GRASS logo

NAME

r.pi.fnn - Determines patches of given value and performs a nearest-neighbor analysis.

KEYWORDS

raster

SYNOPSIS

r.pi.fnn
r.pi.fnn help
r.pi.fnn [-aq] input=name costmap=string output=name keyval=integer method=string[,string,...] number=num[-num][,num[-num],...] statmethod=string [dmout=string] [adj_matrix=string] [title="phrase"] [--overwrite] [--verbose] [--quiet]

Flags:

-a
Set for 8 cell-neighbors. 4 cell-neighbors are default
-q
Run quietly
--overwrite
Allow output files to overwrite existing files
--verbose
Verbose module output
--quiet
Quiet module output

Parameters:

input=name
Name of input raster map
costmap=string
Name of existing raster file with path-cost information
output=name
Name for output raster map
keyval=integer
Key value
method=string[,string,...]
Operation to perform on fragments
Options: distance,path_distance,area,perimeter,shapeindex
number=num[-num][,num[-num],...]
Number of nearest neighbors to analyse
statmethod=string
Statistical method to perform on the values
Options: average,variance,standard deviation,value,sum
dmout=string
Output name for distance matrix and id-map (performed if not empty)
adj_matrix=string
Output name for adjacency matrix (performed if not empty)
title="phrase"
Title for resultant raster map

DESCRIPTION

Determine the functional nearest-neighbor distance analysis. r.pi.fnn is a patch based ecological/functional nearest neighbour analysis module. It computes distance based on a friction map. This module is related to r.pi.enn but more adequat if the ecological connectivity should be analysed.

NOTES

The calculation of the ecolgogical nearest neighbour is based on a raster with start patches. The actual map can be categorical or continuous but the value defined in keyval will be the basis for the patches to calculate the methods defined below. These patches will also be in the output map. The calculation of the ecolgogical nearest neighbour is based on a costmap (* and 1-infinite) - this map can be binary or continous - high values are considered to have high cost issues and the shortest path is the one with the lowest amount of costs. "null" values can not be traversed, hence these values have to be bypassed. "0" values are not accepted and will result in "0" distance.

e.g. if a binary map(1 and 2) is used, the the path with the lowest amount of "1" is chosen The number is the amount of nearest neighbours to be taken and the calculated distances are processed as assigned in statmethod Operations which r.pi.fnn can perform are:

Distance
The Distance to Nearest computes the nearest edge-to-edge distance between patches. Counting from the focus patch.
path Distance
The Distance to Nearest computes the nearest edge-to-edge distance between patches. Unlike Distance the distance is computed based on subsequent NN not from the focus patch onwards. The 1th NN is the first patch with the minimal edge-to-edge distance from the focus patch, while 2th NN is the patch with the minimal edge-to-edge distance from the 1th NN patch and so on.
Area
The Area computes the size of the nearest edge-to-edge distance patch. It is based on Distance not on path Distance.
Perimeter
The Perimeter computes the Perimeter of the nearest edge-to-edge distance patch. It is based on Distance not on path Distance.
SHAPE
The SHAPE computes the SHAPE Index of the nearest edge-to-edge distance patch. It is based on Distance not on path Distance. The statsmethod operators determine calculation is done on the distance. Average, Variance,Stddev and value can be used.
Average
The Average computes the average value defined in Operations to perform .
Variance
The Variance computes the variance defined in Operations to perform .
Stand. Dev.
The Stand. Dev. computes the stddev value defined in Operations to perform .
Value
The patch Distance computes the nearest edge-to-edge distance between two patches. The output of value is the actual value. E.g. NN==5 of area gives the size of the 5th NN while Average gives the average of the area of 1-5th NN. The input options are either one NN: 1 or several NN separated by ,: 1,2,5,8 or a range of NN: 1-6.

Merging these options is possible as well: 1-5,8,9,13,15-19,22 etc.

EXAMPLE

An example for the North Carolina sample dataset: Computing the functional or ecological distance to the first to nth nearest neighrbours using a cost matrix: 
r.mapcalc "cost_raster = if(landclass96==5,1,if(landclass96 == 1, 10, if (landclass96==3,2, if(landclass96==4,1,if(landclass96==6,100)))))"
r.pi.fnn input=landclass96 keyval=5 costmap=cost_raster output=fnn1 method=distance number=10 statmethod=average

SEE ALSO

r.pi.enn, r.pi.index

AUTHORS

Programming: Elshad Shirinov
Scientific concept: Dr. Martin Wegmann
Department of Remote Sensing
Remote Sensing and Biodiversity Unit
University of Wuerzburg, Germany

Last changed: $Date$

Main index - raster index - Full index

© 2003-2016 GRASS Development Team