Point occurrence data for all plant species in NSW were accessed from digitised, vouchered herbarium specimens held within the Australian Virtual Herbarium (AVH). Data for all species occurring in NSW was downloaded from the Atlas of Living Australia (ALA)-AVH webservice in August 2014 (accessed here: http://api.ala.org.au). Whilst species were limited to those occurring in NSW in many cases their distributions extend beyond the state boundaries and these occurrences were always included in range metric calculations. Taxonomy was standardised to the Australian Plant Census following the protocols of the ALA.
A preliminary dataset of 1,902,937 species occurrence records was cleaned to remove taxonomic and spatial errors. Taxonomic cleaning removed records that were not identified to species level - including manuscript names and hybrids, records that were not for vascular plants (i.e. mosses, bryophytes), and records which were for species not native to Australia (i.e. introduced species identified in both an Australian exotic plant checklist (Randall, 2007) and with the tag 'naturalised' provided in the APC). Records lacking latitude and longitude coordinates were then discarded and the remaining spatially georeferenced records were cleaned by removing occurrences from outside Australia, duplicates (i.e. non-unique combinations of latitude, longitude and species name) and specimens taken from cultivated plants (either flagged as such in the AVH or containing the search terms cultivate*, garden, horticulture* or agricultur*. Data were generated for each intraspecific rank (i.e. subspecies, variety) of each taxon.
Following these procedures, the final dataset of 1,844,041 occurrence records represented 7,421 species from 1,188 genera and 201 families. Across all species, the number of occurrences per species ranged between 1 and 4,673, with a mean of 249 and a median of 140.
Range size was calculated using two methods - extent of occurrence (EOO) and area of occupancy (AOO). Broadly, EOO measures the area inside a polygon encompassing the occurrence records, whereas AOO sums the number of grid cells of a fixed size occupied by a species. AOO is considered a more conservative measure of range species than EOO, but may severely underestimate range size for species which are poorly represented in herbarium collections.
Prior to analysis all occurrence records were projected to an Albers equal-area projection. EOO was estimated by calculating the area in km2 of the α-hull derived from all occurrences (Burgman & Fox 2003). The α-hull is a modified minimum convex polygon created by linking all occurrence records with a set of non-intersecting triangles and removing those edges whose length is greater than the mean edge length of all triangles combined, multiplied by α (i.e. Li > L x α; where α = 0.3). The remaining area is taken as the range size in km2. This method provides a more conservative estimate of range size than does a polygon encompassing all records (i.e., minimum convex polygon) and minimises the underestimation of range size inherent in area of occupancy calculations (i.e., summing the area within occupied grid cells). These features make α-hull estimation particularly suited to studies using occurrence data collected in a non-systematic way, such as herbarium records (Burgman & Fox, 2003).
AOO was calculated following protocols used in the IUCN Red List. That is, the sum of the number of 2km x 2km grid cells occupied by all species occurrences.
Predominant range dimension was determined by calculating the maximum N-S (latitudinal) and maximum E-W (longitudinal) distance within the 5th and 95th percentiles of all occurrences within the range of the species. This provides a measure of range shape which is distinct from area.
Altitudinal range was calculated as the difference between the lowest and highest altitude locations occupied by species across its Australian range. Occurrence records were overlaid on a high-resolution (90m) digital elevation model of Australia and height above sea level (m) was extracted using the raster package in R.
Climate niche breadths were calculated as the difference between the lowest and highest values of climate variables encountered by species across its Australian range. Occurrence records were overlaid on gridded climate datasets and values were extracted using the raster package in R. Niche breadths were measured for 20 climate variables; the standard set of 19 bioclimatic variables and an aridity index (see Table 1 for full details). The 19 bioclimatic variables represent a collection of extreme and average measures of temperature and rainfall variation. The bioclimatic variable data was accessed from ANUCLIM1.0 via e-MAST. The ANUCLIM dataset resolution is 0.01° and is derived from interpolation of baseline point meteorological observations between the period 1970-2013. More details can be found about the derivation of these datasets on the EMAST portal. Data on aridity index (AI) was accessed from CGIAR-CSI. AI is calculated as the ratio of mean annual precipitation to mean annual potential evapotranspiration, where higher values represent regions of greater humidity (i.e., lower aridity). Values of AI < 0.5 are considered indicative of arid conditions.
Soil niche breadths were calculated as the difference between the lowest and highest values of soil variables encountered by a species across its Australian range. Occurrence records were overlaid on gridded soil datasets and values were extracted using the raster package in R. Soil data was extracted from the National Soil Attribute Maps available in the Soil and Landscape Grid of Australia. Twelve soil variables were targeted; available water capacity, bulk density, clay content, depth of regolith, depth of soil, effective cation exchange capacity, total nitrogen, pH, total phosphorous, silt, sand, organic carbon (see Table 2 for full details).
References:
Burgman, M. A., & Fox, J. C. (2003) Bias in species range estimates from minimum convex polygons: implications for conservation and options for improved planning. Animal Conservation, 6, 19-28.
Table 1: Climate variables used in niche breadth calculations
| VARIABLE | AKA | UNITS | DESCRIPTION |
|---|---|---|---|
| Mean diurnal range | dnrg | ° C | The mean of all the monthly diurnal temperature ranges. Each monthly diurnal range is the difference between that month's maximum and minimum temperature. |
| Isothermality | isot | The mean diurnal range divided by the annual temperature range, multiplied by 100. | |
| Precipitation of coldest quarter | pcld | mm | The coldest quarter of the year is determined (to the nearest week), and the total precipitation over this period is calculated. |
| Precipitation of driest month | pdmt | mm | The precipitation of the driest month. |
| Precipitation of driest quarter | pdry | mm | The driest quarter of the year is determined (to the nearest week), and the total precipitation over this period is calculated. |
| Annual precipitation | prec | mm | The sum of all the monthly precipitation estimates. |
| Precipitation seasonality | psea | The Coefficient of Variation is the standard deviation of the monthly precipitation estimates expressed as a percentage of the mean of those estimates (i.e. the annual mean). | |
| Precipitation of wettest quarter | pwet | mm | The wettest quarter of the year is determined (to the nearest week), and the total precipitation over this period is calculated. |
| Precipitation of wettest month | pwmt | mm | The precipitation of the wettest month. |
| Precipitation of warmest quarter | pwrm | mm | The warmest quarter of the year is determined (to the nearest week), and the total precipitation over this period is calculated. |
| Annual mean temperature | tavg | ° C | The mean of all the weekly mean temperatures. Each weekly mean temperature is the mean of that week's maximum and minimum temperature. |
| Mean temp of coldest quarter | tcld | ° C | The coldest quarter of the year is determined (to the nearest week), and the mean temperature of this period is calculated. |
| Mean temp of driest quarter | tdry | ° C | The driest quarter of the year is determined (to the nearest week), and the mean temperature of this period is calculated. |
| Max temp of warmest month | tmax | ° C | The highest temperature of any monthly maximum temperature. |
| Min temp of coldest month | tmin | ° C | The lowest temperature of any monthly minimum temperature. |
| Temperature annual range | trng | ° C | The difference between the maximum temperature of the warmest month and the minimum temperature of coldest month. |
| Temperature seasonality | tsea | Standard deviation of mean monthly temperature, multiplied by 100. | |
| Mean temp of wettest quarter | twet | ° C | The wettest quarter of the year is determined (to the nearest week), and the mean temperature of this period is calculated. |
| Mean temp of warmest quarter | twrm | ° C | The warmest quarter of the year is determined (to the nearest week), and the mean temperature of this period is calculated. |
| Aridity index | arid | The ratio of mean annual precipitation to mean annual potential evapotranspiration, where higher values represent regions of greater humidity (i.e., lower aridity). Values of AI < 0.5 are considered indicative of arid conditions. |
Table 2: Soil variables used in niche breadth calculations
| VARIABLE | AKA | UNITS | DESCRIPTION |
|---|---|---|---|
| Available water capacity | awc | % | Available water capacity computed for each of the specified depth increments |
| Bulk density - whole earth | bdw | g/cm3 | Bulk density of the whole soil (including coarse fragments) in mass per unit volume by a method equivalent to the core method |
| Clay | cly | % | < 2 um mass fraction of the < 2 mm soil material determined using the pipette method |
| Depth of regolith | der | m | Depth to hard rock. Depth is inclusive of all regolith. |
| Depth of soil | des | m | Depth of soil profile (A & B horizons) |
| Effective cation exchange capacity | ece | meq/100g | Cations extracted using barium chloride (BaCl2) plus exchangeable H + Al |
| Total nitrogen | nto | % | Total nitrogen |
| pH - CaCl2 | phc | pH of 1:5 soil/0.01M calcium chloride extract | |
| Total phosphorous | pto | % | Total phosphorus |
| Silt | slt | % | 2-20 um mass fraction of the < 2 mm soil material determined using the pipette method |
| Sand | snd | % | 20 um - 2 mm mass fraction of the < 2 mm soil material determined using the pipette method |
| Organic carbon | soc | % | Mass fraction of carbon by weight in the < 2 mm soil material as determined by dry combustion at 900° C |