Habitat Analysis

From Anatrack Ranges User Guide
Revision as of 12:03, 6 November 2014 by Admin (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


Habitat analyses in Ranges can use data from vector maps (shapes, points) or raster maps (e.g. remote sensed imagery). Habitat content of a map can be assessed for the whole map or in rectangular or circular sub-sections. Maps can also be used together with edge files to estimate Habitat content of ranges as alternative indices of habitat availability or of habitat use, both for areas of habitat and for important Points within ranges. The most precise estimates of habitat use come from analysing Habitat at locations, with combined Location & range habitat analyses to estimate preference analyses within the same run. All results can be saved to output files that are suitable for further processing in spreadsheets. For analyses using raster data there are different options for estimating habitat areas in partial cells around shape edges. The make selections button can be used to restrict the analyses to a subset of the ranges and/or locations within the input files.

Habitat content of a map

This option leads to a submenu which gives the opportunity to assess the whole map, or sections of it

map rectangle

This allows you to enter the north, south east and western limits of a rectangular sample of the map. Ranges then estimates the habitat available within that rectangle.

map circle

You can manually enter coordinates for the centre of a circle and its radius.

Habitat content of ranges

This requires a range edge file (created in location analysis). It estimates the habitat composition of each range and finishes with a summary of the content of all ranges. The percentage of each habitat in each core is available in the Stats Viewer and output files. There are two output files:

  1. xxx_Hab_mapname.csv shows the area of the core followed by the percentage of each habitat within the range core
  2. xxx_Hab_mapnameareas.csv shows the area of the core followed by the area of each habitat within that core.

These .csv files with column headers can be double-clicked to open in Microsoft Excel, imported to an alternative spreadsheet, or loaded back into Statistics at a later time.

Output clip file

If you have chosen a vector shape (.ves) map file, then an output "clip file" will be created. The clip file will be the intersection of the edge and vector shape files. The file is output in edge format. This can be used to "clip" range edges, to exclude areas that individuals are unable to visit. For example a river outline can be used to excluded areas outside the river from fish home ranges. For example using fish.loc in the fish directory of sample data, you could create an edge file using one of the home-range estimators, then you could load \fish\river.ves into this option in habitat analysis and use it to create a new clipped edge file that does not contain areas outside of the river. This clipped edge file could then be used for further analyses such as overlap.

Note that if you clip an edge file using a more complex vector shapefile (e.g. <RangesFolder>\samples\blackbird\blackbird_map.ves), the resultant file will contain all of the overlapped polygon boundaries from the vector file, and hence may not be very useful.

Points within ranges

This requires a vector points file (e.g. of nests or feeding sites) and a range edge file (created in location analysis). It calculates which points/sites are within each range and finishes with the mean number of points contained in all of the ranges.

Habitat at locations

This requires a location file and a vector or raster map file. It estimates the habitat at each location point or in an ellipse around each location.

location points

This calculates the percentage of the locations in each range that are in each habitat type and produces a summary of the percentage of all of the locations that fall within each habitat type. If locations fall on a habitat boundary, it shows this on the display as "habitat1label/habitat2label" and attributes a half-share of each habitat to the location in the summary.

buffers around locations

For a conventional location file, you will be prompted to Input circle radius (m). This will determine the radius of a circle around each location. The default value is set to half of the tracking resolution defined in the location file, thus generating circles with a diameter equal to the tracking resolution.

The buffer option is particularly useful for providing an average value round locations on raster maps. Individual raster cells may have been categorised inaccurately but the general area may be representative, if an appropriate radius is used. The routine could also be used for a simple form of edge detection near locations, by gradually increasing the assessment circle round each location. When more than one habitat is included in the results then a location is less than the radius of the circle from that edge.

The percentage of each habitat type within the buffers surrounding all of the locations within each range is output to a statistics file (.csv) and automatically loaded in the Statistics window. A "location buffers" edge file showing the positions of the buffers around each location is also created and is automatically loaded to the main window.

Habitat preference in ranges

This assesses habitat in ranges and at locations during the same run. After the analysis run, the display shows the preference or avoidance of habitat at locations (with Jacob's index, values +1 to -1), compared with habitat in the range as a whole.

Count cell centres, corners or partial cell areas

When calculating areas within raster maps, you will be offered one of the following 3 options. Your choice will depend on your need for speed and the scale of the raster cells relative to the areas you are using. The different methods deal differently with raster cells that are partially included in your areas. If your areas contain hundreds of raster cells then one of the counting estimation methods should be fine. It is only when your areas are closer to the size of the raster cells that you should need to use the ‘partial areas’ option (and then you should be careful that you are not overestimating what information is contained within the raster data).

count cell centres(fast)

Counts only those raster cells for which the cell centre is within the shape.

count cell corners

For partial cells, counts the number of cell corners and divides by 4.

partial cell areas(slow)

Calculates the area of partial cells.