Patterns of variation in Australian alpine soils and their relationships to parent material, vegetation formation, climate and topography
Introduction
Jenny (1941) suggested that soils are a function of parent material, climate, topography, vegetation and time. Fire regimes can also have a strong influence on soil characteristics (di Folco and Kirkpatrick, 2011, McIntosh et al., 2005). Most of the alpine soils of the world have had only ten millennia in climatic conditions comparable to the present, with most loose material previously eroded by glacial and periglacial processes. This youth means that, in general, alpine soils are shallow, coarsely textured and have poorly developed horizons (Retzer, 1974). The characteristics of alpine soils in many different places have been shown to be strongly influenced by varying combinations of parent material, topography, biota and climate (e.g. Archer and Cutler, 1983, Baruch, 1984, Bäumler and Zech, 1994, Belsky and Del Moral, 1982, Diaz et al., 1997, Ives and Cutler, 1972, Messer, 1988, Molloy and Blakemore, 1974, Nolan and Robertson, 1987, Retzer, 1974).
In Scottish moorlands as a whole, Nolan and Robertson (1987) found that soil type was related to topographic position, through its influence on slope and soil drainage, within the over-riding influences of climate and geology. Ives and Cutler (1972) found that whilst soil characteristics were mainly determined by topographic position within a particular geological substrate, they were also influenced by vegetation type and altitude. Archer and Cutler (1983) highlighted the importance of aspect and elevation, surrogates for climatic conditions, in affecting the processes of soil formation. In the Mt Everest region of Nepal, Bäumler and Zech (1994) demonstrated a relationship between soil types and elevation. Some regional studies reinforce climate as the major influence in the formation of alpine soils, due to its effect on biological activity and chemical weathering (Dahlgren et al., 1997, Knapik et al., 1973, Retzer, 1974). Snow cover duration can dramatically affect soil properties in mountain landscapes (Dahlgren et al., 1997), through its effects on erosion rates and vegetation productivity.
In Australia, the alpine soils of the Snowy Mountains (Costin, 1954), Mount Sprent in southwestern Tasmania (Bridle and Kirkpatrick, 1997) and the organic alpine soils of Tasmania (di Folco, 2007) have been described in detail. The alpine soils of Australia are known to be highly diverse, including at least eight great soil groups (Bridle and Kirkpatrick, 1997, Costin, 1954, McKenzie et al., 2004).
There is known to be variation in Australian alpine soils related to variation in climate and substrate (Williams and Costin, 1994). Topography and vegetation type are also known to relate to edaphic variation, as in the case of the soils of fjaeldmark, short alpine herbfield and organic soils in Tasmania (Bridle and Kirkpatrick, 1997, Gibson and Kirkpatrick, 1985, Gibson and Kirkpatrick, 1992, Kirkpatrick and Dickinson, 1984, Kirkpatrick and Gibson, 1984, Lynch and Kirkpatrick, 1995).
The soil-forming factors are also known to interact. For example, the alpine soils of the Snowy Mountains illustrate an interaction between parent material and climate. The parent material is partly constituted of aeolian dust from the western plains (Walker and Costin, 1971), with Johnston (2001) estimating an influx of 1.8–113 kg ha− 1 per annum. The frequency of major dust storms, one in 10–20 years in the 20th century (Walker and Costin, 1971), has increased with four such storms occurring in the past decade (Green, 2012).
Whilst much is known of the alpine soils of Australia, there has been no systematic study of variation over their full range and no national scale quantitative testing of the relationships between the soil-forming factors and this variation. We use repeatable numeric methods to fill these gaps. Our specific aims were, first, to determine whether a numerical classification of soils based largely on chemical characters conforms to existing understanding of variation in Australian alpine soils. Second, we wished to determine which soil characteristics were related to surface geology, climate, vegetation and topographic position throughout the range of alpine environments in Australia. Within the iconic alpine area in the Snowy Mountains of New South Wales we tested whether a high level of aeolian dust deposition has overwhelmed the influence of the underlying rocks on soil characteristics.
Section snippets
Field data collection
Soil data were collected from 166 locations distributed throughout the full geographic range of Australian alpine and treeless subalpine vegetation, including the Snowy Mountains (Fig. 1), as part of a study focused on describing and explaining variation in the flora and vegetation (Kirkpatrick and Bridle, 1999). At each location data were collected from a 5 × 5 m area. Soil depth was measured using a metal probe. Depths were obtained from the centre of the quadrat and halfway from the centre to
Soil groups
The pattern of change of error values with agglomeration suggested that there were 11 soil groups. The characteristics and environments of each of these groups are summarised in Table 1, Table 2, Table 3 and briefly described below. Groups 1–8 occurred in both mainland Australia and Tasmania, whilst groups 9–11 were confined to Tasmania (Fig. 2). The terms ‘low’ and ‘high’ are used in relation to the data sets reported in the present paper.
Group 1 was best characterised by soil shallowness,
Conclusions
Our soil classification, based largely on chemical characteristics, proved largely consistent with the Australia-wide classification of Isbell (2002) and that of Costin (1954) for the region including the Snowy Mountains. Our classification suggested some ways to partition the orders of Isbell (2002) without relying on chemical characters. Our classification also demonstrated the value of repeatable numeric classification. The strong role of surface geology in influencing the characteristics of
Acknowledgments
This research was supported by an Australian Research Council grant to J.B. Kirkpatrick. We would like to thank the conservation agencies of NSW, ACT, Victoria and Tasmania for the provision of collecting permits and access to the parks. Allison Laboratories in Hobart carried out the soil analyses for the Australian wide samples whilst the soil samples from the Snowy Mountains were analysed by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) Division of Soil and Water.
References (62)
- et al.
Soils of the high mountain region of Eastern Nepal: classification, distribution and soil forming processes
Catena
(1994) - et al.
Generalized linear mixed models: a practical guide for ecology and evolution
Trends Ecol. Evol.
(2009) - et al.
Soil development along an elevational transect in the western Sierra Nevada, California
Geoderma
(1997) - et al.
Topographic variation in burning-induced loss of carbon from organic soils in Tasmanian moorlands
Catena
(2011) - et al.
Effect of north and south exposure on weathering rates and clay mineral formation in alpine soils
Catena
(2006) - et al.
The role of fire and nutrient loss in the genesis of the forest soils of Tasmania and southern New Zealand
For. Ecol. Manag.
(2005) - et al.
Soil catena variation along an alpine climatic transect, northern Peruvian Andes
Geoderma
(1992) - et al.
Post-fire hillslope erosion response in a sub-alpine environment, south-eastern Australia
Catena
(2008) - et al.
Pedogenesis and vegetation trends in the alpine and upper subalpine zones of the northeast Ben Ohau Range, New Zealand. 1. Site description, soil classification, and pedogenesis
N. Z. J. Sci.
(1983) - et al.
Dynamics of subalpine vegetation at Echo Flat, Lake Mountain, Victoria
Proc. Ecol. Soc. Aust.
(1983)