1r.li(1) Grass User's Manual r.li(1)
2
6 r.li - package overview
7
9 raster, landscape structure analysis, overview, landscape metrics,
10 landscape pattern, landscape analysis
11
13 The r.li suite is a toolset for multiscale analysis of landscape struc‐
14 ture. It aims at replacing the r.le suite of modules through a client-
15 server, multiprocess implementation. External software for quantitative
16 measures of landscape structure is for example FRAGSTATS (McGarigal and
17 Marks 1995).
18 The r.li suite offers a set of patch and diversity indices. It sup‐
20 ports analysis of landscapes composed of a mosaic of patches, but, more
21 generally, the modules work with any two-dimensional raster map whose
22 cell values are integer (e.g., 1, 2) or floating point (e.g., 1.1, 3.2)
23 values. The r.li.setup module has options for controlling the shape,
24 size, number, and distribution of sampling areas used to collect infor‐
25 mation about the landscape structure. Sampling area shapes can be the
26 entire map or a moving window of square, rectangular or with circular
27 shape. The size of sampling areas can be changed, so that the landscape
28 can be analyzed at a variety of spatial scales simultaneously. Sampling
29 areas may be distributed across the landscape in a random, systematic,
30 or stratified-random manner, or as a moving window.
31 The r.li modules can calculate a number of measures that produce single
33 values as output (e.g. mean patch size in the sampling area), as well
34 as measures that produce a distribution of values as output (e.g. fre‐
35 quency distribution of patch sizes in the sampling area). The results
36 are stored as raster maps.
37 The general procedure to calculate an index from a raster map is two-
39 fold:
40 1 run r.li.setup: create a configuration file selecting the parts
42 of raster to analyze.
43 2 run r.li.'index' (e.g., r.li.patchdensity) for calculate the
45 selected index using on the areas selected on configuration
46 file.
47
49 Also the r.li.daemon has a main function and it can be run, but it is
50 only a template for development of new indices. The function itself
51 has no meaning, it can be used only for debug.
52
54 To calculate a patch density index on a whole 'geology' raster map in
55 the Spearfish region, using a 5x5 moving window, follow this procedure:
56 1
58 CREATE A NEW CONFIGURATION FILE
59 1.1
61 run
62 r.li.setup
63 1.2
66 The main r.li.setup window is displayed, click on "New"
67 1.3
69 Now it is displayed the new configuration window, enter the
70 configuration file name (e.g., "my_conf", do not use absolute
71 paths) and the name of raster map (e.g., "geology"). The other
72 fields are not needed for this configuration.
73 1.4
75 Click on "Setup sampling frame", select "Whole maplayer" and
76 click "OK"
77 1.5
79 Click on "Setup sampling areas", select "Moving window" and
80 click "OK"
81 1.6
83 Click on "Use keyboard to enter moving window dimension"
84 1.7
86 Select "Rectangle" and enter 5 on "heigth" and "width" fields
87 1.8
89 Click on "Save settings"
90 1.9
92 Close r.li.setup window
93 2
95 CALCULATE PATCHDENSITY INDEX
96 2.1 set region settings to geology raster map:
98 g.region rast=geology -p
99 2.2
102 run r.li.patchdensity:
103 r.li.patchdensity map=geology conf=my_conf out=patchdens
104 The resulting patch density is stored in "patchdens" raster map. You
107 can verify the result for example with contour lines:
108 r.contour in=patchdens out=patchdens step=5
109 d.rast patchdens
110 d.vect -c patchdens
111 Note that if you want to run another index with the same area configu‐
112 ration, you don't have to create another configuration file. You can
113 also use the same area configuration file on another map. The program
114 rescale it automatically. For instance if you have selected a 5x5 sam‐
115 ple area on 100x100 raster map, and you use the same configuration file
116 on a 200x200 raster map, then the sample area is 10x10.
117
119 Core modules:
120 r.li.daemon: job launch daemon
122 r.li.setup: Configuration editor for r.li.'index'
124 Patch indices:
125 Indices based on patch number:
127 r.li.patchdensity: Calculates patch density index
129 on a raster map, using a 4 neighbour algorithm
130 r.li.patchnum: Calculates patch number index on a
132 raster map, using a 4 neighbour algorithm
133 Indices based on patch dimension:
134 r.li.mps: Calculates mean patch size index on a
136 raster map, using a 4 neighbour algorithm
137 r.li.padcv: Calculates coefficient of variation
139 of patch area on a raster map
140 r.li.padrange: Calculates range of patch area
142 size on a raster map
143 r.li.padsd: Calculates standard deviation of
145 patch area a raster map
146 Indices based on patch shape:
147 r.li.shape: Calculates shape index on a raster
149 map
150 Indices based on patch edge:
151 r.li.edgedensity: Calculates edge density index
153 on a raster map, using a 4 neighbour algorithm
154 Indices based on patch attributes:
155 r.li.cwed: Calculates contrast Weighted Edge Den‐
157 sity index on a raster map
158 r.li.mpa: Calculates mean pixel attribute index
160 on a raster map
161 Diversity indices:
162 r.li.dominance: Calculates dominance's diversity index
164 on a raster map
165 r.li.richness: Calculates dominance's diversity index on
167 a raster map
168 r.li.shannon: Calculates Shannon's diversity index on a
170 raster map
171 r.li.simpson: Calculates Simpson's diversity index on a
173 raster map
174
176 New indices can be defined and implemented by any C programmer, without
177 having to deal with all basic functions (IO etc.). The computing archi‐
178 tecture and the functions are clearly separated, thus allowing an easy
179 expandability. Every index is defined separately, placed in a directory
180 along with its Makefile for compiling it and a file description.html
181 which describes the index including a simple example of use.
182
184 McGarigal, K., and B. J. Marks. 1995. FRAGSTATS: spatial pattern analy‐
185 sis program for quantifying landscape structure. USDA For. Serv. Gen.
186 Tech. Rep. PNW-351 (PDF).
187
189 Claudio Porta and Lucio Davide Spano, students of Computer Science Uni‐
190 versity of Pisa (Italy).
191 Commission from Faunalia Pontedera (PI)
192 Last changed: $Date: 2007-02-26 15:54:04 +0100 (Mon, 26 Feb 2007) $
194 Full index
196GRASS 6.3.0 r.li(1)