The database I patched together from the large number of rock shelters that exist in the Pilbara region, Western Australia, has a great deal of potential for future research on these sites. This is because the database contains nearly 600 sites and potential archaeological deposits (PADs) with quite a bit of information available for each. There are always going to be limitations, though, because the database does not include every rock shelter and PAD in the Pilbara.
If nothing else, Australian Cultural Heritage Management (ACHM) could use the database I created and expand on it with more data fields and rock shelters and create their own large scale database. Some data fields that the archaeologists at ACHM said they would like in a database are information on which mining tenement each rock shelter falls within and if there are any hazards or difficulties in physically accessing the rock shelters. This would assist the work with rock shelters in the Pilbara dramatically, because archaeologists would have additional information easily available during their research. Hopefully the database I have created will have features added to it and its data expanded by future researchers so it can reach its full potential for assisting rock shelter and PAD research in the Pilbara.
I have found some interesting results now that I have come to the end of the data analysis in regards to rock shelter sites and PADs in the Pilbara. The aspect (direction rock shelter opening faces) of rock shelter sites and PADs was most often east, at 22.8% and 17.4% respectively. Rock shelter sites and PADs are also within 100m2 in area at 82.4% and 89.5% respectively. The aspect and area analyses of rock shelter sites and PADs helps to indicate how common rock shelters with these features and physical dimensions are in the Pilbara.
Blog post 3
I have finished the data analysis of the database comprised of many of the unexcavated and excavated rock shelters, and potential archaeological deposits located in the Pilbara region, Western Australia. This information should be very handy for future archaeological work in the area. One interesting pattern I found is the direction many of the rock shelters are facing. In the data that I have looked at, most of them are facing in between east and south. Bednarik (1977 p 55) has performed a considerable amount of work in rock shelters relating to rock art and takes a considerable interest in the ability of humans to build representations of reality. He found some results in the research performed in the Pilbara which are similar to the ones I located during data entry. He found that most of the rock shelters faced east and north-east. The results I have had are loosely similar to Bednarik’s, indicating some level of consistency across the greater part of the Pilbara .
Another data set I analysed was the metre-squared area of rock shelter sites against the numbers of stone artefacts. From this I found that most of the rock shelters were less than 60m2 and contained between one and four artefacts on the surface. I found that, for the better part, larger rock shelters had very few artefacts. This indicates that smaller rock shelters which contained a low number of artefacts are much more common than large rock shelters with numerous artefacts.
A single bailer shell was found in one of the rock shelters, which is very rare in the Pilbara (Lynley Wallis 2012. pers comm.). This site could be interesting to research in the future because of its uniqueness in the Pilbara.
Bednarik, R.G. 1977 ‘A Survey of Prehistoric Sites in the Tom Price region, North Western Australia. Archaeology and Physical Anthropology in Oceania. 12:51-80.
Bednarik, R.G. n.d Robert G. Bednarik. Retrieved 12/6/12 from http://home.vicnet.net.au/~auranet/auraweb/web/index.html
Blog post 2
I have started entering rockshelter data into the Excel spreadsheet and have found that the database is not going to be as useful I hoped it would be. This is because not all of the texts that I have found contain the data needed. The information on rock shelters in these texts varies greatly in quantity, which sometimes depends on the purpose of the project. In one report, three rock shelters were observed but not described, other than their rough locations and the fact that they exist. Rockshelter names were not mentioned in the report, which can be attributed to the fact that the survey was primarily concerned with other site types.
Other reports focus on the excavation of rock shelters but do not include all of the useful information, such as rockshelter dimensions. With information like this missing, it will be difficult for archaeologists to use the database to assist in rock shelter identification. I understood I would run into this problem at some point because of the number of reports and publications that had to, and will, be searched, but hopefully enough information will be given in the future to make the database useful.
Directed study in cultural heritage management
This semester, I’m undertaking a Directed Study in Cultural Heritage Management as part of my Graduate Diploma in Archaeology. My industry partner is Adelaide-based heritage firm Australian Cultural Heritage Management Pty Ltd (ACHM).
ACHM do a lot of work for the mining industry in the Pilbara region of Western Australia. One of the issues out there is that often it’s not easy to tell if a rockshelter has been used by Aboriginal people in the past. My job is to collate all the available information to see if there are some characteristics of rockshelters that will help us to work this out.
The key issue here is a site with Potential Archaeological Deposit, which usually means it displays features that are related to use by Aboriginal people, such as proximity to water sources. There may be no evidence of the use of the site by Aboriginal groups on the surface.
I am going to use information in previous reports to create a database of all these PADs and genuine rock shelters, recording such data as shelter dimensions, number and type of surface artefacts recovered and excavation details where applicable. Hopefully, when I analyse this data, we will be one step closer to working out how to distinguish between legitimate archaeological rock shelters and those that haven’t been used at all in the past. This information will help all future archaeological work in the Pilbara region.
Blog post 4
Looking at the dry sifting and floatation processes I have used on the midden material, they seem to have mostly served the purpose of cleaning the material. These processes cleaned the materials very well, because it is reasonably easy to identify faunal remains simply by looking at them. This means that I do not have to keep lightly wetting the materials to see if the water is absorbed, indicating porous bone.
However, the subtraction of vegetation and charcoal—the purpose of the floatation procedure—does not seem to have helped much in narrowing down the amount of material likely to contain faunal remains. This is because, although floatation has proven to be extremely effective at separating vegetation and charcoal, these materials do not seem to make up a great deal of the material in the lab as indicated in the photo:
Left: Light fraction. Right: Heavy fraction
It might have been more effective simply to have performed the deflocculation procedure on the midden materials, because this is designed specifically to clean them. Although I never tested this technique, I think it would have been more helpful in the project as compared to floatation. This is because the primary aim in the project is to identify faunal remains. Therefore, a technique which is designed to make materials more visible is more useful than one which separates out charcoal and vegetation.