Estimates of late Holocene soil production and erosion in the Snowy Mountains, Australia
Introduction
On a global scale, alpine landscapes are recognised as regions of relatively high geomorphic activity (Dedkov and Moszherin, 1992, Milliman and Syvitski, 1992). This has been ascribed to their climate, which is typically cold and wet, tectonic activity and corresponding high elevations and steep slopes (high potential energy), which in combination promote rapid physical weathering, erosion and sediment transport (Milliman and Syvitski, 1992, Syvitski and Milliman, 2007, Vanmaercke et al., 2011, Walling and Webb, 1996). For example, rivers draining mountain basins transport a disproportionately large proportion of the global sediment yield, that is, 870 t/km2/y compared to 115 t/km2/y for the rest of the World's rivers (Milliman and Farnsworth, 2011). Especially in tectonically active mountain ranges, erosion can be so rapid as to equal or exceed the rate of uplift (Brozović et al., 1997, Koppes and Montgomery, 2009, Mitchell and Montgomery, 2006).
Despite high rates of geomorphic activity, the historical perception was that in cold mountain environments, rates of chemical weathering and, therefore, soil formation were low (e.g. Peltier, 1950). Contrary to this assumption, rapid rates of soil production have been measured in alpine environments, especially in regions experiencing rapid uplift and high precipitation. For example, in the Southern Alps New Zealand, where rainfall may exceed 10 m/y and uplift approximates 10 mm/y, soil production rates may reach 2.5 mm/y (Larsen et al., 2014), an order of magnitude higher than soil production rates measured elsewhere (Larsen et al., 2014). The potential significance of chemical weathering in mountain environments is further evidenced by the existence of extensive soil mantles in a variety of alpine settings worldwide (Dixon and Thorn, 2005, Egli et al., 2014, Norton and von Blanckenburg, 2010, Riebe et al., 2004).
Soils in high mountain environments have been shown to be sensitive to changes in climate and human activity, which alter soil production processes and may accelerate the erosion rates by multiple orders of magnitude (Barsch and Caine, 1984). Episodic changes in climate, vegetation cover, fire frequency and human disturbance are therefore likely to be important controls on the balance between soil production and erosion in some mountain environments (Hewawasam et al., 2003, Kirchner et al., 2001, Koppes and Montgomery, 2009, Schmidt et al., 2002).
In tectonically stable, non-glaciated, high to moderate rainfall, i.e. less than 2 m/y, alpine areas, such as the Snowy Mountains in south-eastern Australia, rates of soil production and erosion have received less attention. The Snowy Mountains have traditionally been viewed as distinct from other alpine regions (Costin, 1989, Kirkpatrick, 1994). This is due to their intra-plate setting and resulting tectonic stability and moderate relief (slopes) (Bishop and Goldrick, 2000). In addition, they experienced relatively limited Pleistocene glaciation (Barrows et al., 2001). These characteristics have facilitated the development of a relatively thick soil mantle (0.6 to < 1 m) over almost the entire alpine area (Costin, 1989). Nevertheless, the Snowy Mountains are considered to have experienced pulses of intensified sediment transport in response to changing climate during the late Quaternary (Costin, 1972, Kemp and Rhodes, 2010, Ogden et al., 2001, Page et al., 2009) and as a result of livestock grazing between the mid-1800s and 1940s CE (Costin et al., 1960).
The quantification of inherently spatially and temporally variable soil production and erosion rates remains a major challenge within geomorphology. Measurement of hillslope erosion has been undertaken by a variety of methods that can be broadly categorised into: plot and survey approaches (e.g. Costin et al., 1960); measurement of sediment yields by either stream gauging, or by measurement of the mass of sediment accumulated in geomorphic sinks, such as lakes (e.g. Neil, 1991, Tomkins et al., 2007); erosion tracer methods using fallout radionuclides (e.g. 137Cs and 210Pb) (Blake et al., 2009, Loughran et al., 1988, Porto et al., 2009, Ritchie and McHenry, 1990, Walling et al., 2003) and; the application of cosmogenic nuclides (e.g. Dixon and Riebe, 2014, Heimsath et al., 2002). These methods are each limited by the challenges of upscaling point measurements in time and space in relation to the representativeness of reference sites, the spatial heterogeneity of tracer fallout and transport and the issues of sediment storage and delivery (Chappell et al., 2011b, de Vente et al., 2007, Zhang et al., 2015). In addition, these methods provide data over different time periods, e.g. stream gauging typically provides short term data (event to decadal scale), radionuclides provide decadal to centennial scale data, and commonly used cosmogenic nuclides integrate over millennial scales. As a result, different approaches will commonly yield very different results (e.g. Tomkins et al., 2007, Wasson et al., 1996) that are then subject to various interpretations.
The objective of this study is to quantify soil development and erosion rates in a tectonically stable, currently non-glaciated mountain environment and to advance the understanding of the relative controls that changing climate and anthropogenic activities place on landscape stability and sediment budgets. In doing so the likely age of these soils is discussed, which in this setting is likely to be constrained by glaciation and/or periglacial processes to at least < 11–16 ka (Barrows et al., 2001, Costin, 1972). This study employs multiple methods to attempt to quantify rates of soil development, hillslope erosion and sediment transport. Hillslope erosion rates are investigated using fallout radionuclides (137Cs and 210Pbex) and by calculating sediment mass accumulation rates in alpine lakes and reservoirs. Soil development rates are examined using geomorphic and paleoclimate evidence combined with radiocarbon analysis. These approaches overlap in time, allowing the balance between soil development and erosion rates to be investigated. Results are placed within the context of past climate variability and athropogenic impact in the Snowy Mountains.
Section snippets
Regional setting
The Snowy Mountains are a high' elevation plateau of moderate (undulating) relief. Despite being the highest region of Australia, they reach an elevation of only 2228 m at their highest point (Mt Kosciusko) and local relief of the alpine area is usually < 200 m. The Snowy Mountains are the erosional remnants of uplift associated with the Cretacous breakup of Gondawana, beginning at 100 ma with most intense tectonic activity centered around 55 ma (Bishop and Goldrick, 2000). Their intraplate setting
Study locations
Estimates of rates of soil development, hillslope erosion and sediment yield were undertaken within the catchment of Guthega Reservoir, located in the headwaters of the Snowy River (Fig. 1). Guthega catchment includes the highest peaks of the Main Divide. Sixty-five percent of the catchment lies within the alpine zone, which is characterised by both high relative precipitation and high runoff coefficients (Reinfelds et al., 2014). Much of this area is comprised of high elevation plateaux
Radiocarbon ages
For the Guthega soil catena, returned 14C AMS ages for the A/BC horizon transition show good agreement between the three profiles with ages for the ridge-crest ranging from 2330–2430 y cal. BP; 2520–2750 y cal. BP for the mid-slope and, 2150–2310 y cal. BP in the toe-slope profile (Fig. 2 and Table 2). Minimum age differences are as small as 30 years between the ridge-crest and toe-slope and do not display a consistent downslope relationship. The maximum radiocarbon age of the organic A horizon is,
The age of alpine and subalpine soils in the Snowy Mountains and its implications for sediment production
It is generally accepted that late Pleistocene sediment erosion and transport rates in the highlands of Australia were greater than present (Kemp and Rhodes, 2010, Page and Nanson, 1996, Page et al., 1991). This is demonstrated in the major rivers which drain the Snowy Mountains (the Murray and Murrumbidgee Rivers), which are known to have experienced markedly different channel forms and sediment characteristics, i.e. braided, low sinuosity channels transporting bedload by comparison to the
Soil development, erosion and sedimentation – summary and implications
The results of this study imply that the Snowy Mountains experience both rapid soil development rates and slow erosion rates by comparison to lowland sites. Maximum and minimum net soil development rates estimated by this study (20–220 t/km2/y) exceed the maximum and minimum estimates of the net soil loss which has occurred over the past 100 years (10–90 t/km2/y) (Fig. 5). This is consistent with the occurrence of widespread shallow alpine soils in the Snowy Mountains.
The 210Pbex erosion rates
Acknowledgments
This research was funded by Snowy Hydro Ltd. and by an Australian Institute of Nuclear Science and Engineering (AINSE) Post Graduate Research Award (PGRA No. 10085). We thank the staff of the Environmental Radioactivity Measurement Centre (ANSTO), in particular, Daniela Fierro for performing the radionuclide measurements and the many people at Snowy Hydro Ltd. who provided field work and logistics support. We also thank Bob Wasson and Ken Ferrier whose thoughtful comments improved this
References (139)
- et al.
Exposure ages for Pleistocene periglacial deposits in Australia
Quat. Sci. Rev.
(2004) - et al.
Late Pleistocene glaciation of the Kosciuszko Massif, Snowy Mountains, Australia
Quat. Res.
(2001) - et al.
Deriving hillslope sediment budgets in wildfire-affected forests using fallout radionuclide tracers
Geomorphology
(2009) Carbon-14 dates from the Snowy Mountains area, southeastern Australia, and their interpretation
Quat. Res.
(1972)- et al.
Chemical weathering and landscape development in mid-latitude alpine environments
Geomorphology
(2005) - et al.
Chemical weathering response to tectonic forcing: a soils perspective from the San Gabriel Mountains, California
Earth Planet. Sci. Lett.
(2012) - et al.
Soil formation rates on silicate parent material in alpine environments: different approaches–different results?
Geoderma
(2014) - et al.
Correlation of erosion measurements and soil caesium-137 content
Int. J. Rad. Appl. Instrum. A Appl. Radiat. Isot.
(1990) - et al.
Radiocarbon dating of quaternary loess deposits, Banks Peninsula, Canterbury, New Zealand
Quat. Res.
(1977) - et al.
Cross-continent comparison of high-resolution Holocene climate records from southern Australia — deciphering the impacts of far-field teleconnections
Earth Sci. Rev.
(2013)