion and total area of suitable habitat in each plot (and zone), stored in two fields in the attribute table.

This tool adjusts the plots and zones to reflect barriers in the landscapes, which can be depicted as polygons and lines. For the polygons, the tool will remove all plots that are completely contained by barrier polygons, and will clip plots that overlap the barrier polygons. Barrier lines are used to split plots into components that do not cross the barrier lines. You can specify barriers as polygons only, lines only, or both lines and polygons.

After the geometry of the plots has been modified by these barriers, an area threshold (expressed as a proportion of the plot area) is used to identify plot fragments that will be merged with adjacent plots. Any fragments greater than this threshold are considered to be close enough to a full plot size that they can be retained without merging in the plot dataset. Each of the smaller fragments is merged with the neighbouring plot with which it shares the longest boundary, provided this does not involve crossing a barrier.

The final barrier-related parameter is a tolerance that defines the distance from the edge of the plot that barrier polygons or lines can be ignored. This parameter is needed because, with certain data formats, there can be very small movements of vertices following the splitting procedure (as a result of coordinate precision). Usually this is only a few centimetres and is negligible, but does create a problem when resolving merges of adjacent plot polygons. The default tolerance is 1, and the units for this tolerance are the coordinate system units (e.g. meters for UTM). If you use a barrier line dataset that contains overshoots, then you may need to increase this tolerance to allow the merging algorithm to resolve acceptable merges. It is particularly important when using a barrier line dataset that 1) lines that are supposed to cross do actually cross (e.g. roads at a T junction), and 2) that overshoots at such crossings are not too long. The former issue results in plots that fail to be split by a barrier line, while the latter issue can result in a failure of the algorithm to deal with plot fragments.


genregionsampleplots(extent, out, dim, [outzones], [dimzones], [snap], [raster], [barrierpoly], [barrierline], [minarea], [tol]);

extentthe reference layer that defines the extent of the vector grid, or a set of four values that define the extent (min x, max x, min y, max y)
outthe output polygon data source for the plots
dimthe dimensions of the plots in coordinate system units, e.g. 100 or c(100,200); specifying one value results in square plots
[outzones]the output polygon data source for the zones (if unspecified no zone output is produced; if specified, dimzones must also be specified)
[dimzones]the dimensions of the zones in coordinate system units (this value should be an exact multiple of the plot dimensions), e.g. 1000 or c(1000,2000); specifying one value results in square zones
[snap]a value > 0; controls whether the vector grid is aligned with a major coordinate system interval (supplying a value of 1000 will result in the grid being aligned to the 1000-mark intervals of the coordinate system); default=0 (no snap)
[raster]a suitability (1/0) raster that is used to characterize plots (see help documentation for further details)
[barrierpoly]a polygon data source representing barriers that cannot be sampled (see help documentation for further details)
[barrierline]a line data source representing barriers that cannot be crossed (see help documentation for further details)
[minarea]the threshold, expressed as a proportion of the plot area, that is used to determine whether a plot fragment will be retained as a fragment, or merged (default=0.8, see help documentation for further details)
[tol]the tolerance that is used in the merging of plot fragments resulting from barriers (see help documentation for further details)


genregionsampleplots(extent=”C:dataregions.shp”, out=”C:dataplots.shp”, dim=200, outzones=”C:datazones.shp”, dimzones=1000, raster=”C:datasuit.tif”);

genregionsampleplots(extent=”C:dataregions.shp”, out=”C:dataplots.shp”, dim=200, outzones=”C:datazones.shp”, dimzones=1000);

genregionsampleplots(extent=”C:dataregions.shp”, out=”C:dataplots.shp”, dim=200, raster=”C:datasuit.tif”, barrierpoly=”C:datalakes.shp”, barrierline=”C:datarivers.shp”, minarea=0.7, tol=5);


Please consider making a purchase to support the continued development of these tools  Read more...

Tips on how to use this interface efficiently 


Open Source GIS

Copyright © 2001-2014 Hawthorne L. Beyer, Ph.D., Spatial Ecology LLC    Connect on LinkedIn