Video
GIS and Historical Imagery at Voyageurs National Park
Transcript
Karen: Hello, and welcome to the NPS Archeology Program webinar speaker series for Fall 2013. My name is Karen Mudar and I'm an archeologist in the Washington office. We're about halfway through a series on geophysical technologies in archeology. After a stutter in October, we regained our momentum, and
you can see the complete schedule now on the screen in front of you. If you can't see the schedule, you need to start investigating the problem, so you don't miss any of the graphics in the presentation today, which would be very sad because it's a great presentation. Since this is nearly our final schedule, I want to take a few minutes to talk about it. Our next speaker on December 5th is Alan Sullivan at the University of Cincinnati, who will be talking about terrestrial and satellite remote sensing. I first heard about his work through an article in the winter 2011-2012 "Park Science" about a project that he and his co-authors did at Grand Canyon National Park. If you're interested in that particular topic, you might take a look at the article.
On December 19th NPS archeologists at Midwest Archeology Center Anne Vawser and Amanda Davey are going to talk to us about GIS applications. Anne gave a presentation at the SAA a few years ago on a project that MWAC archeologists were doing to pull together cultural resource data for geo-spatial reference database for each park in the Midwest region. I hope, I anticipate that she's going to give us an update on this very useful project that they've been working on.
We've rescheduled Lori Collins and Travis Doering's talk which will be on January 23rd, not January 19th as on your screen. I'll send out an announcement. I obviously was looking at the 2013 calendar when I made that out. I talked about their work previously, so I won't dwell on it now except to say “Yay!” I'm really looking forward to their talk. I've also invited Jim Delgado to talk to us about underwater technologies for locating and investigating cultural resources. He's going to join us on February 6th. I think that I've heard him speak before and some of you may have heard him speak before as well. You know that he's a good speaker and we're very pleased to have him join us.
We do have one slot on January 9th that's not yet filled. If you are working on a project that you think would fit with the theme of this series and would like to give a presentation, please let me know.
Before I introduce our speaker for today it seems only fitting, given the series' topic, to announce that the NPS 2014 workshop on archeological prospection techniques will be held May 19th to 23rd, 2014, at Aztalan State Park in Jefferson County, Wisconsin. This will be the 24th year of the workshop, which is dedicated to the use of geophysical aerial photography and other remote sensing methods as they apply to archeological resources. The workshop customarily presents lectures on operation, methodology, processing, and interpretation with hands-on use of the equipment in the field. Our last speaker in the webinar series, Ken Kvamme, who talked to us about geospatial prospecting, is a frequent instructor in the course. If you're interested in taking the workshop, get in touch with Steve Devore. There will also be an announcement about it in the Archeology E-Gram with more details and more contact information.
Steve, if you're on, if you want to give a plug for the workshop, now is a good time to do it.
Steve: I thought you did pretty good, Karen. I don't need to say anything else.
Karen: Okay. Nice that you could join us today. Remember, everybody, that all of the lectures are recorded, so set your phone to mute unless you're going to ask a question. The recorded webcast will be posted on the Archeology Program website. I don't think that any of them have been posted yet for this series and we apologize. We've just been thrown off our schedule by the furlough, so we're still working to catch up. I'll announce it in the e-gram.
If you don't receive the Archeology E-Gram and would like to, let me know and I'll add your name to the mailing list. Our speaker for today, Andrew LaBounty, is an integrated resources technician at Voyageurs National Park. He previously held an archeological technician position at MWAC and he worked in a archeological technician capacity at a variety of parks in the Midwest region. Let me forward this, there we go. He holds an MA in anthropology and an MA in geography specializing in GIS and both are from the University of Nebraska in Lincoln. Today, Drew is going to give a presentation entitled, "Capturing Cultural Landscapes: GIS and Historical Imagery at Voyageurs National Park".
The park, located in northern Minnesota, has a rich and complex history of overlapping cultural landscapes that includes both Native American occupation and the fur trade, as well as colonization of the area. Drew's project involved capturing, displaying, and analyzing the activities of the recent settlements to manage and interpret the physical resources that remain. To do that, he's combined historical aerial imagery with early shoreline surveys and modern archeological information. He has digitized, he says, thousands of cultural features into a geodatabase. His study is a good illustration of the ways that archeologists can use GIS to query and visualize cultural activities, revealing spatial patterns over time and directing future research.
Thanks very much for talking with us today, Drew.
Andrew: Thank you, Karen.
Karen: We're going to switch to your presentation, I hope. There it is, okay. You're away.
Andrew: Okay, thanks very much. And thank you everyone for coming, everyone who's online. I wanted to point out, too, that I'm sitting in the room, actually, with a lot of my colleagues at Voyageurs. I'm very excited to hunch over the speaker phone and tell them a little bit about what I've been doing, too.
As Karen mentioned, my topic here is capturing cultural landscapes. As part of the series this is going to be all geo and no physical. I hope we'll be able to get something out of it anyway. My goal is to just demonstrate how we can use GIS archeologically speaking to do a landscape study like this, complex and long term.
To begin, I would like to talk about some of the challenges and goals we had in capturing a cultural landscape. First, what is a cultural landscape? There are many different ways to define that. My goal in doing this project was to capture the essence of a landscape in a way that's going to allow other researchers to bring their own resource questions and to bring their own data. Then, to develop this database in such a way that things can be added, things can be overlaid. It was challenging, then, to develop a database that could synthesize and display multiple lines of evidence through time and space.
Are you getting a glare?
Speaker 4: No.
Andrew: We turned to GIS, then, because in order to make a product that was of use to archeologists and to land managers, we wanted these landscapes in a sense to become "query-able". In terms of using a geodatabase, because it captures these spatial relationships. Using this method, resources can be overlaid-
Karen: Drew, can I interrupt you?
Andrew: Yeah.
Karen: So sorry. I can hear a lot of rustling and moving around in the background and I just wanted to remind people before you get too far along in your presentation to please mute your phones. Thank you.
Andrew: Thank you. We approached this from a GIS perspective then because our goal was to look at a landscape in a way that would be useful down the road.
Those were our challenges, those were our assumptions in creating this database. Then to talk a little bit about how we really got started. At Voyageurs we implemented this in an effort to understand aboriginal land management practices. The park had purchased some 1927 aerial photographs. Those had been geo-referenced and processed by the Minnesota Department of Natural Resources. The question was, can we use that resource to record, document, and even do some ongoing research with 1927 Ojibwe landscape features.
In early August last year, I spent a couple of weeks at Voyageurs. I was then working at MWAC. We began to look at some of these aerial photographs. Before I get too far into that, though, let me tell you a little bit about where Voyageurs is, if you aren't familiar with it. It's in far northern Minnesota and it's about 35 miles across, that's 320 square miles. Just about exactly half of that is water. This landscape today is rested. It's a natural landscape that's meant to reflect the way it would have looked to the fur traders in the 17th and 18th century. But, in 1927 it was alive with people and culture and a lot of overlapping cultural landscapes.
The way we began this project was really to look for Ojibwe features, including things like pits and caches or ricing jigs, which were shallow pits in the ground that were used to remove the husks from rice. We were looking for burials and rock rings, if possible. There was some question as to whether the aerials would even depict that kind of thing because they were fairly old. Looking for spirit houses or other structures, like houses and docks. Then, as we went through this project we began to see that there were many, many overlapping landscapes, specially logging, but that there was also fishing going on at the same time. The trees had almost all been cut down in the park. A lot of the lowland vegetation remained and that actually exhibited some land management practices, even in the lowland vegetation. Then, at that time, there were much higher lake levels, so we see some evidence of climate change as well.
In order to capture all of these things, as I said we had a series of IJC, that's International Joint Commission photographs from 1927. Those were purchased by the park and actually flown by the IJC in an effort to manage the waterways between United States and Canada. We also added in some IJC surveys and some International Boundary Commission surveys as well. These were surveys conducted in the 1910s. We had historic maps of those as well. As I mentioned we began looking at the known Ojibwe sites and to do this we used an archeological report written by Jeff Richner in 2002. He identified the Ojibwe sites based on the documentary evidence and the feature patterning and the artifacts found at those sites. We started there, looking at places where he knew the Ojibwe were living and trying to examine some of the impacts they had on the landscape. We also see that there were a lot of overlapping landscapes. We started to add those and we've built this database now into something that can be used to examine and to visualize the ways that these landscape features interact with one another.
Upcoming here is just an image showing you the coverage of the 1927 aerial photographs. You see that the planes flew in various flight lines from south to north in the park. You'll also see that we're missing some critical areas. The south shore Rainy Lake we don't own aerial imagery for. That was about 13% of the land area in the park that we don't have a record of in 1927but we supplemented that with other historical maps.
To pause for a second and talk about exactly how that works and what I did to actually examine these aerials, you'll see that as the plane flies overhead it takes successive photographs. Using a stereoscope we can feed one image to each eye and then use the difference in perspective to sort of trick your brain into seeing these images in 3D. This was extremely helpful, the stereoscope was an old technology from 1838. These images were not taken with any thought of 3D. However, just because of the fact that they overlap enough we can use a stereoscope and it was critically important just for determining even the difference between water and land.
You can see in these images the water is black, but a lot of times because of light it looks exactly like the ground. The only way to see the difference between land and water is to actually see the land rising above the water. Plus, looking at just one image you would never know the difference between a tree or a building or just a spot of dark rock. I used the stereoscope to look at these images and then using GIS I plotted the points that I saw, or the polygons or the trails in the GIS, and created the database that way.
As I said we also had a lot of supplementary information. We added the IJC survey and the IBC survey to the GIS as well. I went through and geo referenced these, much like the images. I placed them on the earth's surface and then I used the database to add points like buildings or survey notations to the database as well.
Going forward, we wanted to add some historical images and some historical records, such as these logging records here. They are in a tabular format so we can easily join that information to the database. Then we also have some very nice historical photographs that we would like to add as well. Actually, there was a webinar just about an hour ago about how to add photographs to a geodatabase.
The question really becomes, how do we do this? The rest of my presentation will talk about how we did it, and what we're going to do with it, and some of the other examples and ways to use GIS. It's a complex problem to take a natural landscape and to convert that into something that we can query and visualize. Here's my methods slide. Naturally this is how I feel when I'm doing it. I zap the landscape and then it turns into a glowing energy inside of a box. Just like the ghost-busters, that's kind of a fictional ideal. What I really did was capture what I thought was important about the landscape and, hopefully, everything that comprises that landscape so that other researchers can use it down the road.
The geodatabase or "ghost trap" that we ended up designing is actually based on the National Park Service’s data transfer standards, which calls for a large number of variables for each landscape feature. I ended up implementing only about 25 of these, but they store the standard kind of who, what, and when information, like the source of the feature. Then because it's a geodatabase, that information is attached to a specific point on the earth's surface as well. This is what allows for the queries and the colorful displays and the analysis and so forth.
For example in this particular geodatabase, I implemented resource names as perhaps being the kind of resource that it was, like a structure. Resource type could be culture, like Ojibwe, and so on and so forth. Then you can apply those variables in different ways depending on what you wanted to look at. Perhaps a orange square labeled Joe Biden's cabin would be an Ojibwe structure.
Here's just some quick statistics about our particular geodatabase. VOYA covers about 200 square miles of land and the ICJ aerials cover a certain portion of that. We ended up identifying 143 different "sites" or collections of features. The interesting thing to me, I didn't realize this, but there were nearly 1,100 miles of trails and logging roads in 1927 at Voyageurs National Park.
This is what the user interface actually looks like. After you get all of these features digitized and in a geodatabase, this is what ARC GIS looks like for those who don't know. You can see this down here is the attribute table. It contains all of the who, whats and when information and the source of the feature and so forth. Then the data view above that just shows where on the Earth's surface it's located.
If we zoom in on that, the rest of my talk is going to be examples about the various overlapping cultural landscapes of Voyageurs National Park. The way I'm going to approach it is, as you can see it's a mess. It's a jumble of cultural features. Each of those symbols represents a different kind of feature. Whether it's mining or homesteading or logging or fishing. I'm going to go through each of those, one at a time and use each cultural landscape as a way of talking about a different aspect of the geodatabase.
I'll begin as the other features fade away, we'll be left with just Ojibwe sites. I'll start with Ojibwe because we know the most about that and that's where we started this project.
Bois Forte Ojibwe represents 12,000 years of culture history at Voyageurs National Park. Their historic occupation was from about 1730 to 1940. A lot of the things that we can see in the aerials could be attributed to the Ojibwe people living in the park. It's important to note, though, that the 1866 treaty moved the Ojibwe from Vermilion to Nett Lake. However a number of Ojibwe people actually chose to give up tribal membership and remain in the park by taking homesteads, legal homesteads within the park. In the 1800s there were 4 bands here. Looking at the structures and the land ownership, we'll be looking at about 100 permanent residents in these aerials, specifically.
Again here's the map of the Ojibwe features and I want to take us down right now to Alligator Bay in the southeast corner of the park. It covers an area of about 2 1/4 miles across. This is a modern topo of essentially what it looks like today. What you see here is an overlay of the 1927 aerial photograph. You can see it's fairly grainy and they aren't a whole lot better than that in real life. You can see some trails, some scars. The water, there's a lot of wind action there. You can see it turns white and dark and it's very difficult to see, even in some cases the differences between land and water. But looking at this in a stereoscope you can easily see these features that really pop out. As the trees - well, where there is bare earth you can see trails. This brown striped area up here has actually all been cleared by loggers. I'm going to go back for a moment and you can see the difference between vegetation between the area that's been logged and the area that hasn't.
If I take the 1927 aerial away again, you can see what it looks like on today's landscape. One thing of interest to me is this trail clearly passes through a nice valley here. Another thing of interest is this area here is known as King William's Narrows. It actually is so narrow and the flow is constricted, it does not freeze in the winter. We've interpreted this trail as actually a winter trail from the north shore of Crane Lake which is down here, up to Alligator Bay, which is this area here.
These on this side are all logging roads, and that's why I've symbolized them as brown. That's one of the power of the geodatabase is that looking at the different attributes that I've already recorded, I can change how it's displayed and I can look at different kinds of patterns. That is clearly a logging road, though, because you see these teeth on the road itself. That's very characteristic within the park because the loggers would pull their sleighs, actually the horses would pull the sleighs, and then they would have the ruts filled with ice. They would pull back into the woods, load it up or turn around, and then pull the logs back out and take it to a lumber camp where it would be even milled or dumped in the water or something. Then they would take it out to sell them.
Now what I've overlaid here, these dots are actually known archeological site locations. It's a subset within this area. These two green ones are known Ojibwe sites. I find it very interesting that actually this winter trail connects to otherwise un-linked Ojibwe sites. That could represent some kind of community. It could represent visitation. I think even more interestingly is this spur that goes directly to a completely un-associated site. This is a prehistoric site with no particular cultural affiliation that we know of. That's going to bear some further investigations to see whether it was related, because clearly in 1927 that area is being used. It could be either one of those two; north and south sites.
It was also gratifying to see after I was done digitizing that this logging camp appears archeologically right at the terminus of all these logging roads. Things like that help us to understand what we're looking at and the network and overlapping landscapes that we see here.
I also want to point out, here there's a very stark boundary between logging and this is actually a section right here on the topo. This track was owned by an Ojibwe man and was not ever logged. Even today Alligator Bay is one of the only places left in the park that has virgin pine. I can tell you I've been there this summer and it was pretty awe-inspiring. Those trees are twice as tall as they are elsewhere in the park. It looks like a primordial landscape. It's very interesting because that cabin, the remains are there, as an archeological site. This says something about the way the Ojibwe managed their land. There is no feature here in the 1927 aerials because it is obscured by the timber. It was illegal to cut because that Ojibwe man gave up his tribal membership to homestead that area. That is why we still have trees there today.
That was everything I wanted to show you at Alligator Bay. Now I'd like to move over to the Moose River Indian Village where you see a large cluster of Ojibwe features. This is a larger area we're looking at this time, 5 miles across. As such, it has a great many more aerials. As I was doing this analysis I looked at every pair of these aerials, the overlapping aerials, so I systematically went from one end to another, from north to south, and just recorded everything that looked cultural on them, in the GIS. This is what I came up with. You can see this is kind of a messy jumble as well. I take the aerials out, it's a little easier to read. Once again, the brown areas are logging-related, either clear cut areas where the timber has been cut down and taken out. Then the roads and railroads with the spurs that we can see again. Then the green, again, are the Ojibwe sites and Ojibwe features, whether they're mounds or grades or trails or what have you.
If I now just query just the Ojibwe features out of this, you can start to see, especially right in here, why we refer to this as the Moose River Indian Village because especially right here the features are all facing inward toward one another. If you go out there today and stand on one of these islands you will be able to see and hear people actually just across that narrow bit of water. Looking at it from the top down like this, you can begin to understand the landscape a little better. Even more striking, though, is when you turn on the Ojibwe ownership tracts, you can see that just in this area alone a lot of land was owned by the Ojibwe people.
Looking down here we see some interesting features. This one in particular is going to be something we want to look for later, find out what was going on there. Looking at images like this, I think, is very cool to see why these features are where they are and to know the history behind how those people used the land. Also fascinating, I think, is Cemetery island was not owned by anyone in particular, but many of the people listed here on these homestead records are buried there, or their family members. This takes on a new role as the community island, almost, for this village that is otherwise essentially invisible unless you look at it from top down.
Now we'll move up to Gold Portage and look at something entirely different. This is a very quiet, relatively flat area. It was the same in 1927. This area was owned by the Ojibwe people and so we were hoping to find a lot of structures and a lot of different things indicating the way they used the land. In fact, we were a little disappointed to see that's all there was there. We're not even sure if this is the Gold Portage itself that was used by the Ojibwe or whether it was an old mining road. What is clear, though, is that this land was not cut and, whether it's because the trees are obscuring the features or whether it's because the Ojibwe simply weren't doing something on the land, there are no other cultural features here and it looks very, very different from the rest of the park. The fact that there's nothing there is fascinating in itself. It shows that where we don't have data recorded is also culturally impacted.
My theme there in showing you those examples were- I wanted to talk a little bit about research and discovery. Here, moving on to homestead features, the theme for the database now is talking more about documentation. Homesteading began of course with the Homestead Act in 1862. If you recall the Ojibwe weren't necessarily displaced until about the 1880s. At that time, the Ojibwe could return to the park by giving up tribal membership and claiming a homestead, but most homesteads here are circa 1900. Patenting required you had to do a filing fee, you needed to live on the land and make improvements for five years. So the question became, if all of this happened in 1900, it should be visible in 1927.
To look at that I want to take us to the Kabetogama Community, which is actually outside the park. It's a fairly large area and as you can see from the modern topo, there's actually still quite a community there today. In 1927, we definitely see the beginnings of some farms. There's a lot of almost rectilinear roads. A lot of buildings and fields. My point here is just to show you that even features that are outside the park can be documented in this geodatabase and they reveal something about the landscape of the park at that time. Unfortunately it's not clear that all of these were homesteads. A lot of them probably started that way, but this might not be a great example, actually, for homesteading, per say. It's not even within the park.
We'll move to another place, Randolph Homestead. My point here is, in fact, that the simple answer to my earlier question is no; that homesteads really are not visible in 1927, unless they became something that was reused repeatedly. This homestead actually turned into a commercial fishing camp and as such it was developed. As you can see there are a few structures, that these are not as noticeable as a lot of the logging camps are, which we'll get to later on. I'll talk about fishing next, but of interest to me was that a lot of the homesteads were cleared only for the land. In some cases homesteaders seem to have been paid to own land by logging companies or other commercial interests. In a lot of cases the homesteads here just didn't really pan out. Anything that we can document about those becomes very interesting because we can see clearings towards patenting that was not successful, or we can see structures that actually were constructed and people did live there.
I'll move right along then to fishing features. Commercial fishing happened in part between about 1890s and 1980s. Fishers were here looking for sturgeon and later walleye and northern pike and white fish. Some of the notable characters should definitely be visible in 1927, including Adolph Hilke who was the first, and Emil Torry, Edward Stoffel and Frank Bohman Sr. These are the fish camps within the park that we know about. I want to look real quickly at the Bohman Fish Camp. Note the spelling there, B-O-H. All too common on the topos here, especially, we end up with names that have been changed. Bowman Island now is with a W.
For this one I want to look at some climate change, too. I want you to especially pay attention to the shoreline of this island as I go to 1927. You'll see it gets a lot smaller because as I mentioned in 1927 the water levels were much higher. Whether by design or by accident, these features, and they're difficult to see there in the aerial, but they are partially in the water. We know from surviving examples that a lot of fishing structures were built right next to the water and maybe partially in the water, but not like that.
Evidently Bohman built his fish camp a little bit before the water levels went up. Now it looks like a lot of those features are on land. By documenting these landscape features, too, in a geodatabase and then overlaying that with modern imagery, we can see some of the differences in the way the natural landscape itself has changed over time.
There aren't that many fish camps and as I mentioned before, they are pretty ephemeral. They were difficult to see. Now we're going to go to mining features, which are actually even worse. The 1893 gold rush here involved hard rock shaft mining and it was such a difficult method that it was never particularly profitable. However we do have some major locations, including Rainy Lake City, which was a boom town and one of the major settlements here at Voyageurs, which a lot of those residents later went on to live in International Falls, which is just a few miles west. We even have an actual Rainy Lake gold mining district on the National Register but, unfortunately, this up here is the gold mining district. You can see the aerials stop pretty well south of there, so we really only have two mining sites within the park that are actually covered in the aerials. I'm going to talk about both of them at once, here, because they're actually rather mysterious.
Mica Island, as it name implies, is actually a mica mine rather than gold mining. It has those very strange star shaped roads on them. We honestly don't know what that means. That's a new feature for us and we'll have to do some additional research to figure out what the point of that was. On the right hand side we have what was a miner's rest stop. There were some structures built there so that miners on their way north could stop there for the night. You'll see there it's pretty embedded in a logging landscape. Exactly which features are related to mining and which features are related to logging, we are not sure. Clearly there are some data gaps here. It's also interesting the overlapping landscapes really swamp one another out in some cases, especially logging.
Those are our only two mining features that I can show you today, but we have a wealth of logging features. Logging ran from the 1880s until about 1929. It peaked in 1917 and 18 when millions of board feet were taken out of the park. They were here looking for white and red pine and then later they took out the spruce also for pulp. The major company here was Virginia and Rainy Lake. They recorded, we know of, 30 camps recorded in and around Voyageurs National Park. These included railroad camps and sleigh hauls and mills. In this picture here is V&RL Camp #75, which was known as Hoist Bay.
What you're looking at is all the different features that I attributed to logging. The striped areas are places where all the trees had been cut down, or part of the trees. Then the green lines are all the roads and railroads within the park.
I want to draw your attention to Hoist Bay, which was pictured in the earlier slides. It's not a very large area and today we actually have the remains of a resort and that's what those buildings you see in the topo are. This is what it looked like in 1927. Note this area here, those are floating logs corralled to the shore by a chain that's attached to two points. They would float this chain in the water and then they dumped the logs in and then they would float there and be corralled. Then they would let them out slowly in this little area. Then these are hoists and these are railroad trusses coming out into the bay. The railroads would pull up and the hoists would lift logs from the boom onto the train and then the train would take it to places south.
We have all these features visible in 1927, including these piles of logs that are actually visible in a number of historic photographs that I've seen. What's interesting about this is not only the number of buildings and the number of features that we can record, which is especially useful for archeological research, because we can put this data on a GPS unit and go back out and find these footprints, or the remains of them, on the earth. I would also say that a lot of the logging camps look like this. They are just as complex and have just as many buildings as this one. What is also really fun about this one, in particular, is we have that photograph and using GIS we can actually see where it was taken. This photograph here, you'll see this dock. This shorter one - I'm sorry, this railroad, this shorter one, and this dock, I believe are this railroad, this small pullout, and then this dock here. I think that boat house is probably what you're looking at there.
This is of particular interest to me, probably just because I look at this photograph all the time. It's very cool to see where it was taken from because now I can identify specifically these buildings on the shoreline. We can also see that it appears those may have fallen into the water since then, because there's evidently been some erosion. When we walk out there today it's not quite clear where those buildings are now. This is only one example of using GIS to place historical photographs. I did this one fairly quickly just by looking at it, rotating that triangle around until I could match up the right number of features with what I could see in the photograph. There's a great deal of potential there and we have a huge number of historic photographs that we're going to be adding to the database and using to interpret a lot of the features we see here.
If we zoom out from Hoist Bay, I just want to point out how many logging camps there really are in the park. Zooming out just this much to show you a 4 mile view, we have two more logging camps, each containing a large number of features. V&RL #93 and 115. You can also see all the roads and railroads that are crisscrossing and how haphazard and oftentimes how redundant they are.
If we zoom out even further, you can see throughout the park, those are the 1,100 miles of road I was talking about. Also I want to point out, especially down here, you see the roads become so dense they actually begin to look like polygons. Those I've interpreted as the beginning of mechanical removal of the spruce for pulp, because the roads become so close together and so even. They go into lowland vegetation that in the rest of the park has not been logged yet in 1927. We can even see differences in methods. In some cases I've even seen, I think what are different methods of logging between companies. Known locations, the International Lumber Company Camp versus Virginia and Rainy Lake Lumbering Camp, seem to have different road arrangements and different road styles. There's a lot more research to do here, especially with logging.
For now all I've done is selected the logged areas. You can see how much of the park was logged in 1927. It was well over half, I think I calculated something like 67% of the land that was covered by the aerials was actually logged in some way. As an example of a real quick dirty analysis for GIS, I simply overlaid the locations of all the different V&RL camps. Then I attributed the logged areas to each camp based on the distance only. This analysis could be definitely improved. We may be able to use LiDAR and do some real catchment analysis, look at slope and topography and really begin to attribute logged areas to a specific camp. Then using the logging records that I showed earlier and the arrangement of the roads we can begin to almost model how timber was removed from the park. We can begin to question how much logging traffic there was, how quickly things were logged and in what order they were logged. There's a lot of potential research here that's going to be really interesting from the spatial perspective.
In summary we have multiple landscapes. I've talked about Ojibwe and homesteading and fishing and mining and logging, but those are arbitrary. I've displayed here things that I've selected and queried, but the great thing about having this database now is that another researcher can come back and display different things and look at different patterns, because we also have multiple aspects. I talked about doing research, documentation, we see some climate change. There's definite data gaps, but we can also begin to approach analysis as well. The database has a number of different features, I would say. We hope that it's going to become a real research tool. It's grown from just looking at the Ojibwe features to gathering a lot of information that land managers and archeologists can use going forward.
Using GIS this way I wanted to talk about methods in this presentation. Using GIS applies resource management and aids project planning, but for our purposes it really was an efficient way of combining old and new data, because it represents the past in a way that we can use, but it also informs the present. It directs archeological research. It allows us to look at the resource in a new way and to really learn things, combining, for example, the photograph with the features we see on the groundhelps us just learn more about each component. And it provides another glance into the overlapping cultural landscapes. Remember what a jumble it was in the beginning to look at all those landscapes overlapping? Very difficult to see that any other way. It helps us preserve, protect, and interpret this resource.
It looks like I finished maybe a little early, even. I wanted to thank Mary Graves, the chief of resources here and Jeff Richner, retired MWAC archeologist. Both of them know a lot more about this than me. It's “special thanks” because if you have any questions you should ask them, not me. That's my information, if you do have any questions. Thank you very much.
Karen: Drew, that was a great presentation. Thank you very much for that.
Andrew: Thank you.
Karen: Do people have questions for Jeff?
I have a question, Jeff.
Andrew: Sure.
Karen: I'm sorry! Jeff! I was thinking about Jeff Richner. I'm sorry. Do people have questions for Drew? I have a question.
Andrew: Mm-hmm (affirmative)?
Karen: It looks like you have a lot of information that could be applied to the forest succession?
Andrew: Yeah.
Karen: Do you have interest from the natural resource community in the project that you're working on?
Andrew: Yes, we do in fact. We're working on a fire history at the same time. Overlapping a lot of these things with that is going to be a real fun project. We haven't gotten that far yet to working collaboratively, but I was able to begin to see the difference between logging and fire.
Karen: Oh, really?
Andrew: Yeah, the logging features tend to be really straight edged and they tend to end at tracked boundaries based on land ownership and the legality of pulling out timber. Whereas fire leaves a very different kind of pattern and a very different shape. That hasn't been a focus yet, but hopefully we can use some of this data and look at successions and fire and the different ways the trees have regrown and been removed.
There was an automated, the Minnesota Department of Natural Resources, when they geo referenced and mosaic-ed these aerials together, they ran an algorithm that used, sort of like stereoscopy, to look at tree height based on shadows and so forth. They ran an automated algorithm. We started to look at whether that agrees with what I have coded, essentially. Whether I've coded it as logging or fire. It's been inconclusive so far, but that's part of the project, too, for sure.
Karen: Wow, how interesting. What's been happening with the rice harvest there? Have you been able to track that at all in terms of climate change through GIS?
Andrew: That I don't know. I haven't, I didn't see, personally, any evidence of ricing harvest in the aerials, but plotting wetlands and so forth and the size of wetlands would be easily added to the geodatabase. Certainly if not added, at least we can overlay it. I think that's a good idea too. I'm not aware of the GIS. We may have the wetlands that probably do, though. Yeah.
Karen: It might be hard to get from some aerial photos. There might be more ground truthing that's required.
Andrew: Could be.
Karen: What an interesting project. Do other people have questions?
No?
Andrew: I will say it was difficult to describe this as archeology. I'm not sure that it is. Certainly it isn't geophysics, but it interfaces with a lot of different things. It's a tool that I think archeologists can use. That we will be using soon.
Karen: This is an excellent example of the way that archeologists are using some of the technologies that have been more recently developed. I've got another question for you.
Andrew: Sure.
Karen: Can you give us an example of the way that your study has influenced management actions in your park?
Andrew: Well, I don't know that it has yet. I don't know. I've only been here a few months and this project only started last year. We have used it already a little bit to look at the logging camps. We did an inventory this summer.
Karen: Uh-huh.
Andrew: Looking at the aerials and then using that information on the ground with the GPS units, we started to relocate and map some of those to manage that appropriately. Yes, I suppose already even in the short time that we've begun to use this data to manage the resource better. Certainly to interpret it and understand it's location.
Karen: I think it has great implications for the interpretation of the cultural resources in your park.
Andrew: Yeah. I hope so, yeah.
Karen: Do you have anything else that you'd like to add to your presentation?
Andrew: I don't think I do. I'm sure I forgot things and I went through pretty quickly. If anybody has any interest or any questions, please let me know.
Speaker 5: I've got some questions. I put up multiple ones on the text box. Are those not being recognized?
Karen: I'm sorry. I'm not monitoring text boxes. Just speak up everybody who wants to ask a question.
Speaker 5: Okay. I was curious about, you used a stereoscope to identify resources on the ground and then you digitized them in GIS. I was just wondering what your methods were for actually digitizing.
Andrew: For digitizing I was using the Minnesota Department of Natural Resources' mosaic. They stitched these all together for me and then they geo-referenced them. I'm not quite sure how they did that. There's metadata about it. I simply used what they gave me and I overlaid that on the computer screen. When I saw something on the aerials I used their index to navigate to that aerial. Then I squinted a lot at the computer screen, honestly. When I found the features on the two-dimensional aerials, after seeing it in three dimensions I was able to digitize it that way. That was my method there.
Speaker 5: I guess I'm confused on how, you said you were using a stereoscope?
Andrew: Yes.
Speaker 5: Or is that just the nature of the image itself?
Andrew: No I was using a stereoscope which will work on, really, any overlapping image. I had hard copies of the aerials as well. I had, I don't know how many, images in hard copy. I laid two successive images down next to each other and used the stereoscope to view the overlapping images in 3D. Then after identifying a feature that way after just seeing it, then I turned to my computer and digitized it on the aerial. It's not visible unless you know that it's there already. Let me see if there's a better way to describe this. If there's a shadow on two different aerials, in 3D you can tell whether it's cultural or not and if it is, you can digitize it then on the computer screen. Does that make more sense?
Speaker 5: Yeah, yeah.
Andrew: Okay.
Speaker 5: I was just curious about it. If you had to, I don't know, print something out, re-scan it, then digitize that. It seems like your multiple computer methods seems sufficient
Andrew: Yeah, that's what we ended up doing. Just having to look from one to the other. Finding the same spots on a computer screen and then digitizing it that way. Essentially a stereoscope was sort of a ground truthing in this case, because if there was something suspicious that looked like it might have been cultural in the GIS, in the geo-reference copy, then we went to the hard copy, looked at it in 3D, determined whether it was or not and then we were able to digitize it and code it properly. After we could see it better.
Speaker 5: Okay.
Andrew: In the hard copy.
Speaker 5: I have a follow up question if nobody else has any?
Karen: Please, go ahead. I found the question pane so I can go through them, but please ask your question while I'm sorting these out.
Speaker 5: Okay. I was curious about, in your Gold Portage example, I was wondering just in general how much archeological survey on the ground did you have in the study area? And can you use this research to maybe maximize potential results for crews that might go out in the field in the future?
Andrew: The second part of your question, absolutely. We hope to use this to go out and actually find things that we didn't even know were there. We'll direct a lot of research that way. As to how much archeological study has been done at Gold Portage, I don't think very much. A lot of the work on the park has been done just on the shore line. That's a good question. I don't know how much has been done at Gold Portage, specifically.
Absolutely knowing now that there is what looks like a trail through that specific part of Gold Portage, we can definitely direct our archeological research in that direction and see if there's anything there. That will be true of logging camps, too, and other homestead features. In one case there was what appeared to be a cabin that was partially submerged. So a flooded cabin. We suspect, based on aerial photographs and on the historic photographs, too, that may have been a man named Joe Whiteman's cabin. We'd like to go out and look at that too, because I don't believe that place has been visited either, necessarily. In a lot of ways, kind of solving some mysteries here.
Speaker 5: Okay, thanks.
Karen: I found the question pane. Somebody from Yosemite has asked "What is your funding source for this project, Drew?"
Andrew: I think we - We put in for money through PMIS for this one. We started that way. Now the project has evolved somewhat and we're working with a contractor to write some narrative to also add to the GIS on this. To my knowledge this all comes through culture resources funding through PMIS. That true?
Speaker 6: Yes.
Andrew: That is true.
Karen: Okay, thanks. Teri Heightman says, "You discussed various research potentials for this geodatabase. Could you talk a little bit about who will have access? And how this access will be granted?"
Andrew: That is a good question, too. Right now it's still in the development phase and we're building the geodatabase. For it to be really complete, like I said, we brought in the contractors. They're going to add a lot of information to this. They're going to do a lot of the background research and develop some more information thatwe can add to it, particularly in the form of narrative. They'll write essentially stories and this will become essentially a landscape report, but in the form of the spatial database.
When that's complete - we don't have a whole lot of cultural staff here - but certainly the three of us will have access to it. We'll also grant access probably to any other division that would be using it for research purposes. Of course we're going to bring in interpretation as well and do a lot of that. Eventually, and this is a long way off, we want to put it online, some form of it, to make it public so that public researchers can access it and use this information. That's going to require a lot of thought, though, as to the sensitivity of the data.
Karen: Mm (affirmative). Yeah.
Andrew: The simple answer is we're going to ramp up, hopefully to everyone eventually will have access to it. As to answer how it will be granted, I honestly don't know that.
Karen: All right. Is Teri still on? Is she gone? I hope she didn't leave before we got to answer her question. And Sheldon has a comment. He says, "Yeah. It's archeology. It's very interesting."
Andrew: That's good news. Sometimes, like I said, I sit there and I use my ghost trap on it and that's all I ever feel like I do. It will be exciting when they get out in the summer and find some of these things and do some actual ground truthing.
Karen: Yeah, I bet the ground truthing part of it is going to be very interesting. Does anyone have any additional questions?
Andrew: Oh, there's a question here. Go ahead.
Speaker 7: Have you shared some of this to the Bois Forte people and if so, are they interested in their history here?
Andrew: The question is have we shared some of this with the Bois Forte people and are they interested in their history here. I don't know. Mary Graves, the chief of resources, does the consultation here. I think we have used a lot of this information in doing consultations for some of the projects we've done so far this year. I don't believe it's been shared directly. Although I'm sure they'll be glad to see it. I'm not familiar enough with that yet to know for sure.
Mary: They were very [crosstalk 00:56:25].
Andrew: Mary's here. She said yes, they are very interested and yes we have shared and will be sharing more. Thank you, Mary.
Karen: Are there any more questions or comments? I want to wish everyone a very happy Thanksgiving. Stay safe and we will see you the first week in December. Drew, thanks again for a very interesting talk.
Andrew: Thank you.
Karen: Good-bye, everyone.
Andrew: Good-bye.
Description
Andrew Labounty, 11/21/2013, ArcheoThursday
Duration
49 minutes, 49 seconds
Credit
NPS
Date Created
11/21/2013
Copyright and Usage Info