Hello everyone, good afternoon. Welcome to the last session before the closing session. We will have presentations on use cases, different use cases. We will start by Elena

in the field with the use of open source technologies for Antarctic logistics, so I hope you enjoy it. The floor is yours, thank you. Hi everyone, thanks for dragging yourself away from coffee. I kind of wish I had one now.

Hi, I'm Elena. I work for the British Antarctic Survey and this is a scientific research organisation whose aim is to deliver world class polar research.

To do that it requires an awful lot of geospatial information and generally quite a lot of work from everybody. Our main hub is in the UK, so it's in Cambridge and we have research stations in both the Antarctic and the Arctic. We work together, both research staff and science and support staff to come together and support really good science in both poles.

So I'm part of the mapping team, so it's actually the mapping and geographic information centre, or MAGIC for short, which is the best acronym I've ever found.

And my main role is actually to provide geospatial support to operations, so that's what I'm going to be talking to you today. So, everyone knows Antarctica, but what we do is we operate three research stations within the 60 south latitude, so that's Rothera, Halley and Signe, so the red dots you can see there.

And then the two red dots at the top of your screen are maintained by us on behalf of the south Georgia government. And their research is quite different, it's based on more biological aspects, but what I'm going to focus on today is actually the geospatial support we provide for people on the continent.

So, the three stations that we provide there are Rothera there on the left, and that's our main hub, so it's where all of the science kind of starts. It has the air bridge from southern Chile, and the Falklands, and it's where everybody kind of starts, either when they're

passing through, or where they're based and they take short flights and trips out, so it's a quite busy airport actually. The top right there, which some of you may recognise from the news, is called Halley. It's built on an ice shelf, and it has a modular structure, and it's actually built on stilts with skis,

and the design is actually so it can be moved, and that may have been where you heard about it. Last year it was moved 23 kilometres, dragged across the ice shelf. The last one there is called Signe. You can see it's quite green, so it's quite a different

atmosphere to the other bases, and it actually is only accessed by ship, so it's potentially even more remote. So, to give you an idea of context, a bit stretched, but we'll get there. This is actually the raise that a Twin Otter can travel

along, so a Twin Otter is our standard aircraft that is used throughout the continent, and the crosses you can see are actually either stations or depots, so places where one of these aircraft can land, and the routes between them are actually just areas where they can make one hop at a time. So you can see the amount of different hops you'd have to get to get from, rather at the

top left there, all the way down to McMurdo at the bottom right, which is actually a US station. And if you superimpose this route network onto Europe, you can see the scales we're operating at, and this is exactly why geospatial data is so important for this area.

So, to support this, we have a couple of ships and one in construction, so these two on the right are the ones currently being used, so it's the James Clark Ross and the Ernest Shackleton. These are due to be replaced next year, hopefully, by the Sir David Attenborough, also known as Boaty McBoatface, if anyone recognizes that one.

So yeah, it's just been announced that we've finally got the top section installed, and so we're well on the way to getting a ship this time next year. I mentioned the Twin Otter aircraft, we've got four of these, the British Antarctic Survey has four on the right, and they're

used for deep field deployment most often, so they're quite small, there's only a pilot and a co-pilot, which actually is anybody. So, I know I've been one, so you can't really, don't trust me flying. So they do very short hops, and they're the ones that are deployed into the very deep field.

In contrast, you've got the Dash 7 there, which is much larger, but actually it can only carry 18 passengers, so actually pretty small. And that's the one that does the hops, the hops or the leaps between South America, the Falklands, and South, as we call Antarctica.

So, as you can see, this is quite a small operation, but from the range of stuff we do, it's actually pretty significant. So, we also have backup in the form of tractors, so these are piston bullies, technically, so these are the red machines you can see there.

And they're capable of carrying huge amounts of cargo, so upwards of 50 tons each. They're used for really large research traverses, so they travel upwards of about 4,000 kilometers per season. And they only travel at 15 kilometers an hour, so you can imagine just how many days that requires.

But you can see, this is a really good image actually, it shows how much they can carry. So, I mean, there's enough, that container there is actually one of the massive containers you see on the ships. And it's carrying one of those and plus six fuel bladders, which carry liters and liters and liters of fuel, which is actually the lifeblood of our organization, unfortunately.

So, these are deployed from ships, so the ships dock near an ice shelf usually. They deploy all of the cargo and then it's dragged across ice shelves to support large projects or just to resupply.

So, but we don't all have nice ships and aircraft to take us around. More often than not, you find two people in a tent, and that is still the case as it was 100 years ago. So, this is the traditional pyramid tent on the left here, and that sleeps two. So, you often have a field guide and a scientist, and they work together for upwards of about 120 days.

And, you know, that's a bit cozy, but they also have skidoos there where they can do overhand travel and short regional, like local trips. So, it's quite a varied range of scales that we work at as well.

And finally, we have large static camps. So, these are camps where multi-year surveys are carried out, such as ice core drilling or monitoring of ice shelves. And you can see the scale is huge. There's a piston bully on the left over there, and you can see the tents.

And these are there for months at a time. So, after that little whirlwind tour of everything, how do we support all of this? Quite well, considering. There's only nine of us in the mapping team, and we're responsible for all the mapping products and support for the British Antarctic Survey.

So, we're kept very busy. We all have our individual tasks and individual kind of niches, and mine is the operations support. So, you can read all of this, but the main asset that BAS has is that it's got both science and logistics under

the same roof, which means that you can have meetings with the principal investigator, and you can really plan what they're going to do. And it takes so much less time than if it was an email, phone calls, etc. So, that's the support.

And to do all of this, we have used open source technologies. These are a few of the ones we use. I'm sure you're all familiar with them. I won't go into too much. The main reasons for this is actually the flexibility that open source allows, and also for licensing issues, etc. Given people are there without, with limited connection to anything, for months at a

time, having licenses and dependency on anything like that is just not a good idea. So, we've developed really good tools, and this is something I'm going to run through today. So, the first one we use is QGIS. So, this is actually, I think, quite an

older, a bit older version, but I just snagged this off one of the field guide's laptops. So, this is just, as you can tell, a screenshot of QGIS, and it's actually the field operation manager's desktop device. So, he has all of the information that they need to plan where people are going in the field.

They talk to the field guides and the scientists when they're actually at Rothera, at the station. And from there, they make a plan of where they're going to go, how they're going to support it, who does what. So, it's vital for what we do there. This information is all stored in databases by us, and so we have a link from a database on the station to the database in Cambridge.

And this is one of the products. So, the general flow is pretty obvious, but you go from QGIS and you add data through, in PostGIS, usually either through a plug-in DB manager in QGIS, or directly using PG admin.

It's then published through GeoServer, styled, and then published to this web map, which is actually developed in-house. Yeah, the reasoning behind this, as I've mentioned to a few people here, is that this projection is not very popular among most people.

So, this is the Antarctic polar stereographic projection, generally quite a painful projection to customize or include in much. So, this is why we've kind of gone our own route.

But it has allowed us to develop quite a good way of personalizing things and making bespoke maps, but with this very familiar user interface, look and look and feel. So, this is actually just the front end of the Antarctic digital database, which I think you may have seen in yesterday's GeoServer talk, briefly.

We maintain this on behalf of the scientific community at large, and so there's links at the end if you want to go and have a mess around. But this is where the global repository of Antarctic topographic data is held. It's held in a web map, and you can have a look at what you want, zoom in and out, and then go to a download repository as well.

And the little buttons you can see at the top are actually all essentially plug-ins that we've developed. And it gives the user, you can just pick and match, mix and match all the drop-ins, and then pick what you want to use for each web map.

This is a bit blurry, but you can kind of see what we mean. So, all of this has been styled in GeoServer using just simple SLDs. I'm quite glad that works. But you can see how this is actually South Georgia here. You can see how as you zoom in, it gets more complex and labels appear and disappear.

So, it gives it a really nice look and feel and makes the user quite enjoy using it, I hope. So, behind the scenes, this is what we in the mapping team can see. And this is just our interface to creating individual custom maps.

It's quite an easy interface, so it's quite blurry there, but you can see it's a traditional GIS layer panel style. And then each layer can be imported from different feeds. The good thing about this is that because all of our data is fed by WMS, we're

actually just creating about 20% of the data for a new map is actually brand new. The rest of it is actually sourced from previously existing feeds. And so, this means that we can prevent data getting out of data, out of sync due to forking and duplication.

So, it's really useful. It's also quite a good interface for people like the majority of our team who aren't programmers and are not very familiar with it at all, myself included. So, this gives us all an option to create and build our own web maps. And this is one we made earlier.

So, this is the operations GIS, which is another bespoke map which was created a few years ago now as a planning and visualization tool for operations. It allows the users, there's a huge amount of data on the left, and it allows the users to pick and choose which layers they want to show, and there's more detailed information when you zoom in.

So, I've just picked a couple of layers at random here, but here you can see there's actually, this is the point of no return for a flight of a Twin Otter aircraft from each of the areas we know they have decent runways from. And then the lines there are actually tracked to traverse routes as well.

So, this is all kind of static data we have on the map. We actually have incorporated quite a bit of live data as well, and these are sourced from API feeds. So, the top bar, you can see there's a little aircraft symbol and a ship symbol. You can kind of see that. They're just like toggles, so you can see if there's any aircraft flying, and it's got a traditional red if it stays still, green if it's moving.

I haven't been able to get a screenshot of it just yet because it's currently winter there and there's no aircraft flying. I know. But a recent addition is the live tracking of the piston bullies, so the tractor traverse.

So, you can see this is actually them at work pulling the station that I said earlier. So, what it did is just created a short script to talk to the geotab API, extract the data, and then add it to a postage table. And this is fed through GeoServer and then plotted on a map.

You can see there B2 and B3 currently parked in our station. This is something we're developing now. So, this live is a bit questionable here, but this is the most up-to-date location of all of the field guides when they're moving.

So, every 12 hours or so, they're required to call in to the communications tower in Rothera and check in and say where they are and if everything's going okay. This information has been added to a web form, stored in postage, and then published on one of our web maps. And you can see we've actually got pop-outs available as well where you can see the name of the team and everything like that.

We also support quite large scientific projects. So, you may have heard of the Thwaites Glacier Initiative. It just started earlier this year. It's about a $40 million project spanning four years and involving hundreds of scientists who all want to get onto the ice shelf.

And Prophet, Dominoes, Charsan, they're all names of actual research groups, so it's a huge project. And this is a multinational project between the UK and the US, who are the primary partners, and there's a few others as well.

But what they want is this, oh you can't tell, there's the Thwaites Glacier, so it's actually enormous. What they wanted, and they came to us for, was a bespoke map which allowed both the people in the UK and those in the US to look at the same data at the same time. And so what better way to do this than a web map? And you can see this has kind of been swiveled a bit to focus on the area, and there's a lot more specifically operations data in here.

We're actually looking at packaging this up into a virtual box, and then this is going to be deployed to laptops in the field as well. All the way over there, and there's definitely no internet connection.

This has different base maps involved as well, so that's just the basic Landsat imagery mosaic of Antarctica. And there's more and more information as you zoom in. So I think that's pretty much it. Thank you for listening. Here's some contacts, there's some links there to the maps that I've shown.

And you can find some more information on some other web maps that we provide over there. I'll leave that for a couple of minutes. Thank you. Thank you very much. Are there any questions?

Thank you for keeping the time, because we don't want to be late for the final closing session. Are there any questions on the audience? Yes, one? Quite interesting by the way.

Thank you. Thank you very much, it was a very interesting presentation. I work for Land Information New Zealand, and I'm just wondering if there is some collaboration between us in terms of data sharing or tool sharing or whatever. Yeah, this is something that we're actually very interested in starting.

Our Director of Operations has been over to New Zealand for these discussions as well. We've created a number of the web maps which are publicly available anyway. And the Thwaites project is kind of the start of the collaboration for web mapping particularly.

One of the things we're really interested in doing is tool sharing and actually just open sourcing all of this in the future. I can't guarantee when, I'm not the developer so I cannot speak for them. But yeah, we're really interested in collaborating further with all the Antarctic programs. Thank you very much and looking forward for that.

Any additional questions? One more? I think on one of the slides with the QK's desktop there was a north arrow. How do you decide which way is north? Yeah, that's a good point.

Well, everywhere is north from the South Pole for starters. Yeah, we generally just go with magnetic north and this actually is the polar stereographic projection. And so its north is technically just directly like the North Pole as we know it.

But yeah, you could rotate this anyway and it would stay the same. Yeah, it's kind of meaningless having the north arrow there. You have to start using maybe a circle and then north is everywhere. Exactly, yeah, anywhere.

Additional questions? Yeah. One more? How do we manage offline editing? If you do editing here where you are now or there during operation and so on? Yeah, this is a subject we've had a lot of trouble with in general.

And the way we've managed it so far is by not managing it. In that any offline editing is then sent when it is online to the people in the team and then we are the ones responsible for making any changes.

I was there last year and that was partly to update all the web maps in the station and to make sure it's all up to date. And so the aim is to any changes that are made that are important then they get synchronised. They get sent over to the station or back to our headquarters in Cambridge.

Usually either over email and then added manually or we are working on having common layers that are synchronised. But this is something that we are trying to kind of iron out the details on at the moment. Does that answer your question? It's kind of a supervised editing sort of.

One more question? I have a question about the live tracking. Because I didn't fully understand. Is it really live or from time to time you send back the information to the server and then you update the web service?

Yeah that's a good question actually. Live is very difficult to do in Antarctica. And because often the connections are through Iridium which is extremely expensive.

So we have had to kind of manage it and deal with the frequency of the pings depending on each unit. So the air unit they have a ping that's every minute and that's updated on the web maps every minute. Because they travel at much greater speeds so we need to know. However the field guides and the overland travel they move much slowly so every 15 minutes to every 6 hours.

It really does depend on the users. Alright. Okay thank you very much. Then we move to the next speaker. It's about crowdsourcing the georeferencing of historical pictures.

And who is going to talk? Nicola. Thank you very much.