Hawth's Analysis Tools for ArcGIS

 You are here: Home > Hawths Tools > Tools Descriptions & Help

 Summary

CONDITIONAL POINT SAMPLING TOOL
Input: a point layer
Output: a point layer
Features:

• generates sampling points specifically for "case controlled" (conditional / descrete choice) analyses e.g. case controlled logitistic regression approaches to habitat selection using telemetry data
• ties the original point to the sample points using a unique identifier field in the input attribute table
• offers a variety of options for generating points (random, regular, sampled from movement parameter distributions)
• the number of points to generate can be constant across all cases, can vary across cases (value obtained from attribute field), or can be specified as a constant density (area based calculation)
• the "domain of availability" (the circular area in which points are generated) can be of constant or variable radius (value obtained from attribute field)
• the output is a shapefile that is designed to then be used with other tools (the Intersect Point Tool for instance)
• records in a text file the options and parameters used to create the sampling point shapefile (in case you forget how a layer was generated)
• numbers the generated points consecutively within each case in a "PNTID" field
• fast: generates around 5000 - 10000 points per second depending on the speed of your computer

Limitations:

• assumes that the data is in a projected coordinate system (it is inappropriate to use this tool with point data in a Geographic Coordinate System)
• can generate output files with a very large number of features, which can be cumbersome to work with
• the form is tall, and requires that your screen resolution is greater than 800x600

[Click for larger view]

 Example

Figures A - F. All figures show conditional sampling points generated with this tool using four original locations (shown as larger red dots in all figures). (A) a constant number of points generated in a fixed radius using the random sampling form; (B) regularly spaced points generated in a fixed radius; (C) a constant number of points generated using variable radii in an attribute field (note that density varies widely between cases in this scenario); (D) points generated using the same variable radii as in C, but using a constant density (note that the number of points varies between cases in this scenario); (E) a constant number of points generated using movement parameter (step length and turn angle) distributions (note that for clarity only one case is shown in this figure, that not all points are shown as the spread is quite large, and that in this rather complex example no step lengths were permitted from 0-50m which creates the "donut" shape observed); (F) the same as E, but with all four cases shown.

Summary report. The following is an example of a summary report that is created in order to help you keep track of the options and parameters used to generate each set of sampling points:
Conditional Sampling Point Generation Tool
Layer Information
Layer: condsamp_elk1
Unique ID field: STEPID
Sampling method: constant number of sampling points per case (150)
Domain of availability: attribute field based (RADIUS)
Sampling form: random
Set of input data: all points used
Processing...
Number of input points processed: 4
Number of output points generated: 600
Processing completed successfully

 Help

Getting started. At least one point layer must be loaded into ArcMap in order to use this tool.

Sampling Intensity. These options determine how many sample points are generated per case. The simplest option is to create a constant number of points across all cases. The trouble with this method is that it results in different sampling densities IF your domain of availability varies. Thus the second option is to define the density of sampling, and the number of points to generate is calculated on a case by case basis based on the sampling area (the domain of availability). The third option is to define the number of sampling points in an attribute field, which allows you to create your own sampling scheme rules (in cases where the two options I provide are not adequate). Note that some Sampling Form options will override the Sampling Intensity options (see the Sampling Form section for details).

Domain of Availability. This is the area in which sampling is allowed to occur and takes the form of a circle centered around each case point. The radius can be fixed or variable (in the latter case, the radius is read from an attribute field). Using a field representing the step length is an example of a variable radius. Note that some Sampling Form options will override the Domain of Availability options (see the Sampling Form section for details).

Sampling Form. The random sampling form generates points entirely at random within the domain of availability. This option does not affect any of the Sampling Intensity or Domain of Availability options.

The regular sampling form generates evenly spaced points at a specified interval. This option overrides all of the Sampling Intensity options (i.e. given a radius that defines the domain of availability, the number of points to generate for each case is determined by the regular interval between points).

Finally, the sampling can be based on movement parameter distributions (binned step length and turn angle distributions, and a "Bearing" field that represents the bearing of the movement from the previous point to the current point). This option overrides all of the Domain of Availability options, and the "Constant Density" option for Sampling Intensity. The Domain of Availability is determined by the upper limit to the step length distribution.

Linking data to the generated points. If you need to acquire data from the attribute table of the original point layer (i.e. representing the "cases") and attach it to the attribute table of the generated points, you can join the two tables together using the common unique ID field you specified. To perform the link, right-click on the generated point layer in the ArcMap table of contents, select "Joins and Relates...", then "Joins". In the new form that appears, you are asked to specify the common field name in each of the tables. Press OK. If you are asked whether you would like to index a field, press Yes - it only takes a few seconds. Now if you open the attribute table of the generated points you will see all the fields from the original attribute table have been added. Note that this is a temporary join. To make the join permanent, you need to create a new shapefile (right click on the layer in the table of contents, select "Data", then "Export Data").