Input file type
Output prefix
Genus column
Species column
X column
Y column
Cell size
Species threshold
Range threshold
Template file
Do dropper
Species richness threshold
Iterations
Matrix file
Number of classes
Merge type
Use taxa
Acceptable forms are:
The ORGD format is as follows, and contains keywords and data.
BEGINCELL Xcoord Ycoord
genus species
genus species
genus species
ENDCELL
BEGINCELL Xcoord Ycoord
genus species
genus species
ENDCELL
There are as many "genus species" pairs between the BEGINCELL and ENDCELL keyword lines as there are unique species in the cells. An example with real data is:
BEGINCELL 2175000 1125000
Pultenae pedunculata
ENDCELL
BEGINCELL 2225000 975000
Pultenae daphnoides
Pultenae densifolia
Pultenae scabra
ENDCELL
BEGINCELL 2225000 1225000
Pultenae elachista
Pultenae pedunculata
ENDCELL
BEGINCELL 2275000 1025000
Pultenae densifolia
ENDCELL
The transposed version is like this, and will only work in the analyses that are not explicitly spatial (do not require spatial neighbours).
BEGINCELL Pultenae parrisiae
3275000 725000
3375000 725000
3375000 825000
ENDCELL
BEGINCELL Pultenae elachista
1975000 1375000
1975000 1425000
2125000 1275000
2225000 1225000
2275000 975000
ENDCELL
BEGINCELL Pultenae myrtoides
3475000 1025000
3625000 1525000
3675000 1175000
3675000 1225000
3675000 1475000
3725000 1625000
3775000 1275000
3775000 1325000
ENDCELL
The output prefix is used in all the output filenames, including the zip file into which all the results are put.
If your output prefix is "biodiverse" then example output names would be "biodiverse.orgd", "biodiverse.log", and "biodiverse_specrich.txt". The ZIP file would be called "biodiverse.zip".
This is the column in the input file that represent the genus name (or whatever you are using in its stead).
Remember that counting starts at zero, so the default of 1 is actually the second column, 2 is the third column etc.
This is the column in the input file that represent the species name (or whatever you are using in its stead).
Remember that counting starts at zero, so the default of 2 is actually the third column, 3 is the fourth column etc.
If you only have one name for the input data (eg the genus), then set this to be equal to the genus column. The names in the output file will be "Eucalyptus eucalyptus", but you can edit that later if need be.
The column which represents the X coordinate. If you have data in latitude/longitude coordinates, then this should be the longitude (but note that it is a bad idea to use lat/long data because they are angular bearings and not distance units -- there are some pretty major distortions as you approach the poles).
The column which represents the Y coordinate. If you have data in latitude/longitude coordinates, then this should be the latitude (but again note that it is a bad idea to use lat/long data because they are angular bearings and not distance units -- there are some pretty major distortions as you approach the poles).
How large an area should be used to aggregate the units. At the moment it is a square cell. The value should be in your map units, eg 50,000 represents an area of 50x50 km for a data set in a Universal Transverse Mercator projection.
Exclude cells that have a richness less than this value. Default is one to include all cells. Note: The ranges are not adjusted to reflect the loss of any species from the data set.
NOT YET IMPLEMENTED. Exclude species that have a range greater than than this value. Default is one to include all species.
The set of neighbours to use for the explicitly spatial analyses.
Option for the randomisation analyses. If turned on then the randomisation will use the circular model, where species are assigned to cells in a circular pattern (very structured, very conservative). If turned off then a model of Complete Spatial Randomness is used (there is no structure in the randomisation, not very conservative).
NOT YET IMPLEMENTED. Force the randomisations to adhere to the species richness in the data set. A value of 1 is exact, values greater than 1 allow more flexibility in the allocation of species to cells. A value less than 1 will not allocate all species records (and may or may not work).
How many iterations to run the randomisations for.
A file containing a symmetric matrix of similarities between species. Must be comma or tab delimited but can contain any numerical value. The system ignores any values along and above the diagonal, as well as any blank lines before the first record. The system assumes that higher values are more similar, but at the moment the matrix values are inverted (subtracted from one) - see known issues on the main page.
1,Almaleea,cambage,-
2,Aotus,carinata,0.04408,-
3,Aotus,cordifolia,0.05258,0.02071,-
4,Bossiaea,buxifol,0.11886,0.12957,0.13027,-
9,Daviesia,mimosoi,0.14957,0.16061,0.16096,0.13407,-
How many classes do you want in the dendrogram. The system is still pretty basic, and the plotting routine is yet to be interfaced with the online version (it runs under ArcInfo Grid). Until the system generates output files compatible with other dendrogram plotting systems (eg the newick format), we are stuck with this approach.
The heuristic used for merging clusters. Average is a standard approach, the min is the complete linkage, and the recalc is the option where the entire matrix is recalculated after each merger (those that are affected by the merger, anyway).
When set, the system will use taxonomic clustering, but exclude those species that do not occur in the similarity matrix. You should be able to just generate a list of species to include, and not worry about all the additional similarity values, but I haven't tested it.
.