Webinar on Standards for Use of Aerial Sources for Heritage

Speakers: Helen Winton, Matthew Oakey (referred to as ‘Matt’), Matt

Helen: Well, as I said, greetings from sunny Swindon and Matthew in York. I'll just introduce ourselves, if you haven't met us before. Matt and I are the managers in the Aerial Investigation Team at Historic England. I have a national remit and Matt covers the North. We’ll mainly be talking you through slides today, but I'm hoping this will prompt some questions. So as Matt said, please do put these into the chat and we will have time at the end for a few questions as well. So our main aim for this webinar is to show that effective use of sources of aerial sources is important, and that presenting them in the form of a map is the best way to communicate the extent and form of archaeological landscapes seen on aerial sources. So if you just look at the map, some of you might recognize this. This is the Stonehenge landscape. Here is Stonehenge. We've got the avenue extending eastwards, we’ve got the cursors, but what this is a compilation of archaeological features seen on aerial sources, so we have things that survive in the landscape that you can still see. Like the cursors like the avenue, parts of the Avenue that is, but you've also got features such as this, which is a First World War aerodrome. The buildings associated with that and we've got field systems that have long since disappeared. So this is what we want to have you to take away from today: Maps are great, so OK, let's just move on. First of all, we've got a question quick and easy question for you, which is did you attend Webinar 1?

Matt: And don't forget everyone if you didn't attend it, you can find the recording for the first webinar on aerial photography on our website. Again, a link for which I will provide you with at the end of today's session. So that's pretty good. It's almost a 60-40 split.

Helen: That’s cool. Fantastic, it doesn't matter today if you didn't attend the first webinar. So there we go. Just to recap, for those of you who were there, there's a little taster in the hope that you will catch up on the website if you haven't seen it.

In the first webinar, we looked at context and uses of aerial sources, and we touched on some aspects of interpretation and Matt took us through 2 examples of our work, a rural example at Belsey in Northumberland, and an urban example in the city of Sunderland.

So even if you're very familiar with aerial photography and its uses, you might not be so familiar with the use in urban context, so I would recommend having a look and checking that out.

So what's this webinar about? We're going to briefly look at why standards are important. We'll look at a few technical aspects of looking at sources, and then I'm going to hand over to Matt who's going to take you through the whys and wherefores of the mapping process, how to get the most out of existing data, and then he's going to cover a few key things to think about, if you're involved in commissioning, or assessing tenders for work involving aerial sources.

OK, so moving on, so why do we need standards? Well, this this applies to any context for your work, whether it's a desk based assessment in a commercial context, maybe one of our projects at Historic England, or a project with the Community focus and also academic work as well. And it's really just so that everyone is clear about what you're hoping to achieve. Again, this is common sense. But also as a government body, Historic England are also keen to promote the work that results in data that has a legacy. So whether that's mapping records or reports arising out of aerial sources.

And we really encourage people to ensure that their data can be used in the long term, and we think that the best platform for this is the local authority historic environment record. Now I realise this may not be possible in some contexts, so there might be client confidentiality in a commercial project, for example, but it is worth asking the client if the data can eventually be in the public domain and Matt's going to come back and explore a few more of these key aspects later.

So this is just again stating the obvious in some ways and obviously most projects follow this process. You decide the project here, you agree things, but it's just surprising how many people don't actually document their methods and their use of aerial sources, and this comes up especially in the context of commercial desk based assessments. Perhaps the people doing the work are documenting it, but when the reports are compiled, often these get these get lost or they get missed. And I would just say again, this comes down to the legacy of the work.

It's very hard to judge the nature of a survey if you do not have the information about how and why it was carried out. So this is a bit of a plea to ensure that all this desk based work is properly scoped and documented.

And in Historic England, we have guidance on project management and we actually have a specific project planning note on using aerial sources for mapping. This is not available online at the moment, but you can contact me if you would like a copy of this. And as I'm sure a lot commercial companies use the Oasis system, which is on the archaeology data service and this is a way of logging and ensuring that the work that you're carrying out the details of it, who did it when they did it, and the report making sure that this is available to the local authority historic environment records. And this is being revamped at the moment, and we're really hoping that more people are going to use this system in the future, and we in Historic England are also adopting the use of this system to record our activities.

So what do our standards cover? These are the basics of our standards. We have minimum requirements for a raft of topics. As you can see, this covers a range of things from staff experience to the availability of the data through the local HER. Sorry I will keep going on about this. And please do contact us. So you can contact myself or Matt or we have a remote sensing at the Historic England website email address, as well, if you want to ask us any questions. You can see on your right, on your left sorry, my colleague Sally’s report has a historic of the techniques that we’ve used on older projects.

We’ve covered about half the country for our project and this document will help you to understand where the existing data comes from and how you might want to use that. So, I’m going to go through the scoping and sources briefly and then Matt is going to cover the remaining topics. So archaeological scope, this is pretty much standard, I would say, in the heritage or professional world.

You can look at as wide a range of archaeological scope as you can and in terms of aerial photography and aerial sources, this going to stretch from when people started building the major monuments in the Neolithic period, that are big enough to form archaeological crop marks or to leave earthwork traces behind. Right the way through to the twentieth century, our cut-off date used to be the Cold War, but since we’ve started working in urban landscapes, so we actually look at the 21st century in some cases and I think, even if we’re interested in one particular period, or one particular moment, or monument type, using aerial surveys, I think, to understand what you’re seeing, you really need to have an understanding of the broad range of things and just looking at everything will help you to understand why you are or aren’t seeing things from different periods. So, yes, a plea to have a look at a wide archaeological scope.

So, another of our basic standards is to use all the available sources you can. I’m not going to give you a comprehensive overview of this and of its use of sources. I’m hoping maybe, it will provide some thoughts and I’m going to cover some less familiar aspects of some aerial sources.

So a key consideration is deciding which sources are most appropriate for your study? So, for example, I'm not going to go into satellite data today as this still has limited use in the UK due to the abundance of other sources here, but also mainly the difficulty of getting hold of free high resolution data. And as archaeologists, certainly we're always looking. We need things to be freely available for us to be able to justify using them. So satellite has been used effectively elsewhere, usually where conventional sources are scarce and you can look on the Aerial Archaeology Research Group website for some examples of publications of this. We do have in our department a current project exploring uses of status through the Space for Smarter Government programme and this has provided satellite data to government bodies to explore its potential.

And you can read a little bit about this in the latest Historic England Research News. I think the key point of this slide is that our standard is to use as many sources as possible. Yes, so I think that's enough about that, so we'll just move on. So what are the key sources? So by far the greatest source for aerial interpretation and mapping are the hard copy prints held in national and local archives. Although we're in an increasingly digital world, the value of these hard copy resources should really not be underestimated. These are a mandatory source if you're doing a Historic England funded project and they should also be for desk based assessments as well. And I think, dealing with this the volume of sources that you look at is a surprisingly challenging area and this is where using experience staff comes into its own.

You need to be really organised to be able to view and identify, the relevant aspects of the landscape and archaeological remains on these collections of aerial photographs. And you're really, really missing a trick if you don't use a stereoscope to view the stereo pairs in 3D, so if you don't have lidar, you can do that virtually by using a handheld stereoscope. So another key consideration in project planning is organizing how you're going to access the archives and how you process the prints when you're there.

So be prepared, is what I would say, as there will be a lot to look at either the local archives of the National Archive in Swindon. You need to think about things like whether you need a camera to record the key photographs that you might want to work on back home. So all things to think about. And again, more information on that in our guidance.

OK, we’re going to move on just to talk about Lidar. We also run courses online or we did when the world was a bit more normal and we are looking at developing online materials. So do let us know if you'd like to know more about using Lidar etc. So even those of you who are familiar with Lidar, I hope there will be something for you to think about the following. So when you're choosing to use Lidar or to invest in commissioning a new survey, you've got to be clear that the area you're looking at is suitable for it. So and there's also the reverse, which is: don't assume that it's just areas of upland that have well preserved earthworks. We found increasingly that Lidar can be effective in picking up low earthworks in lowland arable landscapes. So, for example, we're looking at Newton Kyme. Here's the henge down here, the Roman Fort here and you can see from a slightly different viewpoint, here are the earthworks of the henge, and of the Roman camp as well, surviving as low earthworks, even though they've been ploughed and showing and showing as the subsurface features have been showing as crop marks for decades. Another aspect that is not always appreciated is the way that data is processed. So some algorithms used on lidar for other purposes can in fact remove archaeological earthworks. And so another plea: Please ensure that someone with expertise advises you on the processing of any data that you commission or acquire the raw data and process it yourself. Again we have guidance for this that you can find on our website and we're also happy to advise ourselves or we can recommend people who have experience of commissioning Lidar for archaeology. OK. So back to standards. I mean, Environment Agency Lidar is the most commonly used source of Lidar in the country and we download that ourselves and you can also process it yourselves. And we always recommend that you use multiple visualisations of Lidar. A lot of people are used to seeing hill shading. Matt talked in the last webinar about the importance of multidirectional hill shading, but there are also lots of other visualisations that you can use. And the different visualisations will highlight different elements of the earthworks. And you may find some visualizations easier to map from for example, so it's easier to tell where the edge of something is on one visualisation, whereas it might not be so easy on another. And we use the relief visualization toolbox, mainly because it's free software and you can download the software yourself, but mainly because it was developed by archaeologists for use by archaeologists. So moving on.

Of course, where lidar really comes into its own from an aerial prospection point of view is in surveying woodland for archaeological earthworks. So here you really need to be aware that not all woodland is equal and basically the denser the woodland, the less likely the laser beams are to pass between the foliage and reach the ground to record those all-important earthworks. So, as you can see in this example, the archaeological remains are better recorded in the less densely planted area, so a key consideration for you when you're commissioning Lidar is the type of woodland and the time of year, and so it's worth remembering that you'll need a higher resolution survey if you're going to be looking in woodland than if you were in open ground. I'll just let you absorb that, so I hope you can all see the annotations A, B, C, D showing the different levels of information that you can see, depending on the different natures of the woodland. So again, coming back to the processing of the data, you may not realise that data, especially in wooded areas, can be provided smoothed out. So this is because it's easier to view and can make it easier to understand the pattern of remains. However, this data is from Cannock Chase, and what you can hope you can see. We've got First World War hut bases. We've got First World War practice trenches. We've got a mysterious enclosure that we're still not quite sure what it is, but this is processed data, it’s been processed in this way to help us to map from it.

[They discuss an issue with the presentation]

I hope you can see if we go back that's the smoothed out data. This is actually the data showing where there are actually gaps in the data and so this gives you an idea of where the vegetation is so dense that you can’t actually see the ground surface in these areas. So if I just zoom in a little bit. We've got some hut bases. I don't know if you can see these are the hut bases of the First World War training camp, but if you just look at more closely at one of them. If I go back, you can see I've got a red circle in the middle. This area looks a bit diffuse. There's something funny going on here. And you might assume that's a gap in the earthwork. But actually when you look at the data you can see that these are where the gaps are. And so, it is worth having at least a look at your data with the holes in it. And again, it all comes down to understanding your data and if you understand where these gaps are, it can also help you to target any ground visits that you might want to make so that you can fill in the gaps in your knowledge from the aerial stuff.

We have seen on Cannock Chase where the vegetation was so dense, even on the ground it was really hard to actually tell, and in fact we were getting more information from the Lidar data than we were when we were out on the ground in some of these areas. So, as well as processing, resolution is of course something that you need to bear in mind, and I'm sure a lot of you are familiar with this.

A lot of the Environment Agency Lidar that many of you will have seen, a lot of the older stuff was 2m resolution data and a lot of people are a bit dismisses dismissive of this. But in fact, you can still see archaeological earthworks, so we've got the ramparts of this Roman temporary camp. This is near Hadrian's Wall, other wonderfully named Seatsides and this is the 2m lidar. So it's fine if you can get hold of it then that's great.

But if we look at the 1m Lidar, I hope you can make out on your screens. If you look up here, we've got a patch of ridge and furrow. If I just go back, you can see that the 2m Lidar is just not picking it up, so again, this makes sense. The resolution of the lidar and the height of the earthworks will determine what you can see. And so the difference between the 1m and the 2m is really quite marked. However, if we go to the 50cm, this is still better, but it's not as big a jump as between the 1m and the 2m lidar. And it's worth bearing in mind that actually we're looking at a timeline here. The 2m Lidar was taken a while back then the 1m, then the 50cm and it's worth bearing in mind that in some circumstances the older Lidar data of whatever resolution is now a point in time record of the earthworks, which in certain circumstances, for example, in that lowland context we were talking about, may actually have been gradually ploughed away, and so we now have historic lidar datasets in the same way that we have historic aerial photographs. And so it was worth remembering that the newest data isn't always the best data for what you're trying to understand in terms of the evolution of that landscape.

OK, so that's Lidar. I'm now going to move onto another way of looking at the world in 3D, and I’m going to mentioned drones, mainly because they're increasingly being used in archaeology. We talked a little bit about this in in Webinar 1, but there are lots of there are Civil Aviation issues and I think the main point to make is there are Civil Aviation Authority rules, actually, not just guidance, they are the rules. And if you are commissioning, using, involved in anyway, just make sure that people are following these guidelines and doing things in a safe manner. So I'm just going to illustrate this with a case study. This is Alston lead mines, and we've got a series of shafts. Hope you can see the dark circles and some of which I've got upcast. So we've got you might run from webinar 1. These are shaft mounds, sometimes called bail pits. And this is a scheduled monument, and in 2014 it was identified as at risk due to subsidence and there had been some dumping a material to prevent the stock from falling into these dangerous voids.

So it was felt that detailed survey was required to inform the future management and our specialists in analysing archaeological earthworks at Historic England thought this would be a good case to trial the use of new survey techniques. So, we used a drone that we commissioned a company to do, to take multiple photographs of the site and control points were also taken on the ground and this was so that we could use a photogrammetric technique called structure from motion and this was used to create an orthophoto mosaic and this and the ground control was used to create a highly accurate 3D model. So can you see we're looking at a slightly different viewpoint, but we've still got the slope coming down? I’ll put my green pen on again. The slope coming down here and the shaft mounts with the ground control. So, his model was used to create a 3D model lit from multiple directions, so this might look like Lidar data, but it's not. It's actually a digital elevation model created from that ortho photograph and then lit from multiple directions, and so our colleagues instead of having to go out in the cold and rain, were able to use this 3D model to create, to look at the tops and the bottoms of the slopes and create hazard plan of the site and so they were able to do a lot of this in the office and then they were able to go out onto the ground to fill in the detail, try and understand any subtle relationships between the features so if you want to read about it, we do have a research report that's on the left there.

And yes, so this is a site specific, very intense survey, but this technique of structure from motion is something that we were also interested in. Sorry, I forgot to say there's lots and lots of guidance on this. If you just Google Drone Historic England, it's going to bring up all these three things, but you have got the web addresses there, should you need them. And again you can always ask us that are people here. We've got lots of people with expertise in this technique in House Historic England, so do get in touch if you need to ask us about it. So moving on. As I said, we in aerial survey have also been exploring this technique, and so we've been exploring using structure from motion techniques using aerial photographs taken from a light aircraft. And the advantage of this is that you can cover large areas quickly, and here we're looking at an area of about 8 square km along Hadrian's Wall, centring on Birdoswald Fort, which is about there. And in a relatively short time, in just one flight, we were able to achieve high resolution data at a fraction of the cost and the time. that it would take to do a drone survey of such a large area. And this work was used for mapping for archaeological mapping in a contextual area around the moor as part of work for the English Heritage Trust. So it's definitely worth thinking about what you really need, if you're thinking of commissioning your surveys.

Does the Environment Agency Lidar fulfil your needs? Is that good enough? 1m, 50cm, whatever it is or would aerial photography and structure from motion be a better alternative, or do you need the very detailed information from a drone survey? So, for example, we're looking at here top left of Birdoswald, the Environment Agency Lidar was patchy, and one resolution, but we were able to create a 3D model that we could use in the same way and this we achieved 18cm resolution and of the whole area. However, the drone photography was used as well and this offered an amazing 2cm resolution but concentrated on relatively small areas.

So again, just think about what you need for your work. What can you afford? That kind of thing? Yes. So structure for motion is kind of any quite an exciting new technique that we're exploring. OK so my final slide from me is again just: use as many sources as you can look at the hardcopy print and this is a slide from Webinar 1 which is you're going to get different information from different sources so please use as many as you can. And don't forget the non-digital material. And finally please, please, please make sure that you document which source is used and which you didn't because that will be key to people understanding the survey and the mapping that has been created from it. So I'm now going to hand over to Matt who is going to talk about standards for mapping.

Matthew: Hello everybody, so as Helen said, I'm going to look at the next stages in the process now. So looking at some of the standards from mapping through to recording of the archaeology that's been identified and also archiving and dissemination as well. So we're going to start off with another question from you, just to get a bit of a handle on how many of you have created aerial data, how many of you used it, or commissioned it, and if you were commissioning this, did this also include mapping from aerial sources as well as just an assessment? And remember, this is a multiple answer question, so you can tick as many boxes as you like. Excellent, well that's very interesting. So yes, a lot of people seem to be quite used to using data. There are quite a few people that they have some degree of familiarity with mapping and quite a few people that have created data as well, so we're going to be looking in the last part of my slide in a little bit more about commissioning surveys and also assessing tenders when you've actually put workouts to tender as well. That's fantastic. Thank you.

OK, so moving on to our next slide then. Moving on from what Helen was talking about with the mapping, we're going to look at the next stage that we take after that initial interpretation of the images, and this is taking the aerial photographs and transforming those images so that they actually fit a map base and this is a process that we called rectification. And the rectification process basically takes all of the distortions out of the images and it aligns that image to a specified coordinate system. So in our case it would be the OS National Grid.

A very important point to make here is that digital rectification in the way that we're discussing it today is not the same as rubber sheeting in a GIS. So rubber sheeting will just distort an image to make it fit, so you'll use similar techniques of selecting control points and stretching the image, but it's just distorting the image so that it fits onto the map base.

Digital rectification, however, uses specialist pieces of software to carry out a full 3D geometric transformation of the image, so this very accurately removes all of those distortions and creates a very accurate product at the end of it. For archaeological survey, a couple of programmes are in common use.

The first of these is Aerial, which is the programme that we use in house and this was developed by John Hague at Bradford University. There's also an equivalent programme called AirPhoto, which was developed by Erwin Scholar. Both of these have very, very similar functionalities. One of the main differences between the two is that aerial was actually specifically designed to work with the OS National Grid, whereas AirPhoto has much more flexibility with various different world grids. The process is fairly simple really, so you take an image.

This can be a scan of a wet film photograph, or it can be a digital image, so that's represented on the left hand side of the screen here and then on the right hand side of the screen we have some Ordnance Survey mapping. So this this mapping is already georeferenced, so in this case we're dealing with OS master map data, which inherently has the georeferencing within it. The process is then to take control points, so these are points which we can identify on both the aerial photograph and on the map. So a series of control points surrounding the area that we're interested in. The programme then does a series of quite complicated calculations and produces a rectified image. So this is the same photo in this case, fitting perfectly underneath that map base.

To increase the accuracy of our models, we can also incorporate digital height data, so this will take out all of the distortions that are caused by topographical variation as well as all of the tilts, distortions that you're getting from an oblique photo or even to degree in some vertical photos as well.

So this is the site of Castle Hill up in Lancashire, just on the edge of the Pennines. This is a rectification that we would term a plain rectification, so this is a rectification which assumes that everything is entirely flat, so this doesn't look too bad, but we can see that when we look at the rectification which incorporates the height data, we can see how much that image is shifting.

So if I just flick between the two, and you can see how the position and even the shape of that enclosure changes with that much more accurate rectification using the height data.

So in terms of our standards we would always specify that specialist software must be used to rectify aerial photographs, so just rubber sheeting in a GIS is not appropriate. So we talked about getting our control from an OS map base. This is a very, very important point to note, because the source that you use for control will actually have a direct relationship with the eventual absolute accuracy of the rectified aerial photograph. So if we take the Ordnance Survey mapping, for example, this is an example of 1:10 000 mapping.

We can see that some of the buildings are perhaps a little simplified in this compared to the higher degrees of mapping that you're getting, and also you do get some standardisation of certain elements. So for example, the road widths are very often standardised, so if you're picking control that intersects with the edge of the road this might be 2, 3 or 4 metres out from its true ground position.

But if we look at the OS Master Map data so this is at 1:2 500 scale, we can see that this is much more accurate. You've got some extra features being depicted that weren't depicted on the 1:10 000. You're getting extra and much more accurate detail on the buildings and also the widths of the roads are far more accurately depicted. It is possible to get hold of orthophotography, so those in the public sector may have come across this as part of the Aerial Photography for Great Britain or the APGB agreement that we discussed briefly last week. So if you are getting all the proper orthophotography from a website, then this is actually a very, very accurately orthorectified and georeferenced beforehand, and so we can actually use this orthophotography as control to rectify aerial photographs. Or of course we can directly map from it, and so also photography can actually be a very, very accurate source to use for archaeological mapping.

Occasionally, we might find that older versions of maps are used, so sometimes features may be depicted on older versions of maps that you may want to use for control on particular historic photos, but it is always worth treating these with caution because historically different map projections were used and also the methods of surveying were far less accurate than they are today. So if you are using historic mapping in any form this this will introduce that extra level of potential positional inaccuracy in your final data. So just a couple of examples of this.

So on the left hand side of the screen we can see an area of some crop marks in Nottinghamshire. These have actually been mapped separately on two different occasions for two different mapping projects. The depiction of the crop marks that you can see in grey is from a mapping project that was hand drawn and we'll look at some more hand drawing examples in a moment. So this is manual transcription and in this case it used a mix of manual and digital rectification, so there was a very early version of aerial that was used, but there was also manual rectification, so the two techniques are the paper strip or the network methods that you can read about in David Wilson's Orthophoto Interpretation for Archaeologists. The marks that you see in green are mapping that was undertaken using digital rectification in an entirely digital environment so we can see that the difference between these two is off by kind of 20-30 metres in places.

An important point to make is that not all hands transcription is equal. So sometimes hand transcription will be sketch plotted, so this is just literally using the skill of the interpreter to estimate the position and the morphology of features. But it can actually also use digital rectification. Sorry, yes digital rectification, which will be much more accurate. At the other end of the scale, on the right hand side of the screen, we can see an example from Wiltshire of an enclosure and some various related features. In this case, the features that you see in magenta are features that were digitized from a geophysical survey which was surveyed in using a GNSS device. So this is positionally accurate to around, about 10 centimetres and once again in green is the AirPhoto mapping that has been undertaken of this site. And what's very interesting here is just how accurate that AirPhoto mapping is using the error rectification packages in the main enclosure. You can't actually really make out the green line around the outside, because it perfectly coincides with the results of the Geophysical survey. So in this instance the AirPhoto mapping is actually of a high enough level of accuracy to use to lay out trenches for excavation, for example. Helen has talked about the historic mapping projects that have given us around about this 50% coverage of England. These have been going on for around about 30 years and throughout that time the standards and the methods, for these projects, as well as their scope has gradually changed and evolved to give us the standards that we have today. And as Helen alluded to, actually understanding what these standards, methods and scope were is very, very important for understanding the data if you intend to use it in any way.

So the very early projects were hand drawn projects, so this was taken an overlay putting it over an OS map at a scale of 1:10 000 and physically drawing with a fine line pen the archaeological features. So what we find here is that there may be considerable inaccuracies in some places, particularly on upland landscapes where there's not very much control to derive from the OS mapping, sometimes they will have limited scope, so the mapping of the Yorkshire Wolds for example was only mapping cropmark archaeology up to and including the Roman period. You will also find that there are some schematic conventions used.

So here we can see that an asterisk is used to denote a shaft, because these features were just too small to properly draw using a pen and we've got some other some other abbreviations such as LK for lime kiln for example. The early digital projects you can actually trace some of the lineages back to the hand drawing quite clearly in these because they tend to use a mix of polygonization of features and a single line depiction. So features that were over 2 centimetres were depicted as a polygon and features less than 2 metres wide are depicted as a single line. So these are very accurate in terms of the ratification process because you're using digital rectification, quite modern maps, but you're actually losing some of that subtlety of the archaeological features depicted as single lines.

The standard that we have today is almost entirely polygonised, so this means that you can pick out those real subtleties of the morphology, and the relationships of features to one another. We do retain a couple of schematic depictions, so ridge and furrow, for example.

The extent of the ridge and furrow is outlined and then a direction arrow or a single line is used to depict the direction and the morphology of the ridge. And we also use a schematic t-hasher convention as well. So the modern standard for mapping is ideally undertaking mapping in a GIS environment, which enables you to undertake the mapping with access to lots of other datasets at the same time. Also using attribute data is really key, so this is data that sits behind the spatial data and can give you essential information about the features just by clicking on them in a GIS. We have minimum standards for what attribute data should be included within digital mapping, but these are very much minimum standards.

For some projects it may be appropriate to include a lot more attribute data than we normally would do. One of the things to consider is the subsequent users of your data and how that data is being used. If it's going to be used by people out in the field on a digital device, for example, it may be very, very useful for them to have all of the data within the GIS so they can just click on a feature and see all of the information that they require.

So good reporting is as good as important as good mapping, and this really is a very true statement. You can see a quote at the top of this slide, and this is actually a genuine quote from a monument record. You’ll see as the user of this record, we know that it's come from an aerial photo and it's crop marks, but beyond that, it's really not very much use at all. I'm not going to go through this in detail, but once again we do have minimum standards for what should be recorded within the textual monument record that accompanies the spatial data. So just simple things we talked last week about thinking of these features that you're seeing as archaeological sites, so they're not just abstract shapes that you're seeing in a crop, so the form and the data and the function should be discussed wherever possible.

Also, another really important point is referencing and sourcing, so ideally an individual should be able to trace back to the original source material, the monument record, so they should be able to identify the exact photograph that this particular site was recorded from. And these standards are very much outlined in detail in existing sector guidelines. So these are the guidelines that are available on the archaeology data service websites, The Informing the Future of the Past. So if you Google that you can access this on an ADS and Matt’s put a link to this in the chat as well.

Now dissemination and archive. Helen has already alluded to on a number of occasions, how important getting data into the public domain and in particular into HERs is. So the data should ideally be available. People should be able to find it. They should be able to search the data and it should be in a form that's usable to them. So getting it into the HER is the absolute perfect situation and this can be accessed via Heritage Gateway. Also, you could consider other ways of disseminating and archiving that data. So for example the archaeology data service and the report are also a very key element and Helen's already discussed this. So, as well as that narrative discussion on the archaeology, also including how the survey was undertaken, what was looked at, and how it was recorded. Very, very important elements to include in any subsequent project reports. So, defining briefs and assessing tenders. So what we're going to look at here are just a few pointers for the kind of elements of defining a brief, or maybe assessing a tender that somebody has submitted in response to a brief that may hopefully just make sure that you get the most out of the data.

So we're actually going to start this off with a question, so I'll put this up for a minute or so and then Matt will actually upload a question so you can do that in response. So we've got an executive summary, which is entirely made up. I’d just like you to take a minute or so to read through that and just think are there any areas of this which maybe raise a bit of an alarm bell with you? Any areas for concern? So thinking about things like the accuracy of the data and who's involved in the projects, the kind of scope that they've defined, and what kind of sources that they’re going to look at, for example. So we'll leave that up for a minute or so and then Matt will upload a question. Once again this will be a multiple ticking one, so you can tick more than one answer if you wish.

Matt: While we're waiting for people to read that there's a couple of comments in the chat. There seems to be an appetite for an online lidar course, so there’s a takeaway for you, but also, some questions regarding accessing raw lidar data and people were interested to know how they might be able to access raw lidar data.

Matthew: Yes, we can certainly get back, I’ve noticed a couple of people who’ve put email addresses in the feed so that we can get back to them on that. Great. OK, let’s bring a question in again, so again, there you go: vote away.

Matt: Now, I’m not revealing the results, so the voting won’t sway people’s decisions, but once the voting seems to have settled down, we’ll reveal everything. OK, that’s an interesting result. Let’s reveal. Here we go. What do you think about that, Matt?

Matthew: Well, we’ll go through and have a look at some of these in detail, although I was a bit mean, because there are a couple and you could go either way and they are perhaps moot points. I’ll take you through my thinking on it as we move through the next slides. So, thinking about the kind of thing that you may want to define when you’re putting a tender out. Existing data is a really important one and like many things that we’ve been talking about, this might seem obvious, but it’s surprising how often this is overlooked.

So, if you’re considering commissioning any work, a really good first question to ask is what mapping already exists and could this actually meet your requirements?

We've talked about the various different scopes and methodologies that have been used historically in the past to create this aerial mapping data, so it might mean that the mapping data that already exists does need to be enhanced in some way so it's worth having a think about what is required just to bring it up to the standard that you require.

What kind of extra sources may need to be looked at? Whether it's a matter of completely redigitising for example. If you're assessing a tender, it's also worth inquiring whether the contractor has actually looked at existing data. This isn't always looked at as part of a DBA, for example, but it could well provide that extra data about that broader archaeological context, insight of particular archaeological interest. So staff. So in this case we have a member of staff who actually seems quite experienced, a couple of years on the job, but will be supervising somebody who's relatively inexperienced.

As Helen said, a really important point is that AirPhoto interpretation mapping is a specialism, and it isn't something that everybody can just pick up without any training whatsoever. And the quote on the right hand side of the slide is a genuine quote that came up in a conversation I was having with somebody last year. In this case, I think it's actually pretty strong on the staff, as I say we've got somebody who's worked on, in this case, one of our projects, so they've hopefully been very well trained. And by other stuff on that project, they've worked for a couple of years, so we generally recommend around about six months plus, but they are supervising somebody who hasn't worked on this kind of project before, but that's actually a real big plus point, because if we can continue to skill the sector bringing through the next generation of aerial archaeologists, then all the better. So project area is another really interesting consideration, particularly with commercial projects. This will come to a degree down to a cost benefit decision, but there may be advantages that actually outweigh the cost in doing larger areas.

So in this particular example, we've got a hypothetical example of a school being extended. In blue, this is what we would term our red line boundary for the site so we can see that we've just got a couple of fairly fragmentary ditches in here, so we don't really know what these ditches are. We don't know their function. We don't particularly know their dates, and we can maybe make an assumption that they might be more extensive than is indicated by the crop marks alone, but we can't really be certain. However, if we look at a broader context for this. Then this has a couple of effects on our perceptions of this field. So for a start, it's really improved our understanding. We now know that these are actually just fragments of a much broader iron age Romano-British field system and also it does have implications for the archaeological potential of this field.

We can see how extensive these features are elsewhere, so it's fair assumption that similar extensive survival may exist underneath the grass of the current school playing field, and so this can be really, really useful for informing risk management mitigation, and particularly informing subsequent project stages. So moving on to the scope then. So in this case they're using a scope which you hopefully will recognize from slide one, so we're looking at all features up to and including 20th century military features and industrial remains. So that's the kind of broad scope that we would that we would use for all of our projects. It's always worth considering, though, whether the scope can be expanded to include other useful information.

A couple of examples here at Thornborough Henge where the paleochannels have actually been mapped to better understand the position of the archaeology, but also any potential masking effects that that has. And then on the right hand side, an example from the Lincolnshire Wolds where not only this ridge and furrow have been mapped in greater detail, but also geological features and even recent field trains have been mapped to provide that extra level of information for subsequent geophysical surveys and excavations. So project area, this was quite a tricky one, sorry sources. So thinking about the sources. So here we have a project area that is largely surviving as archaeological earthworks, but some of it has been in arable cultivation since the late 1940s, which actually begs the question: what potential information could we be getting from those early 1940s RAF photos? Could there actually be earthworks surviving on those photos that have since been levelled? So we need to be thinking about what sources are being assessed, how they are being assessed, and in particular in viewing in stereo, and if they're using Lidar in an appropriate way and the illustration on the right hand side, we can see the AirPhoto here fairly flat arable landscape and then the light or data of the same area, just by way of illustrating that you do need to be careful about assumptions about how useful sources are going to be. And then finally looking at methods. So in this case they're using aerial as a rectification software, a big tick there, but control is derived from the free Ordnance Survey mapping data. Now free data is never going to be that accurate, otherwise they'd be charging for it.

So this is the alarm bell here is that they're using the right process, but their control data is likely to be quite inaccurate, so the kind of things that we're considering here, the assessment, so once again how? How are they actually looking at those sources? What is that rectification and mapping process? Are they using attribute data? Are they undertaking monument recording and are they creating data that will be appropriately archived and disseminated? So that's the end for us. So thank you very much for listening, and I'm sure that Helen will unmute at this point, and it's really a handover to you to ask if you have. Any particular questions? And I think Matt is you going to put up a question in the Q&A window that people can fill out.

Matt: Indeed, yes, and Matt and Helen. Thank you so much. As always, an absolutely fantastic presentation. If you do have any questions, I know we're pretty much up to the wire at the moment, but please, please do feel free to type a question into this Q&A box that you can see on the screen. Now there's a tiny little text box at the bottom of that window. Just type in your question now, we will respond to those as they come in. I'm also going to provide you here. There you go. There's some. There's some links as well, so if you actually type, sorry, click on the title slides or webinars, etc. you'll see the URL appear at the bottom of that window, and then you can navigate to that URL So slides, if you'd like a copy of today's slides. That's where you'll find them clicking on that that link for slides. ‘Webinars’ is a link to our Historic England pages that lists all our future webinars, and indeed lists all the recordings for our previous webinars. So if you would like to revisit or visit anew the first aerial photography webinar. That's where you'll find it. And indeed all our guidance documents as well. If you click on guidance or the HP guidance documents will be there as well. So it's for me just to say thank you so much to Helen and Matt for an absolutely wonderful presentation. I hope you all found it as fascinating and informative as I have, and I certainly look forward to an online course on Lidar. That's just going to be awesome. Sorry, no pressure guys, no pressure. And it is also for me to say thank you to all our guests today for taking the time to join us. Thank you very much. We've got a few questions coming in which we can respond to very quickly, I'm sure. Oh OK, so Stuart. sorry you missed the first 20 minutes, apologies. Was there any information on the use of multispectral aerial photography? I think we covered structuring in motion, didn't we or sorry, you’d better answer that, Helen.

Helen: We didn't cover multi or hyperspectral aerial photography, mainly because it's not widely used in the UK for the kind of mapping projects that we do, it's definitely going to be a source of the future and we are exploring it through the current space. For Smarter Government or Smarter Government in Space Initiatives we are going to be looking at that, mainly in terms of prospection and we're going to try and see whether the hyper in multispectral data might bring up crop marks or give us indications for crop marks by appear before they become in the visible spectrum, but it's a kind of watch this space, I mean people have done a lot of work on it and again if you look at the Aerial Archaeology Research Group website if you just Google AARG, there's quite a lot of papers published there, or there are people that you can contact. You want to do the next one, Matt?

Matthew: Yes, sure, just to follow on from that that previous question as well. It's worth having a look. We undertook a project a few years ago when we were still English Heritage called Minor Farmer Landscapes of the North Pennines A&B. And as part of this, Birmingham University works on multispectral imagery that we had commissioned for that project and there is actually a separate report out on the use of the multispectral imagery in the lidar, so it's worth it's worth routing that out and that that should give you some more ideas as well. So moving on. Melissa, you're asking about the various NMP projects at a glance. Well, a couple of couple of ways that that you could actually look at this data up. Firstly, if you if you go onto our websites and look at the Historic England Research Report Series. We've actually got an interactive map of the research reports, and this includes polygons of all of the past NMP projects on it. Also, it's I think we were mentioning in webinar 1 we are in the process of making all of our aerial mapping data freely available online and all of this should be making appearances in online GIS portal next year and this so this will give you access to all of the mapping but also include those kind of project polygons as well so you can look at that not that big scale and see where we've covered.

Helen: Somebody's been asking about what's the best source for height data. We get ours through the air photography for Great Britain Agreement that we have. That's the APGB agreement. I don't. I'm not aware of many free sources of height data, though I don't know if anybody out there, or people who work in commercial contexts, or you Matt?

Matthew: No. I think. I mean there that the OS is an alternative source of height data and I have it in my mind that there might be a free OS topographical layer. The problem that you come across is it really depends on what scale you're working at. So the height data that we use as part by rectification process is the five 5m interval data. We have actually tried using 2m data, but we found that this this actually caused more problems than it solved through a variety of reasons that I won't go into. So we would say yes, the f5m interval contour data is what we use, But you might find that the free data tends to be at a cruder scale than that, but it is worth I think that there could be some height data available by the OS.

Helen: OK, somebody's asked, do we have any standards or advice for scanning air photos? Again, that's a bit of a how long is a piece of string question in the sense that it depends on the size of the photograph and the size of the features that you're looking at on the photograph. So it tends to mean, I've used 400 to 1200 DPI depending on what I'm looking at, so it is, it depends on the size.

Matthew: Right, yes, I think I think yes, just scanning at an appropriate resolution. really is the message. I mean you do you do actually get distortions introduced by using scanners, but at the level that we're working at, they are fairly insubstantial I've talked to one of my photogrammetric colleagues and his eyes roll when you talk about using scanners because he’s thinking in millimetres all the time, or even or even less, but for yes, for our kind of purposes, then normal flatbed scanners are fine.

Helen: Yes, and a lot of people working in commercial contexts are now photographing photographs to work on them as well, and that's something that we haven't really explored in terms of the effects that that has on accuracy. But again, as Matt says, the people that are doing that are working in a commercial context and are very experienced, and if it's good enough for them, then it's probably OK. Somebody is asking about advice on training volunteers. Well, that's something that we can help with. We do courses or we can help you develop training material your sales. So I would say yes, have a chat. We've also published. We had the Cannock Chase example that I showed you with the First World War military camps, that was part of a community focused project called ‘The Chase. Through Time’, and we published a report on that. They included some reflections on how we interacted with the volunteers, but yes, so we generally do workshops. And so we can do specialist workshops on LIDAR. Matt and his team have done that for some community projects in the North, and we've also we've done quite a lot of work with community groups in the South as well. So yes, just get in touch if it's short answer, I think I'll let Matt do the next one. If that's alright.

Matthew: Yes, certainly. Data format. Yes, we do use shapefile. That's as our standard. One of the main reasons is really that this is just a kind of an industry wide standard and is something that can broadly be accepted into anybody's GIS. There are, yes, as you as you allude to, there are some limitations to shapefiles, so in particular things like attribute data. But yes, on balance, it's just the interoperability of the shapefile is a real global standard. It means that we can just hand out to people and use that as part of our data exchange. So yes, but that is what we use for sector data. We did up until relatively recently. We did use to map in AutoCAD as a standard, so we were relatively late adopters to GIS to actually create our mapping. Yes, that's what that's what we were using.

Matt: OK, fantastic, I think that's all questions answered. So Matt and Helen, thank you so much for your time today. Absolutely fantastic presentation. And to our attendees, thank you very much for taking the time to join us today.