Home range creation and analysis using

Geospatial Modeling Environment and

ArcGIS software
Alexandra Locher1 and Matt Lindenberg2
Grand Valley State University, Biology Department, Allendale, Michigan
1
lochera@gvsu.edu; 2lindenbm@gvsu.edu

Description: Many analyses in natural resources or ecology-related fields, specifically wildlife-focused

disciplines, investigate wildlife habitat use and movement patterns for population and habitat
management. Kernel density estimates and home range analysis are commonly used by wildlife
biologists for such investigations. Methods to conduct home range analyses are often complex and
require use of multiple software programs. In the literature and on the web, it is difficult to find
comprehensive instructions on how to create home ranges and proceed with analyses. The purpose of
this manual is to describe the process of synthesizing raw location point location data, creating home
ranges (kernel density estimates), and begin initial analyses involving habitat selection.

Key Words: geospatial modeling environment, GIS, home range, isopleth, kernel density, R, wildlife

0
 Home range creation and analysis using Geospatial Modeling Environment
and ArcGIS software

Alexandra Locher1 and Matt Lindenberg2

Grand Valley State University, Biology Department, Allendale, Michigan
1
lochera@gvsu.edu; 2lindenbm@gvsu.edu

Many analyses in natural resources or ecology-related fields, specifically wildlife-focused disciplines,

investigate wildlife habitat use and movement patterns for population and habitat management. Kernel
density estimates and home range analysis are commonly used by wildlife biologists for such
investigations. Methods to conduct home range analyses are often complex and require use of multiple
software programs. In the literature and on the web, it is difficult to find comprehensive instructions on
how to create home ranges and proceed with analyses. The purpose of this manual is to describe the
process of synthesizing raw location point location data, creating home ranges (kernel density
estimates), and begin initial analyses involving habitat selection.

Software used for home range creation and analysis

In this manual, we used the following software for home range creation and analysis.

1. Geospatial Modeling Environment (freeware) (http://www.spatialecology.com/gme/)

2. R-Studio (freeware) (https://www.rstudio.com/products/rstudio/download3/)
3. ArcGIS 10.2 (license required) (http://www.esri.com/)

Formatting tabular data and creating a comma -delimited file

If the point locations for individual animals are already in a shapefile, you can skip to step 5. The
following steps explain how to create a shapefile from point location coordinates stored as a tabular file
such as in a spreadsheet.

1. Input point locations (coordinates) into Excel for all individuals. Data must be formatted with at
least 2 columns: one column indicating the X coordinate (easting) and one column indicating the
Y coordinate (northing). You MUST know the coordinate system in which the point locations
were collected. For instance, in the table below (left), point locations were collected in a
geographic coordinate system in units of decimal degrees. In the table below (right), point
locations were collected in a projected coordinate system (Universal Transverse Mercator) in
units of meters. You may also include an ID column indicating a unique identifier for individual
animals.

1
 x-coordinate y-coordinate x-coordinate y-coordinate

Individual Longitude Latitude Individual UTMx UTMy

1 92.54985 43.265201 1 444782 545567
1 92.54973 43.265004 1 444793 545409
1 92.54899 43.264951 1 444856 545123
2 92.43665 43.465006 2 445002 540998
2 92.43591 43.464996 2 445096 540336
2 92.43602 43.465011 2 445147 540191

MAKE SURE THAT ALL CELLS ARE FORMATTED AS NUBERS and NOT TEXT!! If there are cells
formatted as text, they will not plot as points when you create a shapefile. To ensure that cells
are formatted as numbers, select all cells, right-click and select format cells. Choose NUMBER.

2. Save the file as a text file with comma-separated values (i.e., CSV), and close Excel.

3. Use ArcMap to create a shapefile from the CSV because a shapefile MUST be input into
Geospatial Modeling Environment (GME).
a. Open ArcMap.

b. Add data  navigate to where the CSV is stored and add it into the ArcMap document.

c. Open the attribute table (right-click on the file name in the Table of Contents to the left
of the map document, select “Open Attribute Table”) to make sure the coordinates look
correct.

d. In the Table of Contents, right-click the file name and select Display XY data.

e. Specify the correct X and Y coordinates AND define the appropriate coordinate system.
For instance, if the point locations were collected in lat-longs, then you would specify
the correct coordinate system as a Geographic Coordinate System (e.g., GCS WGS 1984).
If the point locations were collected in UTMs, then the correct coordinate system would
be a Projected Coordinate System (e.g., UTM Zone 16 North; refer to figure below).

2
 Unprojected Geographic Coordinate Projected Coordinate System
System

Click here to DEFINE the

appropriate coordinate system

f. Click OK in the “Display XY Data” dialog box. Then click OK again in the box indicating
that the “Table Does Not Have an Object-ID Field.” This step creates an Event Theme
(not a shapefile yet). An Event Theme displays the points, but it is not an actual
shapefile. Thus, you will not be able to conduct any analyses until a shapefile is created.

g. To create a shapefile,
right-click on the name
of the Event Theme in
the Table of Contents.
Navigate to Data 
Export Data. Make sure
you are specifying the
format as a shapefile
(and not a File
Geodatabase unless you
have created a
geodatabase).

3
 VERY IMPORTANT: THE SHAPEFILE MUST BE PROJECTED! If your point locations were collected
in lat-longs, and therefore your shapefile was defined in a geographic coordinate system, YOU
MUST NOT LEAVE IT IN A GEOGRAPHIC COORDINATE SYSTEM! Home ranges CANNOT be
created using GME for a shapefile that is unprojected.

i. To reproject a shapefile in ArcMap, use the “Project” tool in Data Management

Tools  Projections and Transformations  Feature  Project. Select an
appropriate projected coordinate system (e.g., UTM).

Creating point shapefiles for individual animals

4. If the locations of several animals are included in the shapefile you created in the steps above,
you must create a separate shapefile for each individual. Isolating points for each animal may
require an attribute query. For instance, you might select all records for “Animal 1” and create a
separate shapefile. If you have 5 animals total, then you need to create 5 separate shapefiles:
each shapefile would represent all the locations of one animal. Name the shapefiles something
descriptive such as “ID_001.shp”, “ID_002.shp”, “ID_003.shp”, etc.

Setting up R-studio

5. Launch R-studio.

6. In R-studio, activate the “Packages” tab in the lower right box and click on “Install”. Install ks.

4
 7. Make sure to check the boxes next to KernSmooth, ks, misc3d, mvtnorm, rccp, rgl.

Geospatial Modeling Environment (GME)

8. Launch GME .

Calculating Home Ranges (kernels and isopleths) in GME

9. Run the KDE (kernel density estimator) command in GME. You can access the KDE command by
typing “kde” into the command search box, and selecting the KDE command.

10. Input the point shapefile for Animal 1 (i.e., ID_001.shp). [RECALL that the shapefile MUST be in a
projected coordinate system.] Specify the output location and SAVE the output with an *.img
extension. Use LSCV for bandwidth estimation [see Horne and Garton (2006) article for
justification of using this method]. Specify an appropriate cell size. For instance, if you were
analyzing selection of vegetation classified from Landsat TM imagery, you might consider
selecting a cell size of 30 m to be consistent with the resolution of Landsat imagery. Appropriate

5
 cell size should be based on the ecology of the species you are studying and the resolution of
the spatial data you are using for the home range analysis. Once all inputs are specified, click
“Run”.

11. Once the command has run, you may wish to record the x variance, y variance, and xy
covariance values of the bandwidth. This information will be displayed in the “Output” tab.

If you would like to compute isopleths, then you MUST NOT CLOSE GME!!!! OTHERWISE
ISOPLETHS WILL NOT RUN and you will have to re-do all the KDEs.

12. Run the ISOPLETHS command in GME. Input a KDE file. The name of the output file MUST BE as
short as possible (e.g., 4 characters) otherwise the tool will not run. Specify desired quantiles
(e.g., 50%, 75%, 90%, 95% percent volume contours). The format of the input code for quantiles
should be as follows: c(0.5,0.75,0.9,0.95). If you would like a polygon shapefile created in
addition to a line shapefile depicting the isopleths, you must specify that you would like optional
polygons created. The polygon shapefile will be useful for computing area of the home range
and each quantile. In the example below, the polygon shapefile of isopleths is named
0142isopleth.shp. Once all inputs are specified, click “Run”.

6
 Name of output file
must be SHORT.

Creates a polygon
shapefile of isopleths

13. TROUBLESHOOTING: creating isopleths seems temperamental at times. If you get an error
message, try saving the shapefiles in a different folder.

Home Range Analysis using ArcMap

Calculation of Area

14. In ArcMap, view the kernels and isopleth files.

Kernel, isopleths (line file), and point Kernel, isopleths (polygon file), and
locations for animal 0142 point locations for animal 0142

7
Determining AREA of home ranges

15. GME does not create a field in the attribute table of the polygon isopleth file indicating percent
volume contours. The only field created in the polygon file is “ID”, but the records are all 0. To
specify percent volume contour, select the first record, right-click on the ID field, and use the
field calculator to input “95”. Repeat for the other contours.

16. To determine area of each home range, add a new field in the attribute table by clicking the
drop-down menu button in the upper left of the table and selecting “Add Field”. Name it
“Area”. Specify the new field as a “FLOAT” with Precision of 10 and Scale of 2.

17. Once the new field is created, right-click on the field name in the attribute table, navigate to
“Calculate Geometry”, and specify the appropriate units (e.g., hectares, square kilometers).

8
18. To determine the total area of the home range, right-click on the “Area” field and select
“Statistics”. The sum of the area between each isopleth will be given.

Vegetation Analysis – Determining “Available” Habitat

In ArcMap, at this point, you should have your isopleth shapefile of the home range for an individual
animal, and a layer indicating vegetation type. In this example, we use a raster file depicting land
cover and ultimately convert it to a vector shapefile for additional analyses. These spatial analyses
may also be done using all rasters if you are familiar with raster analyses, but the methods will be
different than those indicated below. For this analysis, we defined “available habitat” as the
vegetation available within the boundaries of the home range. We analyzed available habitat for
each percent volume contour (isopleth). For instance, the steps below describe how we determine
available habitat in the area between the 90 and 95% isopleths.

19. The land cover raster file obviously spans a larger spatial extent than an individual home range.
To “Clip” the land cover raster with the extent of the home range, use the “Raster Clip” tool.
Navigate to Data Management Tools  Raster  Raster Processing  Clip. Input is the land
cover raster. The Output extent is the home range isopleth shapefile. BE SURE TO CHECK THE
BOX SPECIFYING THAT YOU WANT TO “Use Input Features for Clipping Geometry”. Name the
file something descriptive like 0142_lc.img (i.e., indicating animal ID 0142 and land cover).

9
20. Convert the clipped raster to vector by using a conversion tool. Navigate to Conversion Tools 
From Raster  Raster to Polygon. Input is raster file from previous step (0142_lc.img). The
“Field” input should be the “value” field or other field in the raster file indicating land cover
types. Name the file something descriptive like 0142_lcov.shp (i.e., indicating animal ID 0142
and land cover).

Land cover raster “clipped” with home Land cover converted to a vector data
range. model (shapefile).

21. To create a shapefile indicating home range, isolines, and vegetation, you now must intersect
the newly converted shapefile of vegetation (0142_lcov.shp) with the home range isopleth
shapefile of the same individual (i.e., 0142isopleth.shp. See step 12). The “intersect” tool may
be accessed by navigating to Analysis Tools  Overlay Intersect. Input the 2 files. Name the
output something descriptive such as 0142_intersect.shp.

Input features include a polygon

file of isolines (created in step 12)
and land cover within home range
(created in step 20).

10
22. Now that you have a shapefile depicting percent volume contours (isopleths) and land cover
within a home range, you are ready to determine the amount of area available between
isopleths. Any time you perform a geoprocessing function such as an intersect, ArcMap does not
automatically update area. To update area, open the attribute table for your shapefile of the
land cover – isopleth intersection (i.e., 0142_intersect.shp). UPDATE the AREA FIELD by right-
clicking the “Area” field and selecting “Calculate Geometry”. Choose appropriate units (e.g.,
hectares, square kilometers).
Right-click and select
“Calculate Geometry”

NOTE: the sum of the AREA will exceed the total area of the home range because polygons are
essentially duplicated if they occur within the smaller isopleths. For instance, polygons
occurring in the area between the 50 and 75% isopleths ALSO occur within the area between
90-95% isopleths. DO NOT USE THE SUM OF THE AREA TO COMPUTE TOTAL HOME RANGE
SIZE. Use steps 15-18 to determine home range area.

23. To effectively summarize the amount of area within

each land cover category between isopleths, first
conduct an attribute query to select the area you are
interested in analyzing. For instance, conduct a query
for all polygons in the 95% isopleth.

24. Next, Right-click the field indicating land cover type (in
the example, land cover is indicated in the
“GRIDCODE” field) and select “Summarize”. In the
”Summarize” dialog box, check the box next to “Sum”
of “Area”. Save the output table as a *.txt file. Name it
0142_VegAvail95.txt. Saving the file with a *.txt

11
 extension will allow you to open the file in a spreadsheet such as Excel for further analysis. This
table will indicate the total area available for each land cover type.

Vegetation Analysis – Determining “Used” Habitat

We assumed that “used” habitat was vegetation corresponding with the point locations of an
animal.

25. To determine “used” habitat, you need

the shapefile of point locations for an
individual animal (i.e., ID_0142.shp; see
step 4) and the shapefile of the
intersection between land cover and
home range (i.e., 0142_intersect.shp; see
step 21.) Make sure both of those files are
added to ArcMap.

26. Determining “used” habitat requires

conducting a spatial join between the
point locations and the shapefile depicting
land cover. To determine habitat used for
locations occurring within the entire home
range (i.e., between the 0 and 95%
isopleths), right-click the point file in the
Table of Contents, and select Joins and
Relates  Join. Specify “Join attributes
based on spatial location.” Join it to the

12
 intersected file. Name the output something like 0142_join95.shp to indicate a spatial join
involving all points within the home range boundary. This step will create another shapefile
indicating the vegetation type associated with each location.

All land cover within the land cover selected within the
boundary of the 95% isopleth boundary of the 50% isopleth

27. In the resulting attribute table, conduct an attribute query to select all records in the 95%
isopleth. Next, right-click on the field in the attribute table indicating land cover type (e.g.,
GRIDCODE) and select “Summarize”. None of the boxes need to be checked. This process will
indicate the number of point locations occurring within each land cover class. Save the output
table with a *.txt extension so you may open it in a spreadsheet. Name the output table
something like 0142_VegUsed95.txt.

28. If you are analyzing habitat use within various isopleths, you must first isolate point locations
occurring within isopleths. For instance, in this example, let’s determine land cover types used
within the boundary of the 50% isopleth. Therefore, conduct an attribute query on the
intersected shapefile (0142_intersect.shp) to select all records composing the area within the
boundary of the 50% isopleth.

29. Next, in the Table of Contents, right-click the intersected shapefile and navigate to Selection 
Create Layer From Selected Features. The new layer will be added to the top of the Table of
Contents. If you would like to rename it, click the name of the layer, and then click once more.
For instance, we can name the new layer something descriptive like “landcov50.” This process
does NOT create a new shapefile, it simply isolates selected records from a shapefile.

13
 30. Now, you can join the shapefile of point locations (ID_0142.shp) to the shapefile depicting
vegetation within the boundary of the 50% isopleth by right-clicking the name of the point file in
the Table of Contents and selecting Joins and Relates Join. In the dialog box, specify “Join
data from another layer based on spatial location”. Join it to the newly created layer, landcov50.
Name the output 0142_join50.shp.

31. Summarize as in step 27. Continue obtaining “Used” data for each animal.

Analysis of Use vs Availability

32. At this point, you will have data on habitat available and habitat used. Now you must conduct
quantitative analyses assessing selection or non-selection of specific vegetation types. There
are many methods of analyzing selection *e.g., Jacob’s index (Jacobs 1974), compositional
analysis (Aebischer et al. 1993)], and appropriate statistical methods may vary depending on
data normality, sample size, or specific objectives. For example, we used a Jacob’s index (Jacobs
1974) to determine selection or non-selection of vegetation types according to the following
formula:

D = (r – p) / (r – p + 2rp)

where r is the proportional amount of a vegetation type used, and p is the proportional
availability of the vegetation type. The resulting values range from -1 to + 1, where -1 indicates
non-selection and +1 indicates selection. To determine significance, we used a Kruskal-Wallis
non parametric test (because our data violated assumptions of normality.) A Wilcoxon Rank
test, with a Holms adjustment was used when significance was detected.

References

Aebischer N.J., Robertson P.A. & Kenward R.E. 1993: Compositional analysis of habitat use from animal
radio-tracking data. Ecology 74: 1313–1325.

Horne, J. S. and E. O. Garton. 2006. Likelihood cross-validation versus lease squares cross-validation for
choosing the smoothing parameter in kernel home-range analysis. Journal of Wildlife Management
70:641–648.

Jacobs J. 1974: Quantitative measurement of food selection. A modification of the forage ratio and
Ivlev’s electivity index. Oecologia (Berl.) 14: 413–417.

14