Australia: The Land Where Time Began

A biography of the Australian continent 

Keep River Region, Northwestern Australia, Comparison of Histories Inside and Outside Rockshelters

In this paper comparisons are made between the archaeological evidence of Aboriginal occupation between inside rockshelters and the sand sheets outside these shelters, focusing on 2 locations in the Keep River region, northwestern Australia. It was revealed by radiocarbon and luminescence dating that sequences inside rockshelters are generally younger (<10,000 BP) than outside the same rockshelters (<18,000 BP). It was also found that differences in the chronology of occupation and artefact assemblages inside and outside rockshelters result from depositional and postdepositional processes, as well as shifts in function. Ward et al. suggest that the late buildup of sediments within rockshelters, increased accumulation of artefacts, and reduced postdepositional disturbance in some settings may be accounted for by an increase in the regional rate of sedimentation from 10 cm/1,000 years to 20 cm/1,000 years in the Holocene. In the Late Holocene it is indicated by a change in hunting technology and greater rock art production that there was more intense use of the rockshelters. It is indicated by these results that some cultural interpretations might be flawed unless archaeological evidence from rock shelters and open-site excavations is integrated.

Rockshelters provide a fundamental source of archaeological evidence in many places in the world as they act to contain debris of occupation in a relatively limited area. It is often a problem to date the human use or occupation of such sites as a result of the complexity of the sedimentation (Farrand, 2001), including potential disturbance by humans (Stockton, 1973; Hughes & Lampert, 1973; Hughes & Lampert, 1977; Villa & Courtin, 1983; Theunissen et al., 1998: Walthall, 1998).

In Australia archaeological studies commonly focus of rockshelters, even though it has been inferred from ethnographic and archaeological data that open camp sites were occupied much more frequently than rockshelters (Smith & Sharp, 1993; Lourandos & David, 1998). The archaeological perspective that Aboriginal people to have lived under shelter in open sites has been suggested (Attenbrow, 2002, p. 105; 2004), arguing that around the coast in the Sydney Sandstone country they commonly lived in caves and rockshelters. With the exception of any differences in the use of the site, the preservation of cultural material inside and outside the rock shelters are not likely to be similar because they each have a distinct suite of sedimentary processes that control the nature of preservation (Farrand, 2001; Ward & Larcombe, 2003). Also, it is clear there has been a bias favouring the dating of archaeological deposits from rockshelters rather than from open deposits, as well as inadequate sampling of open site occupation that at present constrains the interpretations of the history of settlement in Australian archaeology (Ulm, 2004; Ward, 2004). There have not been many comparative studies of cultural deposits or sedimentary processes inside and well outside the dripline of rockshelters (e.g., Morwood, 1981; Jones & Johnson, 1985; Morwood et al., 1995; Boer-Mah, 2002) with the result that understanding of site formation and settlement history remains incomplete. There is a general assumption that in rockshelters and caves conditions prevail that favour  preservation and recovery of intact archaeological deposits than in open sites (Walthall, 1998; Ulm, 2004). Devil’s Lair, a limestone cave in southwestern Australia, has provided a history of about 19,000 years  (Dortch, 1986), compared with a deep sequence inside the cave that is now believed to span at least 43,000 years (Turney et al., 2001). It has been suggested that whether depositional sequences are longer inside or outside of cave or rockshelters may depend of cultural, sedimentological, and postdepositional processes (Farrand, 2001; Attenbrow, 2002, 2004; Ward, 2004; Ward et al., in press).

This case study of the Keep River’s lower catchment, northern Australia questions that rockshelters necessarily provide better conditions for preservation of longer human occupation records than in open sandy environments. It has been indicated by previous research in the Keep River region that there are major discrepancies between apparent age of some rockshelters and adjacent sand sheet deposits (cf. Fullagar et al., 1996; Roberts et al., 1998, 1999; Galbraith et al., 1999), and between subsurface archaeological sequences and the painted and engraved rock art (Watchman, 1999; Taҫon et al., 2003). These discrepancies highlight the need to identify the spatial and temporal scales of deposition in rockshelters and adjacent sand sheets, as well as to determine how rock art sequences are linked, if at all, with subsurface archaeological remains. In this paper Ward et al. have focused on 2 site complexes that are archaeologically rich, Karlinga and Goorurarmum,  comparing the records of deposition and disturbance inside of the rockshelter with the sand plain outside. In order to assess the implications for interpreting long-term changes in site function and the history of settlement Ward et al. attempted to distinguish cultural, sedimentological , and postdepositional processes.


A new framework within which the archaeological record can be interpreted has been provided by dating of sand sheet sediments and rockshelter sediments, though there are some discrepancies between TL, OSL and radiocarbon age determinations for the Keep River region. A record from the Late Pleistocene with a relatively abundant assemblage of artefacts, has been preserved in thick sand sheet deposits, whereas a Holocene record has been preserved in the rockshelter deposits, in which artefact assemblages are more abundant and varied only in the last millennium.

The presence in the sand sheets immediately outside the rockshelters dating to as early as about 20,000 BP indicates that rockshelters may have been used much longer than has been revealed by luminescence or radiocarbon dating of the shelter deposits themselves. There is an absence of deposits from the Late Pleistocene in the rockshelter sites in this region of the Keep River, which Ward et al. suggest may reflect sparse cultural deposition, though it may also reflect a geomorphological limitation for the accumulation of sediment.

From this study the overarching implication is that patterns and cultural interpretations may be fundamentally flawed if they are constructed predominantly by basing them on rockshelter deposits in similar sandy environments. Also, the environmental and climatic limitations imposed on archaeological reconstructions in northern Australia mean that multidisciplinary studies of rockshelters, sand sheets, as well as other types of open sites, are fundamental to understanding cultural change.

Sources & Further reading

  1. Ward, I. A. K., R. L. K. Fullagar, T. Boer-Mah, L. M. Head, P. S. C. Taçon and K. Mulvaney (2006). "Comparison of sedimentation and occupation histories inside and outside rock shelters, Keep-River region, northwestern Australia." Geoarchaeology 21(1): 1-27.


Author: M. H. Monroe
Last updated: 29/06/2016
Journey Back Through Time
Experience Australia
Aboriginal Australia
National Parks
Photo Galleries
Site Map
                                                                                           Author: M.H.Monroe  Email:     Sources & Further reading