Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Dynamic Communities on the Mesa Verde Cuesta with Kelsey Reese - Ep 118
194 Plays5 years ago
Paul interviews Kelsey Reese about her research at Mesa Verda and a recent paper in American Antiquity. Kelsey is a Ph.D. Candidate at the University of Notre Dame and is studying the long-term effects of sustained climate change on contemporary subsistence-based farming communities. This episode focuses that research on Mesa Verde in the American Southwest.
Links
- Dynamic Communities on the Mesa Verde Cuesta by Kelsey Reese
- American Antiquity, 84(4), 2019, pp. 728-747
- doi: 10.1017/aaq.2019.74
Contact
- Chris Webster
- Twitter: @archeowebby
- Email: chris@archaeologypodcastnetwork.com
- Paul Zimmerman
- Twitter: @lugal
- Email: paul@lugal.com
Affiliates
Recommended
Transcript
Introduction and Sponsorship
00:00:00
Speaker
We're excited to announce that our very own podcasting platform, Zencaster, has become a new sponsor to the show. Check out the podcast discount link in our show notes and stay tuned for why we love using Zen for the podcast.
Podcast Introduction and Guest Overview
00:00:19
Speaker
Hi, welcome to The Architect podcast, episode 118. This is Paul Zimmerman. Chris could not be here today. He's stuck in transit somewhere. So he'll hopefully be back next time. I was going to say I'm flying solo, but I'm absolutely not flying solo because we have a nice interview lined up for you today with Kelsey Reese.
Dynamic Communities Paper Discussion
00:00:39
Speaker
Kelsey is a PhD candidate in anthropology at Notre Dame and recently published a paper in American Antiquity alongside her co-authors Donna Galwacke and Timothy Kohler entitled Dynamic Communities on the Mesa Verde Cuesta. We're going to discuss that and some of Kelsey's ideas about tech and her use of it in archaeology. But before we start, actually, Kelsey, what I want to say is
00:01:04
Speaker
Two weeks ago, we interviewed Sean Field and we had a very nice discussion about the GIS work that he was doing in and around Chaco Canyon.
Community Interaction Challenges in Mesa Verde
00:01:12
Speaker
Afterwards, off air, we were having a little chat and we said to him, if you know anybody that you think would be good to come on the show and talk to us, please let us know. He said, oh yeah, absolutely, you have to talk to Kelsey Reese.
00:01:26
Speaker
We said, hey, you know, she's already signed up. This is perfect. So I guess we're trolling Notre Dame this this month. But Kelsey, how are you doing? I'm doing great. And thanks to Sean Field. You know, that's what good friends are for. To give you a good word. But it's great joining you today and thank you for having me.
00:01:46
Speaker
We're really happy to have you. This is an interesting article that you wrote. Do you want to give us an overview for our listeners what the article is about? And then we can go into some of the details about how you did your work and what some of your findings were.
00:01:59
Speaker
Yeah, this article is basically a product of my master's thesis that I completed in 2014 at Washington State University. The idea of it was to basically try to find an emergent extent of community interaction in the northern US Southwest by using changes in spatial patterns of households through time.
00:02:24
Speaker
And so this whole idea kind of came about as we were all sitting doing fieldwork out in Mesa Verde National Park. And it would take us hours to get to our sites in the back country. And we'd be sitting there on our breaks and thinking about the realities of living in that landscape.
00:02:44
Speaker
and how difficult it really would be to interact with your neighbors or other households. If you were living in those spaces and how you might try to counteract that by, you would choose to live closer to the people that you really wanted to interact with on a regular basis, because otherwise it would be pretty, not impossible, but there would certainly be large impediments to that interaction.
00:03:09
Speaker
Just to point out for those not familiar with the Mesa Verde quest itself, it's this series of mesa tops and canyons.
00:03:19
Speaker
and they vary greatly in elevation. So you basically have all these finger mesa tops running north-south, and then they're all bordered by deep canyons or valleys. And so if you were living on one mesa top, it would take a lot of effort and a lot of investment to regularly interact with the people on a neighboring mesa top. As the crow flies, they may only be
00:03:47
Speaker
a thousand meters away, but it might take you twice as long to actually, you know, go visit those people.
Statistical Models and Data Analysis Techniques
00:03:55
Speaker
So the idea was that if we could assume that those people were actively choosing to live closer to each other if they wanted to interact, then maybe we could identify kind of those self delineating extents that they were willing to travel through time to maintain those those relationships.
00:04:12
Speaker
Right, so practically speaking, if you're in community A and you want to get to community B, you might have to go down a cliff to cross a valley to climb another cliff to get to that community. Is that basically how it would work? Yes, absolutely. That is a very clear way of saying it. So you'd more likely be interacting frequently with people who also lived on the same mesa top. Right, exactly.
00:04:38
Speaker
and then you have developed a number of statistical computer techniques to try to model that kind of behavior.
00:04:46
Speaker
Right. So the whole idea was trying to, again, come up with this emergent extent. And in doing so, trying to remove my bias as a researcher that might just draw circles around clusters of what I see as maybe they might be interacting or whatever. And so it was trying to remove that bias as much as possible to produce extents that were just purely a result of
00:05:16
Speaker
of where those data points were landing through time. Really, this type of analysis was really possible because of a larger project that I was part of called the Village Ecodynamics Project. They worked for
00:05:33
Speaker
years and years and years to collect site data all across the northern U.S. Southwest in the Mesa Verde region. And then a couple of researchers, so originally there's an article also in American Antiquity, Oortman et al. 2007, which produced a Bayesian analysis to date sites based on their surface ceramics.
00:05:58
Speaker
and then a follow-up to that, which is Schwint et al. in 2016. And those two papers were actually, they created the dataset that I was able to use that dated households to very specific time periods. We have one time span that's, I think, only 20 years long. Most of them are about 40 years long. And so we were able to really parse out very specific
00:06:27
Speaker
movements, we just had a really high resolution time data to come out with these patterns through time. So the data itself really lent itself to this type of analysis. But as part of that, in using that data, basically to come up with these clusters or these communities or potential communities, I created a null model, which basically put a bunch of random points on the landscape and said with the idea of, okay,
00:06:54
Speaker
If somebody was moving into this landscape completely at random and they weren't thinking about where their neighbors were living or whatever, what would that look like?
00:07:04
Speaker
And then of course we can look at the difference between that and the actual spatial patterns that we see in households. And they're different, which we expect to see, right? We were like, yeah, of course they're different because people do consider other things when they consider everything when they move into a certain space. But we were able to basically quantify that difference, quantify that social, environmental, political,
00:07:30
Speaker
whatever those influences are, that's what changes the distribution that we actually see from the null model. So rather than an even random distribution across landscape, you have some kind of clustering and you're trying to get at the reasons for those clustering.
00:07:45
Speaker
Right, exactly. Just from the null model, then based on those results, I basically ran a cluster analysis that it's called affinity propagation cluster analysis, which is this really specific type. And it was chosen because you actually do not have to put in an input telling the algorithm how many clusters you want.
00:08:08
Speaker
And so again, it's this whole idea of like trying to take the researchers biases out of the analysis itself.
Community and Household Structure Analysis
00:08:15
Speaker
And so that cluster analysis basically goes out and looks at the data and finds the ideal number of clusters for the data that you give it. And so then that would be how many communities end up popping out at the end. Let me, since you use the word communities here, let me back us up a little bit and I'm going to step us back up to that term. How big is this area that you're looking at?
00:08:37
Speaker
So Mesa Verde National Park, oh gosh, this is a good pop quiz. Sorry. I think it's about 210 square kilometers. Okay. And about how many different households are you looking at there? It varies through time. We go from about, let's see, at the lowest, I think we have about 70 households. And so we're using the term household, which is
00:09:08
Speaker
very mushy in southwest archaeology. So it's this idea that it's a family unit, but there's a lot of different numbers out there of what a household actually represents. So in the paper, we kind of average out that range of ideas. So the lowest end is 3.3 people per household.
00:09:32
Speaker
which is a result of some modeling of one of my co-authors, Tim Kohler. And then the higher range is seven people for that household, which is from other literature and stuff. And so we average that out to create like what the household is itself, what we think about it. Yeah. So the lower end is about 60, 70 households. And then at the higher end, we get over 400, I think, 460.
00:09:59
Speaker
And then if I recall correctly, a household is primarily architecturally based, right? That's what's defining it, or is there other kinds of sites that help you define what a household is? So I try to make the distinction in the paper and I'm careful about how I use the term household versus residence. The residence is basically the physical architecture on the ground that has X many number of rooms, and that will tell you
00:10:29
Speaker
how many households likely occupied that space. And then the household being that social aspect representing actual people. And I have probably conflated that this entire time so far in this conversation. But yeah, really, it's all just the number of residences or the number of households is based on the number of residences that we see in the archaeological record. Gotcha.
00:10:52
Speaker
And then the time period that we're looking at here, you're looking at between 600 and 1200 roughly. That's 612.80. Yes. And that, yeah, I should have specified. I normally work in BCE, so 612. I guess the order that I said it would have made sense. And that comprises a number of different, you know, my specialty is not the Southwest, a number of different cultures. Is that the term still used between Baskamaker and Pueblan?
00:11:19
Speaker
Right. So they're all ancestral Pueblo. It's all the same group of people. It's more kind of just defining like a typical set of characteristics. So, you know, are you basket maker three? You're getting early pottery. People are typically living in subterranean pit structures. And then you have your transition into Pueblo. So it's called the pit to Pueblo transition.
00:11:45
Speaker
where people are starting to create masonry structures above the ground, but then they still have pit structures that they're using associated with that. And then the Pueblo II period, you're getting larger, more aggregated communities, you're getting kind of large, more substantial architecture using larger masonry stones.
00:12:07
Speaker
Uh, and, and you start seeing a more formalization of your pit structures. And then that's when, when they start kind of turning or being referred to as Kibos, which have, uh, you know, ritual, a ritual connotation. Um, so I, I typically, uh, stick with the model of just referring to as pit structures. So I'm not putting any meaning on there, um, that, that may or may not have existed, but the, uh, and then in the, into the Pueblo three period, you're getting.
00:12:37
Speaker
population, you're getting even more aggregated communities, larger structures, and then of course the move into the alcove sites, which is where you get your classic cliff dwellings of Mesa Verde National Park. So it's
00:12:56
Speaker
It's all the same people. It's just, you know, they're going through a lot of changes in development and, you know, figuring out different ways to kind of live and interact in the same space. All right. So the residences are indicators of the households, but they're not the households themselves. And the shapes of the residences changes through this 600 years span of time.
GIS and Spatial Interaction Studies
00:13:22
Speaker
Yes.
00:13:23
Speaker
Okay, that's a lot to try to tease out.
00:13:27
Speaker
You are using these to then look at communities, which you are looking at changes then of time through time of communities. What do you define then as a community in this project? Right. So we define a community based on a definition that has been used for a very long time in Southwest archeology and is initially based on the definition in the
00:13:54
Speaker
The book is called Social Structure, George Murdoch. And so it's this idea that a community is a group of people that you are interacting with on a daily basis. And if not a daily basis, you have the opportunity to interact with on a daily basis.
00:14:13
Speaker
If you're going to be interacting with people very regularly without the modern wonders of cars and planes, you're really just left with your feet. And so that's why it has such a large impact on how these households of people are organizing themselves.
00:14:32
Speaker
So as you hinted earlier, it's easier to get to be in regular contact with somebody that's on the same mesa top as you versus somebody on the next mesa over. And so that's part of how you define what a community is. And you're doing that through computerized GIS techniques, such as what was done on our last interview, what Sean was doing with cost surfaces.
00:14:59
Speaker
The least cost analysis that we did in Mesa Verde National Park, like I said, it's built of this series of mesotops and canyons. The cost surface is going to have a huge impact on the paths that people take to interact with one another. And so, like I said earlier, you can have two households that are maybe only a thousand meters apart, but it might take much longer than that for those people to actually interact with one another.
00:15:26
Speaker
So, in the clusters of households that we get in our results, in a lot of areas, you'll see that the households actually cluster right on top of mesa tops. Now, there are plenty of instances where that's not the case, and you actually get households that are clustering across canyons or valleys, which is kind of an interesting result, but
00:15:53
Speaker
really, I think it's a product of the data that we have. So the spatial data that we have in the household data is limited to the borders of Mesa Verde National Park. Now, while that's a large area, it only is about one third of the entire Mesa Verde landform or the Mesa Verde Cuesta landform. So I think a lot of what when we see those types of effects, it's kind of like an edge effect. You know, if we had data of
00:16:22
Speaker
households across the entire space, you probably wouldn't get that happening as much. So why don't we go to break and when we come back, we can dive in a little bit more to some of the tools that you use and some of the techniques that you applied them to try to get these clusters and these communities and what that was showing you about the change in how people lived in that landscape through time.
00:16:43
Speaker
Chris Webster here for the Archaeology Podcast Network. We strive for high quality interviews and content so you can find information on any topic in archaeology from around the world. One way we do that is by recording interviews with our hosts and guests located in many parts of the world all at once. We do that through the use of Zencaster. That's Z-E-N-C-A-S-T-R. Zencaster allows us to record high quality audio with no stress on the guest. Just send them a link to click on and that's it. Zencaster does the rest. They even do automatic transcriptions.
00:17:11
Speaker
Check out the link in the show notes for 30% off your first three months or go to zencastr.com and use the code ARCHEOTECH. That's A-R-C-H-A-E-O-T-E-C-H.
00:17:25
Speaker
Looking to expand your knowledge of x-rays and imaging in the archaeology field? Then check out an introduction to paleoradiography, a short online course offering professional training for archaeologists and affiliated disciplines. Created by archaeologist, radiographer, and lecturer James Elliott, the content of this course is based upon his research and teaching experience in higher education. It is approved by the Chartered Institute for Archaeologists as four hours of training. That's in the UK for those of you that don't know.
00:17:50
Speaker
So, don't miss out on this exciting opportunity for professional and personal development. For more information on pricing and core structure, visit paleoimaging.com. That's P-A-L-E-O, imaging.com. And look for the link in the show notes to this episode.
00:18:06
Speaker
Hi, welcome back to The Architect Podcast, episode 118.
Computer Modeling in Archaeology
00:18:10
Speaker
We're continuing our discussion with Kelsey Reese about work that she did in the Mesa Vertequesta. Kelsey, right before break, I kind of teed up a strange question for you about the tools and about how we're modeling, how you're modeling, pardon me, the communities. And so I was just wondering, why do you think,
00:18:31
Speaker
And how do you think computer modeling is appropriate to explore human behavior and social relationships? Yes. So I think we have a really great opportunity with the computational power that we have today to kind of tease out different things in the past. And I think
00:18:50
Speaker
that computer modeling you have to think of on a very large scale. It's very zoomed out. It's difficult to pick out or explain very individual behavior with those tools. And they're certainly going to miss the individuals that chose to take a different figurative path. But the repeated social behavior
00:19:14
Speaker
that we see with people, it leaves a physical record, right? Like, we actually see, you know, the buildings that are produced, we see, in, you know, cases, the paths that were, you know, physically walked across. And so that that repeated behavior leaves a measurable physical space that we can identify and use to think about
00:19:40
Speaker
social relationships and human interaction and the human experience in those spaces. So with this paper, a lot of it is based on least cost analysis. And there's plenty of critiques of least cost analysis and what it includes and what it doesn't include, but it still accounts for general human universals of like, it's harder to go down a cliff than it is to walk up
00:20:10
Speaker
a gentle hill, you know, and, and so we can we can start to kind of tease out those behaviors at a large scale. And then and then we can work on, you know, adding in other things that that, you know, might have affected how people
00:20:28
Speaker
moved across their landscapes, like, you know, like thinking about other social relationships, or maybe, I don't know, enemy territory, or something like that, like places that you would either gravitate to or avoid, depending on what is actually going on in the landscape. And that, of course, is, you know, those things are harder for us to see, archaeologically, but if we kind of get a basis down of kind of how people are moving, and that
00:20:58
Speaker
how they're moving about their social space, we might be able to start teasing out those more complicated aspects of that human experience. What specific aspects in this large-scale analysis do you think really reflect human experience? I mean, travel time is the one that you've been talking about mostly, and you're talking about various social aspects. Are there certain things that you wish that you could look at that you haven't been able to, or certain things that you think you can with the right data?
00:21:25
Speaker
Are there certain things that I wish I could look at? Everything. The answer is yes. Yes, everything. I'll take anything that I can get. I think what geospatial analyses are really starting to do in a very strong way is we're starting to place the human back into the landscape that we're assessing.
00:21:48
Speaker
And there's been critiques of GIS for decades now about this separation of the human experience and then your two-dimensional GIS interface that is no real reflection of the landscape that you're studying. And I think the idea of being able to account for the interaction
00:22:15
Speaker
between humans and their landscape on a regular basis and even on a larger basis. So things like a phenomenology approach where you're thinking of larger scales of the horizon and what different peaks might mean and the accessibility to those
00:22:36
Speaker
special places are all thinking about how an individual is actually experiencing the space that they're in. And I think that those analyses become much more powerful in their explanatory power than just kind of measuring distances or simply mapping things out.
Reproducible Research and Open Source Tools
00:22:55
Speaker
Right. Yeah, so I think these really large-scale analyses that are being done more recently are really exciting because you're starting to kind of add in that socialized aspect or trying to really tease out the experience of an individual as they're standing in a space, looking around and deciding
00:23:16
Speaker
Like, do I need to go grab water from this spot or am I going to go talk to my neighbor and what the implications are then for the archaeological record and what we ultimately see based on those decisions being made? Right, so it's not just kind of blind mapping of automaton on
00:23:37
Speaker
on a landscape given certain parameters, but it's trying to get at the ways that people actually interact with those different parameters as humans, with our various motivations and demotivations that we have. Yeah, and certainly doing these large models that we have of putting in inputs and looking for protein need. And like I said earlier, I was part of the Billie G. Eukaryote Dynamics Project, and they created an incredible
00:24:06
Speaker
large, complicated model that looked at human behavior in the Mesa Verde region. And I think all of that is certainly necessary and has done a really good job at answering a lot of questions that we've had. But now, because that is out there,
00:24:24
Speaker
at least in my area, then we can start picking apart little pieces of it and asking very specific questions because we already have that baseline idea of what's going on in the environment and with people aggregating and disaggregating through time.
00:24:42
Speaker
Let me switch gears on you a little bit here and ask you, you were mentioning GIS. I don't remember actually seeing the term GIS in the article. If it was, I missed it or forgot it. It's not foregrounded as these are the tools I'm using, but obviously it
00:24:58
Speaker
played some part of the tools that you were using. And since this is after all the Archiotech podcast and I work in IT, what tools for GIS or GIS related tools were you using for mapping and statistics? So I actually did the entire project in R, which is probably why I never actually mentioned GIS, which I hadn't even thought of until you just brought it up. But yeah, so everything's done in R. My whole analysis is all
00:25:24
Speaker
scripted and available. I put a reproducible script up on my GitHub. So if anybody has a data set that they think that this might work for, I welcome them to copy that code and try to make it work for their area. So I think that creating reproducible work and creating it using tools that are
00:25:52
Speaker
freely and openly accessible to people, to researchers around the world, whether they're affiliated with a university or doing research on their own is incredibly important. So everything that I do, I try to do in free and open source software and
00:26:07
Speaker
and then try to make that as available, or at least as transparent. Even if I'm not posting reproducible code, I'll still put my code up there to make it as transparent as possible of what I'm actually doing to come up with the numbers that I ultimately present. That's great. I've been at IT for so long that
00:26:28
Speaker
I'm really pleased that so many people have that same sort of attitude about the tools that they use, the code that they write, the algorithms they use, whatever they devise, and that they want other people to see them, use them, fix them, make them better, and so on and so forth. Do you mind, we'll be able to link to that GitHub account of yours in the show notes? Yeah, absolutely.
00:26:51
Speaker
Oh, good, good, good, good. I'm going to want to take a look at that because that sounds interesting, even though R kind of makes my skin crawl, but it's mighty useful to all. As programming languages go, that's one of the ones that I'm telling every time I look at it, I'm just like, this is backwards. Why do you do it like that?
Geographical Mapping and Data Momentizing
00:27:06
Speaker
You know, it's so funny. It's really the...
00:27:09
Speaker
Uh, it's the first scripting language that I really learned. I used to clean Java code for my dad when I was little. Um, so I guess that was technically my first introduction, but the, um, uh, yeah, one of my good friends, uh, Dr. Kyle.
00:27:25
Speaker
we worked in the same lab together during my master's degree and he kind of set me up and got me going and I just had this moment where I realized, oh my goodness, I can do anything I want with this interface. Any type of analysis, anything that you want to do, you can do it in R or you can find a package that will do it for you in R and it's a really powerful tool and I would highly recommend anybody using it despite what you're saying.
00:27:54
Speaker
No, no, no, no, absolutely. I recommend everybody take a look at it. I've tried to do myself a bunch of times. I just, I don't have a data that are really amenable to it. You know, if I was, if I was doing archeology full time, then yes, I'm sure I would find a great use of it.
00:28:11
Speaker
because it really is powerful. So thanks for sharing that. But that's interesting. There is kind of an RGIS divide, not really a divide, because they dip toes in each other's waters. Same like there's also Python and R. I see that as kind of a friendly competition.
00:28:31
Speaker
Yeah, so it's different communities, I'm sorry to use that term, of archaeological tech geeks, I guess. Another technical question I had is that in your article you cite, and you did here on the interview today, different prior studies that you're using for your population estimates.
00:28:52
Speaker
And definitely anybody that's interested in those things should read your article to get those references. But since you're plotting these things on a map, what map sources did you have? What data sources did you have for your mapping?
00:29:07
Speaker
So, I mean, the maps themselves, the landscapes are all just downloaded from USGS, the 10-meter DEM. The site data itself is from the VEP, the Biology Dynamics Project. All of that information was basically, it was kind of like a big data pool from the
00:29:29
Speaker
Colorado site database. And so the VEP database is actually much larger than what I'm using. I probably use, I don't know, maybe I don't even know how many total sites through time, but it's a really small subset of, I think there's like well over 10,000 sites in that database that are for the entire Montezuma County area. And I just use the sites that are just on top of the uplift
00:29:58
Speaker
in Mesa Verde National Park, so it's a really small number. So that's where all that data came from. Great. One other question, and this is back more towards your article. You are dealing with a fairly long span of time and with
00:30:18
Speaker
households that are defined by buildings that have variable lifespans. So you talk about in the article momentizing your data that is dividing into snapshots, I guess, of certain times so that you can seriate it and run your analyses.
00:30:40
Speaker
on not everything in a 150 year time span, but things that would have been contemporaneous in that 150 year time span. That I thought was a very interesting part of the article and also a lot of it went over my head. Do you want to briefly describe how you did that? Yeah. Momentizing households is the thing that I get asked about most in relation to this article. It's this idea that
00:31:07
Speaker
We already have really high resolution time periods in this dataset, but those time periods still are not small enough to represent the individual use life of residential architecture.
00:31:25
Speaker
And so, for example, in our first time period, it goes from 8,600 to 725. It's 125 years. It's our longest time period. But during that time period, the average residence was probably only used for about eight years. So, if you assume that
00:31:48
Speaker
all of the sites that we have that are dated to the 600 to 725 time period were occupied contemporaneously, we would have over 1,000 households for that time period.
00:32:05
Speaker
But it's just not the case. There were not 1,000 households living on the questa during that time period. It was in fact the least populated time period. And we just know that from general population estimates of the Mesa Verde region.
00:32:25
Speaker
So we know that that's not right. So momentizing the households is this idea to try to make the total number of households occupied contemporaneously more representative of what was most likely to be the case. And so it's just this idea of taking, okay, so if we have a time period that's 125 years long and our average residents lasted eight years,
00:32:51
Speaker
we divide 8 by 125, and then multiply that with the total number of households we have to that time period. And so for our first time period, the 600 to 725, that would leave us with 62 households that were likely occupied contemporaneously versus about 1,000. And so to then run the analysis on that,
00:33:18
Speaker
I basically did a random selection. And so I had the list of all 975 residences and said, okay, randomly select 62 and then run our cluster analysis on it. And so it's not a perfect solution because you could have had really dense clusters of households. Maybe there were 62 households being occupied, but they were all occupied.
00:33:45
Speaker
on one mesa at one time and then they move to the next or whatever through that 125 years, but it was the best solution that I could think of on this scale to try to best represent what was going on. The residences have a shorter use life than their corresponding time period from 8,600 up until 1,100.
00:34:14
Speaker
So, and that's more than half the time in this study. So, for all of those periods where there's this discrepancy, I repeated the process five times each and then averaged those results together to just try to get a most representative sample that I could to give us the best idea of what was going on at that time.
00:34:39
Speaker
Right. So five times five different random samples of that, for example, those 62.
00:34:45
Speaker
Yeah, and then of course it created five different groups of communities that were all different because they all were based on different residences. And so yeah, actually every single result from all of those runs is actually posted in the supplemental material online.
Innovative Uses of Drone Technology
00:35:06
Speaker
So you can actually see the differences between each of those five results and how they work for the different time periods.
00:35:13
Speaker
and run the maps, I guess, and stats again yourself with what you've posted on GitHub, I take, right? Yeah, feel free. But if I did something wrong, please let me know. I wouldn't be able to access it out, I'm sure. Why don't we take a break, and then we'll come back for a brief wrap up. And yeah, see you in a minute.
00:35:37
Speaker
You may have heard my pitch from membership. It's a great idea and really helps out. However, you can also support us by picking up a fun t-shirt, sticker, or something from a large selection of items from our tea public store. Head over to arcpodnet.com slash shop for a link. That's arcpodnet.com slash shop to pick up some fun swag and support the show.
00:35:57
Speaker
Hi, welcome back to The Architect podcast, episode 118. We're having a great interview with Kelsey Reese. And because Chris isn't here to talk about app of the day, I decided that we're going to have a special extra third segment here to wrap up our discussion. So Kelsey.
00:36:13
Speaker
We were talking about some of the tools that you use, R in particular, and there's a whole crop of different tools that people are using nowadays for actually getting data for archaeological purposes, for landscape studies in particular. Do you see, is there anything that you have used or would like to be using drones, for example, because I can't go through an episode without machine drones, you're welcome, Chris, that are going to be of use to you in your further studies?
00:36:40
Speaker
Yeah, so I think the availability of drones nowadays is phenomenal. I mean, the amount of technology that's packed into these increasingly smaller and increasingly cheaper little things that you can throw into your backpack is incredible. And I think that drones, as far as collecting landscape data, are going to really revolutionize how we can think of spaces.
00:37:07
Speaker
And I say that because, as I was talking earlier, I was talking about how these large-scale geospatial analyses are starting to really incorporate humans, the idea of humans in these spaces, and that I think that that's a good direction that we're going in. And I think drones are kind of how we get there.
00:37:29
Speaker
And I think what they provide is the ability to collect data on a human scale. So we're able to answer questions about the human experience because we're able to capture data on that scale. The types of terrain models that we're going to get from drone, even just photogrammetry.
00:37:50
Speaker
are way more detailed and you can learn a lot more from them than you can the 10 meter DEM that you're pulling down from the USGS. So I think that right now we're very excited about the technology as a discipline and we're doing a lot of things that are
00:38:14
Speaker
We're producing a lot of publications that are strictly about producing data. How do we make data from drones? And I think that the really powerful next step is going to say,
00:38:27
Speaker
Okay, but what can we do with that data that we're producing? And so it's starting to run kind of the same analyses that we're used to. You can do least cost analysis, you can do hydrography, visibility, other site level studies that you can get much more resolution
00:38:45
Speaker
And I don't necessarily mean like pixels per centimeter resolution, but the idea is that it's much more at that human level scale, where if somebody was standing on that landscape, that would be their experience in those spaces versus something that is a much lower resolution.
00:39:03
Speaker
And so I think drones are really a great tool that are going to gap or fill that gap and provide us with a whole host of information and ways to understand archaeological landscapes that we just haven't been used to so far. And it's going to be really exciting to see what comes from it.
00:39:24
Speaker
Yeah, well, there's certainly been a crop of a lot of very interesting studies that are showing, if nothing else, just new ways of mapping archaeological sites with drone data. I'm sure you're right that it is going to allow us to piece together various kinds of different kinds of data and also data of various scales that we haven't always had access to. So we're definitely architect, Chris and I are both definitely on team drone.
Community Dynamics in Historical Context
00:39:55
Speaker
You know, before I go too far afield here, your study found, the one we've been discussing so far, found that there's a fundamental shift in those communities and how the societies are organized in the 13th century AD. Do you want to describe how you found that, how it showed up in your models and what you believe that reflects? So what I think that you're referring to is the area per community.
00:40:22
Speaker
So, one result that we had that popped out that was completely unexpected was this
00:40:29
Speaker
sort of equilibrium of area per household within a community. So the calculation is basically we have our cluster of sites that are grouped into a community from our cluster analysis. And I would just put what's called a convex hull. So straight lines around the outermost points that surrounds all the other points.
00:40:58
Speaker
determine the area of that and then divide it by the number of households, right? So you get area per household within a community.
00:41:05
Speaker
And so what we actually see is that area, it ends up settling right around the amount of area per household that you need to farm to support one household for each year. And as populations increase or decrease on the Cuesta, that metric either stays the same or it gets a little screwy for one time period, but then it returns to that same equilibrium point.
00:41:34
Speaker
And then kind of that changing point, the point of no return almost in the 1200s, right before the area was depopulated, we had the most number of people on the Cuesta at any time in the previous 700 years and people are having the least amount of area for farming within their communities. And so, you know, as the,
00:42:04
Speaker
as the environment wasn't the worst it's ever been, but it was certainly unstable and very volatile. And then you had more people coming into the Cuesta. Even as the population in the Mesa Verde region as a whole was decreasing, the population on the Mesa Verde Cuesta was still increasing. So you're still getting people coming on there even in the worst of years. And they're starting to see a huge drop in the amount of land that they have available to provide
00:42:33
Speaker
their annual food need. So it was kind of this really interesting little metric that came out that I wasn't expecting to see. I just thought it would be interesting to look at that trend through time, but it really does kind of create an interesting spot where you can see where people are really kind of unsettled and getting unhappy in their area right before the quest itself is depopulated.
00:43:00
Speaker
Interesting. I found that actually quite interesting. I'm using the word again. It's like stressing on an already stressed system and people stop, like you said, stop reverting to that equilibrium and then it all just falls apart like one pushed too many.
00:43:17
Speaker
Yeah. And, you know, earlier in the time period, you know, if there's more people coming onto the quest, you know, you could kind of spread out in your communities or the less people you could squeeze in. And they were they were able to kind of maintain that. But I think it just got to the point where you're having too many people packed into the same little space. And it just became unrealistic to to maintain that that annual food subsistence.
Dissertation Focus and Future Research
00:43:44
Speaker
Well, to round out our conversation today, thanks for taking your time to talk to us about this. I thought it was a very interesting use of tech to look at how people in the past lived. You said before this was basically your MA research and now you're working on your PhD. What is your dissertation research on?
00:44:05
Speaker
Yeah, so this whole time we've been talking about the Mesa Verde questa, the landform itself. But my dissertation research actually is focusing on what we call the Mesa Verde north escarpment. So on the northern edge of the questa is a 600 meter cliff face and everything basically north of that that slopes down into the valley to the north is the escarpment.
00:44:32
Speaker
And so it's this really interesting area that has generally thought that there's not too much going on. The general idea has been that it wasn't really heavily occupied. There wasn't too much going on in that space. And it's really just because when you look at a plot of sites across the Mesa Verde region, it's just generally empty.
00:44:57
Speaker
And that idea can be totally explained away about how it's a weird area. It's very vertical. It's north facing versus the south facing slopes of the Cuesta, where you get longer sunlight, warmer temperatures. So it kind of has a lot working against it. But if you talk to anybody that's ever worked out there and did any survey, they would tell you that it was some of the most densely occupied area that you've ever
00:45:25
Speaker
seen and so what I've been doing over the past three summers is going out and I'm working with the BLM so it's all it's outside the national park it's all BLM land and so the local field office there and we've been doing pedestrian survey.
00:45:44
Speaker
And so, kind of filling in that knowledge gap of what might be out there. And we found that it's actually, it's got some huge communities going on and it's really quite exciting. We've found four pretty aggregated, very dense communities across the front of the escarpment so far.
00:46:09
Speaker
and they're all evenly spaced. They all have a P2 occupation component, so about 8,900 to 1,100. Most of them go into the P3 period, so 1,100 to 1,300. But they're all basically contemporaneous. They're all basically evenly spaced from each other. And there's a lot of
00:46:32
Speaker
a lot of evidence to show that they're really harnessing the water runoff that's coming off of the front of the escarpment. So if we think of, well, it's north facing, so it might be a tough place to live because you're not getting as long of growing seasons. Well, that might be true, but you're also getting all of the water runoff off of the entire cliff face of the Mesa Verde questa. So you might be able to have a little bit of an advantage that way by channeling that water.
00:47:00
Speaker
And so there's lots of control features out there. Like I said, there's pretty dense communities. And yeah, so that's kind of what I've been working on over the past few years, just for field work. And then the idea is that all of those, all of the sites that whole survey is kind of going to get funneled into answering the question of looking at the long term effects of climate change over subsistence farming communities.
00:47:27
Speaker
And so the escarpment dataset will basically provide a 700-year longitudinal study of how people are responding to potentially volatile environment when their food is completely subsistence-based communities. So that's what I'm working on now.
00:47:45
Speaker
I assume you'll be using similar toolset going back to R to help model some of these. Yeah, it'll all be done in R. How'd you guess?
Wrap-Up and Contact Information
00:47:55
Speaker
Shot in the dark. Yeah. Well, not going to take up any more of your time today, but Kelsey, thanks so much for coming on. Very interesting discussion. I learned a lot today. Best luck with that dissertation research. We look forward to seeing what you do with that. Excellent. Thank you so much for having me. Thank you. Take care.
00:48:18
Speaker
Thanks for listening to the Archaeotech Podcast. Links to items mentioned on the show are in the show notes at www.archpodnet.com slash archaeotech. Contact us at chrisatarchaeologypodcastnetwork.com and paulatlugol.com. Support the show by becoming a member at archpodnet.com slash members. The music is a song called Off Road and is licensed free from Apple. Thanks for listening.
00:48:44
Speaker
This show is produced and recorded by the Archaeology Podcast Network, Chris Webster and Tristan Boyle in Reno, Nevada at the Reno Collective. This has been a presentation of the Archaeology Podcast Network. Visit us on the web for show notes and other podcasts at www.archpodnet.com. Contact us at chrisatarchaeologypodcastnetwork.com.
00:49:05
Speaker
Thanks again for listening to this episode and for supporting the Archaeology Podcast Network. If you want these shows to keep going, consider becoming a member for just $7.99 US dollars a month. That's cheaper than a venti quad eggnog latte. Go to archpodnet.com slash members for more info.