Biomaps - Mapping Australia's Biodiversity

Site Help

Using BioMaps – four broad steps

Map Help

Navigation Tools

Click on following Icons then interact with the map in the following ways:-

Draw a box on the map to zoom in

Click in center of map to zoom out

Click on Map and drag across to move the map

Roll over point to get specimen information

Click on map to get information for underlying area datasets

Draw line freehand or by clicking on map. Double click to finish line

Draw area freehand or by clicking on map. Double click to finish area.

Zooms to original extent

Interaction with Table

Highlight record on map – click on a record in the table and it will flash the point on the map and highlight that record in the table in yellow.

Exclude records – tick the exclude box and then Redraw map to remove specific points from the map. This is particularly useful when you want to create a distribution model using the points displayed on the map.

Toggle symbol records in table - Click on large Symbol Icon in tab bar at the top of the table to show records for that symbol in the table

Turn off symbol – by removing the tick from a symbols display tick box in the Tabs at the top of the table and then clicking on Redraw in the right hand side tab – all of that symbols points will be removed from the map. This is particularly useful when you want to create a distribution model using the points displayed on the map.

Reducing records in the table by Zooming in on map – when you zoom in on points on the map – the records in the table will reflect just those points visible on the map.

Layers

Turning layers on and off – layers on the map can be turned on and off by ticking or unticking the tick boxes next to the layer name

Modelling Help

Step 1: Select Modeling Technique

Choose either BioCLIM or DOMAIN - details below.

Bioclim

Bioclim (Busby 1986;Nix 1986) uses climatic variables (eg rainfall and temperature) and known occurrences of a species or group of species to predict a species potential distribution. This is often called the species climate envelope and refers to areas that are climatically suitable for a species based on areas we know a species exists (from Museum collections).

Bioclim has been used to model such things as butterflies (http://www.blackwell-synergy.com/doi/abs/10.1046/j.1365-2486.2002.00490.x?cookieSet=1), threatened plants (http://www.deh.gov.au/biodiversity/threatened/species/tetratheca-juncea.html) and weeds (http://invader.dbs.umt.edu/weedfree/methods.htm), long-footed potoroo’s (http://www.publish.csiro.au/?act=view_file&file_id=WR01035.pdf), gliders (http://www.publish.csiro.au/?act=view_file&file_id=ZO98044.pdf) , Superb parrot (http://www.publish.csiro.au/nid/96/paper/MU04057.htm), owls (http://www.sei.org/owl/finalreport/Appendices.doc), and plant distributions under future climate scenarios (http://www.york.ac.uk/res/celp/webpages/projects/bioclimatic/plants/introduction.htm)

Important notes:

Bioclim has a number of advantages over most other models in cases where there is presence-only data (i.e. no absence data), and where the number of known records is extremely low. This makes it a particularly useful modelling system for use with threatened species.
The main drawbacks of BIOCLIM include
- its tendency to over-predict (i.e. predict larger areas than might otherwise be expected),
- its general restriction to climate-based environmental variables (though other environmental variables have been used Chicoine et al. 1985, Walker and Cocks 1991, Sindel and Michael 1992, Shao and Halpin 1995).
- and its unreliability at fine scales.
The model is sensitive to spatial inaccuracies in the data, a collection site which is in the wrong place can seriously affect the modelled distribution.
There are many areas which may have climates similar to the species being modelled where the species does not occur. This is called the fundamental niche (Godown and Peterson 2000 (http://www.specifysoftware.org/Informatics/bios/biostownpeterson/GP_BC_2000.pdf). The areas where the species are actually found within the fundamental niche are called the realized niche.
The fundamental and realized niche’s are different usually because of factors such as hard boundaries like major rivers and mountain ranges, clearing, land use change etc.
Bioclim tends to overestimate the extent of modelled distributions. However because it is common to only have sampled a small subset of a species true range, models tend to be reasonable guesses at actual distributions.

DOMAIN

Domain Modelling Domain (Carpenter et. al. 1993) uses a similarity measure by transforming the known occurences into a environmental space and computing the minimum distance in environmental space from any cell to a known presence of the species. The result of this is a surface which is ranked for likelyhood of having a species occur. The DOMAIN algorithm is described in Carpenter, G., Gillison, A.N. and Winter, J. (1993). DOMAIN: a flexible modelling procedure for mapping potential distributions of plants and animals, Biodiversity and Conservation. 2, 667-680. DOMAIN was originally developed by Guy Carpenter and Andy Gillison at the CSIRO Tropical Forest Research Institute in Atherton, Queensland, Australia.

At its very simplest, this model generates maps of similarity or distance. For example to predict the potential distribution of a particular taxon, DOMAIN maps those regions which are most similar to areas where the taxon is known to occur. The measure of similarity used in DOMAIN is based on the Gower metric. For any location in the mapping area, the values in the layer files define an environmental coordinate. For example if 3 layers are open containing rainfall, vegetation type and elevation, the environmental coordinate for any location is the set of three cell values from the three layers at that point. The Gower metric defines a means of computing the distance between any two such environmental coordinates.

The DOMAIN algorithm creates a new map layer and assigns each cell in the new layer the Gower distance between that cell and the closest point in the training set. If averaging is enabled (by setting the Avg Closest value in DOMAIN/Options) the value stored is the average of the n nearest cells. Generally n is left as 1. Larger values are useful in reducing the effect of outliers in the training points.

The application of DOMAIN is not limited to mapping potential distribution of taxa. The problem of similarity mapping arises in other areas. For example it may be useful to map the regions which are least similar to this set of survey sites when selecting new site locations or examing the adequacy of a sampling strategy. For more details on the Gower Metric or the DOMAIN alogorithm see the research paper described above. For a comparison of modelling methods see here (Pedro Segurado and Miguel B. Araujo. An evaluation of methods for modellingspecies distributions. Journal of Biogeography (J. Biogeogr.) (2004) 31, 1555–1568)

References

Austin, M. P. & Meyers, J. A. (1996) Current approaches to modelling the environmental niche of eucalypts: implication for management of forest biodiversity. Forest Ecology and Management, 85, 95-106.

Beerling, D. J., Woodward, F. I., Lomas, M. and Jenkins, A. J. (1997) Testing the responses of a dynamic global vegetation model to environmental change: a comparison of observations and predictions. Global Ecology And Biogeography Letters, 6, 439-450.

Busby, J.R., 1986. A biogeographical analysis of Nothofagus cunninghamii(Hook.) Oerst. in southeastern Australia. Aust. J. Ecol.11, 1–7.

Carpenter, G., Gillison, A.N.,Winter, J., 1993. DOMAIN: a flexiblemodeling procedure for mapping potential distributions of plants,animals. Biodivers. Conserv. 2, 667–680.

Chicoine, T. K., P. K. Fay and G. A. Nielsen. 1985. Predicting weed migration from soil and climate maps. Weed Science 34:57-61

Ferrier, S. & Watson, G. (1997) An evaluation of the effectiveness of environmental surrogates and modelling techniques in predicting the distribution of biological diversity. Consultancy report prepared by NSW National Parks and Wildlife Service for Department of Environment, Sport & Territories. Environment Australia, Canberra

Godown, M.E.,and Peterson, T. 2000. Preliminary distributional analysis of US endangered bird species. Biodiversity and Conservation 9: 1313–1322, 2000.

Nix, HA. (1986) A biogeographic analysis of Australian Elapid Snakes. In. Atlas of Elapid Snakes of Australia. (ed.) R. Longmore pp. 415. Australian Flora and Fauna Series Number 7. Australian Government Publishing Service: Canberra.

Panella, F. D. & Mitchell, N. D. (1991) Homocline analysis and the prediction of weediness. Weed Research, 31, 273-284.

Panetta, F. D. & Dodd, J. (1987) Bioclimatic prediction of the potential distribution of skeleton weed Chondrilla juncea L. in Western Australia. Journal of the Australian Institute of Agricultural Science, 53, 11-16.

Rogers, D. J. & Williams, B. G. (1994) Tstetse distribution in Africa: seeing the wood and the trees. In Large-scale ecology and conservation biology. (Eds, Edwards, P. J., May, R. and Webb, N. R.) Blackwell Scientific Publications, Oxford, pp. 247-271.

Shao, G. and P. N. Halpin. 1995. Climatic controls of eastern North American coastal tree and shrub distributions. Journal of Biogeography 22:1083-1089.

Sindel, B. M. and P. W. Michael. 1992. Spread and potential distribution of Senecio madagascariensis Poir. (fireweed) in Australia. Australian Journal of Ecology 17:21-26.

Walker, P. A. and K. D. Cocks. 1991. HABITAT: a procedure for modelling a disjoint environmental envelope for a plant or animal species. Global Ecology and Biogeography Letters 1:108-118

Step 2 : Select Climate Scenario

Currently only a "Present" climate scenario is available. In the near future we hope to have a "Future" climate scenario based on that available through the desktop GIS program Diva GIS.

Step 3: Select Climate Variables

Choose one of the available options from the dropdown menu.

These variables lists were derived from the Geoscience Australia / Australian Geological Survey 9 second DEM and Bureau of Meteorology records by the ANUCLIM program. See here for definitions of each variable

Details of the variables for each of these options:-

4 pre:

Annual Mean Temperature
Annual Precipitation
Annual Mean Radiation
Annual Mean Moisture Index

8 pre:

Annual Mean Temperature
Annual Precipitation
Annual Mean Radiation
Annual Mean Moisture Index
Temperature Seasonality (Coefficient of Variation)
Precipitation Seasonality(Coefficient of Variation)
Radiation Seasonality (Coefficient of Variation)
Moisture Index Seasonality (Coefficient of Variation)

class rain:

Annual Precipitation
Precipitation of Wettest Period
Precipitation of Driest Period
Precipitation Seasonality(Coefficient of Variation)
Precipitation of Wettest Quarter
Precipitation of Driest Quarter
Precipitation of Warmest Quarter
Precipitation of Coldest Quarter

class temp:

All:

Annual Mean Temperature
Mean Diurnal Range(Mean(period max-min))
Isothermality (P2/P7)
Temperature Seasonality (Coefficient of Variation)
Max Temperature of Warmest Period
Min Temperature of Coldest Period
Temperature Annual Range (P5-P6)
Mean Temperature of Wettest Quarter
Mean Temperature of Driest Quarter
Mean Temperature of Warmest Quarter
Mean Temperature of Coldest Quarter
Annual Precipitation
Precipitation of Wettest Period
Precipitation of Driest Period
Precipitation Seasonality(Coefficient of Variation)
Precipitation of Wettest Quarter
Precipitation of Driest Quarter
Precipitation of Warmest Quarter
Precipitation of Coldest Quarter
Annual Mean Radiation
Highest Period Radiation
Lowest Period Radiation
Radiation Seasonality (Coefficient of Variation)
Radiation of Wettest Quarter
Radiation of Driest Quarter
Radiation of Warmest Quarter
Radiation of Coldest Quarter
Annual Mean Moisture Index
Highest Period Moisture Index
Lowest Period Moisture Index
Moisture Index Seasonality (Coefficient of Variation)
Mean Moisture Index of Highest Quarter MI
Mean Moisture Index of Lowest Quarter MI
Mean Moisture Index of Warmest Quarter
Mean Moisture Index of Coldest Quarter

Step 4: Select Analysis Cell Size

Choose one of three options from the drop down list.

Each option refers to the length of the side of each of the gridcells that are used in the analysis.

As a general idea - 5km provides a finer scaled analysis than does 20km

Step 5: Species Records Input Window

This window shows the list of records that will be used to create the model. You can edit this list in the window or cut and paste your own list of records.

IMPORTANT - each record must be in the format of - scientific name, longitude, latitude - otherwise the model wont work