 Welcome everybody, I'm Julia Martin from the Australian Research Data Commons. Thank you for joining today's webinar where our guest, John Pring, from GeoScience Australia will present an overview of a wonderful data rescue project using non-traditional means. Yeah, yes, good morning. It's my name's John Pring, as was said. Firstly, I'd like to open with an acknowledgement of the country from where I'm broadcasting this webinar. I acknowledge the traditional custodians of the lands on which I work today, the Ngunnawal people, and pay my respects to the elders past and present. I would also like to pay my respects to all Aboriginal and Torres Strait Islanders present today. A bit of background on GeoScience Australia. It's a federal government agency made up of approximately 600 people. It's responsible for the provision of GeoScience information, services and capability to the Australian government, industry and the community. This presentation explores the rescue of one of our older collections from fading into the annals of history. Okay, so as not to be too radical, this presentation will follow the usual format. So I'm gonna tell you a bit about GA and so forth like that. But firstly, for the dramatic, the DRS-ABCD, courtesy of St. John's, if you've done first aid, this will be familiar, is a way I thought I'd describe how we went about the project. Firstly, dangers. We checked for risk factors that were not considered when the samples themselves were being taken. This would include radioactive specimens, asbestos and things like that. One of these risks that's come to light on a different project was the accumulation of dust on aerial photographs that have been sitting around for 50 or 60 years. In terms of response, they are, where we were reviving a collection that is still useful or just reviving it for sentimentality. Send for help, call for volunteers and call for management support for the project. The airways determine a minimum critical mass to provide the, to prove the viability of the approach and the continued need for the collection, i.e. the borrowings increase and breathing. Was the collection currently being used? E.g. were we getting requests for slides? The answer was just barely, mainly due to a lack of knowledge of the collection and increasing difficulty with gaining access to it. The CPR, well, obviously, applied the project and watched the signs of life returning, including increased awareness of the existence of the collection. One of the unwitting things that we observed it was that the measures of success was to increased borrowing, yes, but we observed that there was increased exposure of citizen science and awareness of citizen science. Unfortunately, we didn't need the defibrillator. In terms of geoscience itself, we, as I said, 600 people, our tagline is applying geoscience to Australia's most important challenges. Here's a sample of what we get up to. Mapping of floods, using satellite imagery, courtesy of the Digital Earth Australia. Note the change in greenness and the Murray-Darling River as it floods a couple of years ago. Management of Australia's involvement in positioning infrastructure, including a network of ground stations, currently 130, rising to 200 over the next couple of years. This will support high accuracy positioning down to plus or minus three centimetres in mobile coverage areas, and has been seen to be useful to things like automated autonomous vehicles, precision agriculture, and emergency management, not to mention your phone. We're also responsible for monitoring of earthquakes and the provisioning of advice on earthquakes and tsunami warning, working with the states and territories in terms of groundwater and the impact of various activities on that. Mapping and mapping related products, in this particular case, of iron collections, sorry, iron deposits. Mapping and documenting geological time scales. Analyzing samples and support of research and industry. This particular instrument is called the shrimp, the sensitive high-resolution iron micro-probe, which uses zircons and zircon crystals for the dating of rocks. We're also the curator of a number of collections, both physical, cores, cuttings, minerals, fossils, and electronic. An example of this is the Landsat Archive, so it's recently been made available. We've got bathymetry, which includes supporting the definition of maritime boundaries and supporting the search for Malaysian airplane MH370. The collection and analysis of data in support of extractive industries in the form of pre-competitive data, in this case, seismic surveys. And we also have a very active education center with in excess of 10,000 people who have access to the data. A very active education center with in excess of 10,000 people coming through mainly school students. Okay, getting started. What's the thin section, you might ask? Well, very briefly, a thin section is a very thin sliver of rock, approximately 30 microns thick, thin enough that in most cases it's translucent. So basically, we take the rock sample and we make a microscope slide. I did say it was going to be brief. Looking at these slides under a microscope allows the geologist to interpret things such as the structure and composition of the rock. This is the same section of rock under a plain polarised and cross-polarised light. I've used rock-thin sections as the example, but we also looked at other slide-based collections such as micro and nanofossils as part of the project. All up there is at least a quarter of a million rocks of thin sections and a similar quantity of other type slides. As you can see, our management system, so to speak, was not the best for today's electronic world. This was due to the historic nature of the collection process over the last almost 100 years. So the issues for the project included the transcription of the hard-copy data and metadata for the slides and the cataloging of and suitably housing of the actual slides themselves. With usage having dwindled largely due to the lack of knowledge of the collection and difficulty in accessing it, it very quickly became evident that I would struggle to undertake the rescue of this collection following traditional means purely based on the cost. That is, the business case for expanding the money that it would take would be very difficult to sell to management. Here are some examples of the types of transcriptions that were needed. The handwriting varies considerably. Okay, so a solution. When the traditional was going to work, we had to look at an alternate means. The solution to this issue was to transcribe the old records, was to use a platform called Digivol. This is a citizen science platform that is managed by the Australian Museum and was implemented by Atlas of Living Australia under increased or national collaborative research infrastructure strategy funding. Briefly, the initial registered user is presented with an image of the items to be transcribed and entry fields to capture the data. The image is transcribed by the first volunteer and validated by a second different one, appointed to the role. Consistency is assisted through the use of dropdown and predictive text fields. This data is then available to the project owner as a CSV file, which in our case, this was also reviewed by a qualified geologist to ensure data integrity. The volunteers involved are varied, but by and large tend to be more senior, but none the less keen. And I might add that Theresa, the lady in pink to the left, lives on the South Coast and has done some 480,000 transcriptions or validations over the last seven or eight years. In terms of results, the overall results indicated here what can be achieved through enthusiastic volunteers over a relatively short period of time, some nine to 12 months. Further work in cataloging the physical slides and data entry was done by local volunteers or on-site. The success of the project has led Geoscience Australia to continue progressing the transcription of these records on an ASICAN basis and also exploring how citizen science might be appropriately utilized in other activities. In recovering data, there are a myriad of data elements that need to be considered, but let's pick one. In this case, the spatial component. Spatial information is critical to the usefulness of a sample. A sample that does not know, sorry, a sample that we don't know where it came from is a very little scientific use for research. It's something akin to road base. Now, for a little history, the geologists in the past would have done estimates of where these things were picked up. They would base those estimates on distances from known things, homesteads and the like, on the names of mines. They would estimate grid references. One of the things they used was aerial photographs. Now, what they would do, they would take the aerial photograph, they would push holes with a pin through the aerial photograph where they identified or picked up the sample. They would then turn the photo over and write down the reference number on the back for these samples. This is a far cry from modern ways of doing things, the modern way of displaying this information. But we'll get to that in a moment. One of the other things that we had to deal with was the differing projections over the course of the time that we'd been collecting samples. This included projections that dated back to 1858, went through AGD 66 and came up to more modern standards. In terms of how you might display this currently, one of my GIS Fluent volunteers painstakingly took the information from the associated aerial photographs and put this together. This is a Google Earth product. The project wouldn't have worked without the volunteers. What drives a volunteer? When working with volunteers, regardless of whether they are local or remote, it is imperative that you understand what makes them want to participate. After all, they are not being paid and there are no formal contracts for their work. From the volunteer perspective, their enthusiasm is indicated by virtue of the fact that I get emails from volunteers encouraging me to put more documents up to be transcribed, requests for background information on the subject matter, aerial photographs, background on geologists, et cetera, and often requests for how they might do more volunteer work with the agency. Volunteers tell me that they use the transcribing to inform their holiday plans, geotourism. One continuing this geotourism to continue to transcribe while they did this trip through Northern Australia, including up the Tannamide track. Volunteers indicate that they like doing the work because it gains some knowledge. It keeps them mentally active, a lot of these people are older and not engaged with the traditional workforce. It gives them a sense of satisfaction, a sense of completing something that's worthwhile. They discover new places. Some of these volunteers discovered places up in the Kimberleys and things like that, which they then went to visit. I'll call it friendly rivalry. There is certainly a rivalry and mostly it's friendly, but don't be mistaken, there is our competitive nature to some of this. Their personal statistics, as I said, Teresa is approaching 480,000 transcriptions and validations. There's a sense of community. DigiVol provides a forum where the participants can actually exchange ideas and answer each other's questions. In terms of where we might go from here, the project has seen the physical cataloging of approximately 95,000 slides, the transcription of 40,000 items from registers and forms and the gradual inclusion of this information into an electronic tool. Currently, this is available internally via a dedicated tool, but the information is also available externally to Geoscience Australia via web services. This external access is provided via OGC Web Services as part of a more general samples service. Where the slides are linked to their parent samples and documentation, etc., via IGSNs. In the slightly longer term, this is a display of where the samples were or the slides were, I think it's about six months ago now. Slightly longer term, I would want to see if we can make the images of the slides available externally to the agency. Ideally, I would hope to provide access to the images in a similar manner to the British Geological Survey. The BGS informs me that as well as assisting people in identifying the slides to request this information and images and the request of the slides be sent to them, because like the BGS, Geoscience Australia will ship the slides anywhere around the world. The images are also being used directly for research. One of the things about being able to access these images would mean that it would reduce the risk to GA of the slides being broken in transit or not returned. As happens when looking through long-forgotten rooms, we, in doing the project, came across some potentially historic, significant finds. Slides made from the residue of someone you might be familiar with, Sir Douglas Mawson, and his Antarctic expedition, 1911, 1914. The slides were made in the 70s. We'd kept the residues for quite some time. The next group, these slides were made as part of the early investigations of the Snowy Scheme, circa 1950. Also, significant to Geoscience Australia and Australia, but better known in Geoscience Australia, there are some significant people involved. I would like to acknowledge all the following people in terms of their participation, particularly Leslie Wyborn, who's been a champion of all this work. This presentation is based on a paper. It's based on the project, but the project was written up in a paper that's also detailed here. Thank you. Thank you very much, John. That was absolutely fantastic. I suspect that what you've talked about will probably prompt quite a few people to consider their own data rescue problems. We know at the moment that data rescue is a significant issue for many people, so the approach and the method that you took could probably be replicated. I encourage people to read that paper that John put in. We could potentially send it around when we send the link to the webinar, because it gives a really good overview of keeping those volunteers enthusiastic. Of an interesting sideline, given the current coronavirus issues, some of the lab staff from Geoscience Australia will be doing things on DigiVol as part of their work-from-home process, because you can't take chemicals and so forth home to do lab work. As part of their stuff, they'll be doing some of the online DigiVol stuff. So we're putting up GA-related projects for them. Here's a question for you, John. How did you actually recruit the volunteers? When we worked out that we wanted to do it via citizen science, we looked around for an existing platform. Rather than trying to produce our own and then going out and marketing for people to come and use it, we found the DigiVol platform. It came with an existing cohort of volunteers. We put up the expeditions. I'm not a marketing person, I'm an electronic engineer by background, but I made sure that the images that were used to highlight the project were bright, were successful, showed things that were interesting, not just a black piece of coal or something of that nature. So they were bright and caught people's eyes. What I believe and from feedback from the volunteers, what I believe also helped was that we were responsive to questions. There were questions around what we wanted volunteers to do with various items. We were also questions around, as I said, background on things like the geologists that picked up the samples. So we were responsive. We'd try and respond within a day by getting the question via the forum to provide answers, and that would mean I'd reach back into Geoscience Australia to some of our volunteers, one of whom joined the agency where it's predecessor in 1961. So it was there for the getting of some of those samples. Thank you. Thank you very much. If there are no other questions, I would like to thank you very much, John, for your time today and everybody else who joined us. Until next time.