GEOSS Ecosystem Map

Quick links

Overview

The physical drivers of ecosystem formation – macroclimate, lithology and landform – along with vegetation structural formations are key determinants of current ecosystem type. Each combination of these ecosystem drivers – each ‘ecological facet’ – provides a unique set of opportunities and challenges for life.

The Australia GEOSS ecosystem mapping models these physical drivers of ecosystem formation and identifies these unique ecological facets. This is the highest resolution (90 m), Australia-wide terrestrial mapping of its sort to-date. While acknowledging that other factors also influence ecosystem occurrence, especially disturbance from anthropogenic and natural sources, understanding the physical drivers should facilitate management and conservation.

By understanding the magnitude and distribution of unique combinations of these drivers, management strategies can plan for their full range of variation, and conservation efforts can ensure that unique ecosystems are not lost. Additionally, by improving our understanding of the past and present conditions that have given rise to current ecological facets this dataset could facilitate future predictive environmental modelling. Finally, this data could assisting biodiversity conservation, climate change impact studies and mitigation, ecosystem services assessment, and development planning.

Method

Ecosystem type is determined by the interplay of several factors operating at a range of scales: macroclimate, lithology and landform. At the broadest scale, ecosystem type is determined by macroclimate, the energy regime resulting at different latitudes, and the influence of this energy regime on water availability (O'Brien 2006; Bailey 2014). At finer scales, landform and lithology become important drivers of vegetation distribution because they modify the water-energy regime by influencing soil, evapotranspiration, precipitation, temperature, wind and cloud regimes, and these in turn determine substrate chemistry, soil water availability and air saturation, heat balance and photosynthetically active radiation (Guisan and Zimmermann 2000; Sayre et al. 2014).

Each of the three categories of physical ecosystem drivers are captured by one or two spatial indicators, described in more detail in their dedicated sections below.

Macroclimate

Open

Bioclimatic regime, or the availability of water and energy, is a key driver of ecosystem function and biodiversity. Indeed, there is evidence that the majority of variation in species richness of plants (Wright 1983; Currie and Paquin 1987; Adams and Woodward 1989; O'Brien 1993, 1998; O'Brien et al. 2000; Venevsky and Venevskaia 2005; Kreft and Jetz 2007), mammals (Currie 1991; Badgley and Fox 2000), butterflies (Hawkins and Porter 2003) and bird species (Currie 1991; Hawkins et al. 2003) at broad scales is determined by climatic variables associated with water and energy availability.

Homogeneous bioclimatic regions were mapped by clustering a large set of un-correlated bioclimatic variables into homogenous classes in a method modelled after that of Metzger et al. (2013). An initial large set of bioclimatic variables and indices (those used by Sayre et al. (2009), as per the method of Rivas-Martinez and Rivas y Sáenz (2009)) was initially calculated from eMast monthly 1976 – 2005 mean daily maximum and minimum temperatures, and monthly mean precipitation grids (0.01 degree resolution). This set was then culled to remove variables dependent on pre-defined thresholds (with no evidence base to support their validity for Australian conditions) and categorical variables, due to the methods reliance on iterative self-organizing data analysis technique (ISODATA clustering)(Ball and Hall 1965).

To prevent the clustering being unduly influenced by more frequently used or correlated variables a correlation matrix was then constructed for the reduced set of 86 variables. Where correlation coefficients greater than 0.90 were identified, variables were removed from further consideration. If a variable was only highly correlated (> 0.90) with one other variable, the retained variable was chosen arbitrarily. If a variable was highly correlated with more than one variable it was retained and the others discarded. However, an exception to this occurred when three variables were highly correlated with each other (It, Tmax and Tp)(and so only one would normally have been retained), but each was also highly correlated with an exclusive set of other variables. In this case, retaining these three variables allowed for the exclusion of 40 other variables.

Lithology and weathering intensity

Open

Previous GEOSS ecosystem mapping has used surficial lithology (Sayre et al. 2008; Sayre et al. 2009; Sayre et al. 2013; Sayre et al. 2014) to capture the important role substrate type plays in determining vegetation distribution (Kruckeberg 2004). Indeed, surficial lithology is strongly predictive of several important soil properties, and therefore potentially predictive of ecosystem type. For instance, Gray et al. (2014) classified lithology into eleven classes based on broad chemical composition, and demonstrated that inclusion of this classification could improve ability of soil property modelling to predict soil organic carbon, pH, sum of bases and sand content. However, given the extreme age of most Australian soils, weathering intensity is perhaps just as important as surficial lithology (Gale 1992; Oliver 2001). Weathering is the process whereby chemical and mechanical processes break-down or chemically alter surficial material. Thus, we include both surficial lithology and weathering intensity datasets to capture both substrate type and degree of weathering.

A Surface Geology dataset produced by Geoscience Australia at 1: 1M scale was used as the surficial lithology dataset. While previous GEOSS mappings have reclassified lithology to few broad categories (e.g., 17 for North America (Sayre et al. 2009)) the Geoscience Australia Surface Geology dataset is quite detailed, distinguishing 244 different lithologies. Future work could attempt a similar reclassification of the Geoscience Australia Surface Geology dataset if less detail is desired. The lithology dataset is not pictured here, due to the large number of attributes.

Weathering intensity was represented by the weathering intensity index (WII) produced by Wilford (2012). The WII is produced from the integration of two regression models based on gamma-ray spectrometry and topography. The WII has a range from 1 – 6, and for the purposes of this was reclassified to four weathering categories: Low weathering (≤ 2), Moderate weathering (> 2 ≤ 4), High weathering (> 4 ≤ 5) and Extreme weathering (> 5 ≤ 6).

Landform

Both the physical shape of the earth surface and effect of geography on moisture availability strongly influence ecosystem type. We capture these with an index of land surface form and topographic moisture potential.

Land surface form

Nine land surface form classes (flat plains, smooth plains, irregular plains, escarpments, low hills, hills, breaks/foothills, low mountains, and high mountains/deep canyons) were modelled from topographic variables (slope and local relief). These land surface forms were derived by reclassification of the Australian CSIRO Slope Relief dataset, created by John Gallant (CSIRO) and John Wilford (GA) in 2011. This dataset was produced through classification of slope relief from the 1 second DEM-S, and the 300 m focal median percent slope product. The slope and relief classes in the product were derived from Speight 2009 with some modification. Finally, a drainage channels class was derived via the slope position classification method of Weiss (2001).

Topographic moisture potential

While the long-term climatic effect on water-energy balance is characterised by the isobioclimate mapping, the available moisture is also influenced by topographic redistribution of rainfall and the influence of aspect on evaporation. To account for these influences a topographic moisture potential index was developed (example map(s)). This index was produced from a combination of topographic index, and remotely sensed inundation frequency.

The topographic wetness index (TWI; also known as the compound topographic index (CTI)), is derived from slope and flow accumulation, and expresses the relative wetness of each point. This index can potentially capture how likely a given part of the landscape is to shed or accumulate rainfall runoff or subsurface flow, and previous work in North America had established a strong link between the TWI and areas of known wetland versus non-wetland (Sayre et al. 2009), and to subdivide the non-wetland landscape in to several categories of dryness.

However, comparison with known wetlands revealed that TWI was a poor indicator of the wettest parts of the landscape. Thus, TWI was used to define drier moisture potential categories; 'Lower slope', 'Mid slope' and 'Upper slope' (in increasing order of dryness). Additionally, to capture the important role aspect plays in determining evaporation on steeper slopes (Wu et al. 2006; McVicar et al. 2007; Wang et al. 2011), the ‘Upper slope’ category was reclassified to ‘Dry upper slopes’ where slope was greater than 25% and aspect was northerly. Slopes greater than 20 % were calculated from the 3” version of the CSIRO product “Slope derived from 1” SRTM DEM-S”(Gallant and Austin 2012b), and northerly aspect was derived from CSIRO product “Aspect (3” resolution) derived from 1” SRTM DEM-S”(Gallant and Austin 2012a).

To capture wetter areas of the landscape the Geoscience Australia Water Observations from Space (WOfS; Mueller et al. 2016) product was compared to the above defined wetland / non-wetland areas (Table 8). The WOfS product summarises long-term inundation frequency (1987 to present) for all of Australia, and is produced from all clear Landsat 5 and 7 observations for that period. Two additional categories, ‘Intermittently flooded’ and ‘frequently or permanently inundated’ were defined as 'inundation frequency ≥ 2 % and < 10 %', and 'inundation frequency ≥ 10 %' respectively. The TWI and WOfS defined categories were then combined, with the WOfS derived categories overriding the TWI derived categories.

Vegetation structural formations

Whereas the previous sections have modelled the drivers of ecosystem formation, vegetation cover is the vegetative response to those drivers (Sayre et al. 2014). The density and structure of the dominant vegetation type plays a significant role in influencing ecosystem type. The amount of vegetation is a determining factor of system-wide primary productivity, and all higher trophic levels are dependent for their energy requirements on this primary productivity. In conjunction with the magnitude of primary productivity, the vegetation structure determines the number and type of niches present. Thus, the final step in producing ecological facets was to produce a map of vegetation structural formations for Australia.

This was done by re-classifying the National Vegetation Inventory System (NVIS) Major Vegetation Group (MVG) data to capture the major vegetation structural formations. The original NVIS MVG contained 33 classes, and each class was a combination of dominant upper story species and vegetation structural formation (e.g., Eucalypt Open Forests), and the reclassified major vegetation structural formations contained 6 vegetation structural classes, and four non-vegetation classes.

Our reclassification transformed the NVIS MVG data to the vegetation structural formations outlined by Specht (1972), ignoring height classifications since many of the original NVIS MVG classes did not contain height descriptors. Additionally, the vegetation structural formations layer was reprojected to the same projection and resolution as the other data layers (GDA94 and 3” respectively). This approach was taken to keep the focus on vegetation structure. An alternative and equally defensible approach would have been to simply use the NVIS MVG data. As it is, our approach produced 369,439 unique ecological facets, and using the full NVIS MVG data would have produced approximately five times as many.

Ecological facets

A continental dataset of ecological facets – unique combinations of ecosystem drivers and vegetation structural formations – was produced by combining the spatial indicators of macroclimate, lithology, landform, and vegetation structural formations described in the sections above. This final dataset contained 369,439 unique ecological facets at a resolution of 90 m. All attribute values and descriptions for each input indicator were retained for every pixel.

While the ecological facet dataset is incredibly rich in detail, this richness may be perceived as a negative for practical and management perspectives (Sayre et al. 2014). However, this dataset retains all of the detail of the component indices, and hence allows examination of each ecosystem driver across continental Australia either individually or in combinations. Further, based on the specific need, detail may be reduced by aggregating some of the detail in a lower priority ecosystem driver index. By gathering these data into one source and ensuring spatial consistency, this ecological facets dataset will allow for better research and management of the biophysical variation within and across Australia.

Metadata summary for each layer

Macroclimate class

Lithology

Weathering intensity

Land surface form

Topographic moisture potential

Vegetation structural formations

Ecological facets

References

Adams, J.M., & Woodward, F.I. (1989). Patterns in tree species richness as a test of the glacial extinction hypothesis. Nature, 339, 699-701

Badgley, C., & Fox, D.L. (2000). Ecological biogeography of north america mammals: Species density and ecological structure in relation to environmental gradients. Journal of Biogeography, 27, 1437-1467

Bailey, R.G., (2014). Ecoregions: The ecosystem geography of the oceans and continents. SpringerLink: (Online service)

Ball, G.H., & Hall, D.J. (1965). Isodata: A method of data analysis and pattern classification. Office of Naval Research. Information Sciences Branch, Stanford Research Institute, Menlo Park,United States.

Currie, D.J. (1991). Energy and large-scale patterns of animal-species and plant-species richness. American Naturalist, 137, 27-49

Currie, D.J., & Paquin, V. (1987). Large-scale biogeographical patterns of species richness of trees. Nature, 329, 326-331

Gale, S.J. (1992). Long-term landscape evolution in australia. Earth Surface Processes and Landforms, 17, 323-343

Gallant, J., & Austin, J. (2012a): Aspect (3" resolution) derived from 1" srtm dem-s.2012a v5. CSIRO. Data Collection. http://doi.org/10.4225/08/5109FC2B2C21D.

Gallant, J., & Austin, J. (2012b): Slope derived from 1" srtm dem-s.2012b v4. CSIRO. Data Collection. http://doi.org/10.4225/08/5689DA774564A.

Gallant, J., & Austin, J. (2012c): Topographic wetness index derived from 1" srtm dem-h.2012c v2. CSIRO. Data Collection. http://doi.org/10.4225/08/57590B59A4A08.

Gray, J.M., Bishop, T.F.A., & Wilford, J.R. (2014). Lithology as a powerful covariate in digital soil mapping. In D. Arrouays, N. McKenzie, J. Hempel, A.R.d. Forges & A. McBratney (Eds.), Globalsoilmap: Basis of the global spatial soil information system (pp. 433-439): CRC Press 2014

Guisan, A., & Zimmermann, Niklaus E. (2000). Predictive habitat distribution models in ecology. Ecological Modelling, 135, 147-186

Hawkins, B.A., & Porter, E.E. (2003). Water-energy balance and the geographic pattern of species richness of western palearctic butterflies. Ecological Entomology, 28, 678-686

Hawkins, B.A., Porter, E.E., & Diniz-Filho, J.A.F. (2003). Productivity and history as predictors of the latitudinal diversity gradient of terrestrial birds. Ecology, 84, 1608-1623

Kreft, H., & Jetz, W. (2007). Global patterns and determinants of vascular plant diversity. Proc. Nat. Acad. Sci. USA, 104, 5925-5930

Kruckeberg, A.R., (2004). Geology and plant life: The effects of landforms and rock types on plants. University of Washington Press: Seattle, WA, USA, 362 pages.

McVicar, T.R., Van Niel, T.G., Li, L., Hutchinson, M.F., Mu, X., & Liu, Z. (2007). Spatially distributing monthly reference evapotranspiration and pan evaporation considering topographic influences. Journal of Hydrology, 338, 196-220

Metzger, M.J., Bunce, R.G.H., Jongman, R.H.G., Sayre, R., Trabucco, A., & Zomer, R. (2013). A high-resolution bioclimate map of the world: A unifying framework for global biodiversity research and monitoring. Global Ecology and Biogeography, 22, 630-638

Mueller, N., Lewis, A., Roberts, D., Ring, S., Melrose, R., Sixsmith, J., Lymburner, L., McIntyre, A., Tan, P., Curnow, S., & Ip, A. (2016). Water observations from space: Mapping surface water from 25 years of landsat imagery across australia. Remote Sensing of Environment, 174, 341-352

O'Brien, E. (2006). Biological relativity to water-energy dynamics. Journal of Biogeography, 33, 1868-1888

O'Brien, E.M. (1993). Climatic gradients in woody plant species richness: Towards an explanation based on an analysis of southern africa's woody flora. Journal of Biogeography, 20, 181-198

O'Brien, E.M. (1998). Water-energy dynamics, climate, and prediction of woody plants species richness: An interim general model. Journal of Biogeography, 25, 379-398

O'Brien, E.M., Field, R., & Whittaker, R.J. (2000). Climatic gradients in woody plant (tree and shrub) diversity: Water-energy dynamics, residual variation, and topography. Oikos, 89, 588-600

Oliver, C., D. (2001). Evolution of the australian landscape. Marine and Freshwater Research, 52, 13-23

Pachepsky, Y.A., Timlinb, D.J., & Rawlsa, W.J. (2001). Soil water retention as related to topographic variables. Soil Science Society of America Journal, 65, 1787-1795

Rivas-Martinez, S., & Rivas y Sáenz, S. (2009). Synoptical worldwide bioclimatic classification system (summarized table). Centro de Investigaciones Fitosociológicas [Phytosociological Research Center]. Accessed: 2017. Available at: http://www.globalbioclimatics.org/book/namerica2/table.htm.

Sayre, R., Bow, J., Josse, C., Sotomayor, L., & Touval, J. (2008). Terrestrial ecosystems of south america. In J.C. Campbell, K.B. Jones, J.H. Smith & M.T. Koeppe (Eds.), North america land cover summit. Washington, D.C.: Association of American Geographers

Sayre, R., Comer, P., Warner, H., & Cress, J. (2009). A new map of standardized terrestrial ecosystems of the conterminous unitied states. U.S. Geological Survey Professional Paper 1768, 17 p. (Also available online.),

Sayre, R., Dangermond, J., Frye, C., Vaughan, R., Aniello, P., Breyer, S., Cribbs, D., Hopkins, D., Nauman, R., Derrenbacher, W., Wright, D., Brown, C., Convis, C., Smith, J., Benson, L., Paco VanSistine, D., Warner, H., Cress, J., Danielson, J., Hamann, S., Cecere, T., Reddy, A., Burton, D., Grosse, A., True, D., Metzger, M., Hartmann, J., Moosdorf, N., Dürr, H., Paganini, M., DeFourny, P., Arino, O., Maynard, S., Anderson, M., & Comer, P. (2014). A new map of global ecological land units — an ecophysiographic stratification approach. Association of American Geographyers, Washington DC.

Sayre, R.G., Comer, P., Hak, J., Josse, C., Bow, J., Warner, H., Larwanou, M., Kelbessa, E., Bekele, T., Kehl, H., Amena, R., Andriamasimanana, R., Ba, T., Benson, L., Boucher, T., Brown, M., Cress, J.J., Dassering, O., Friesen, B.A., Gachathi, F., Houcine, S., Keita, M., Khamala, E., Marangu, D., Mokua, F., Morou, B., Mucina, L., Mugisha, S., Mwavu, E., Rutherford, M., Sanou, P., Syampungani, S., Tomor, B., Vall, A.O.M., Vande Weghe, J.P., Wangui, E., & Waruingi, L. (2013). A new map of standardized terrestrial ecosystems of africa. African Geographical Review.

Venevsky, S., & Venevskaia, I. (2005). Heirarchical systematic conservation planning at the national level: Identifying national biodiversity hotspots using abiotic factors in russia. Biological Conservation, 124, 235-251

Wang, L., Wei, S., Horton, R., & Shao, M.a. (2011). Effects of vegetation and slope aspect on water budget in the hill and gully region of the loess plateau of china. Catena, 87, 90-100

Weiss, A. (2001). Topographic position and landforms analysis. Poster presentation, ESRI User Conference, San Diego, CA.,

Wilford, J. (2012). A weathering intensity index for the australian continent using airborne gamma-ray spectrometry and digital terrain analysis. Geoderma, 183-184, 124-142

Wright, D. (1983). Species-energy theory: An extension of species-area theory. Oikos, 41, 496-506

Wu, W., Hall, C.A.S., Scatena, F.N., & Quackenbush, L.J. (2006). Spatial modelling of evapotranspiration in the luquillo experimental forest of puerto rico using remotely-sensed data. Journal of Hydrology, 328, 733-752