I'm happy to announce that my recent paper on the minimum viable population (MVP) size of hunter-gatherer populations is now officially published and available for download from the Journal of Artificial Societies and Social Simulation (JASSS).  JASSS is open access, meaning anyone can access any paper at any time. It's the way to go, and I wish all journals could figure out how to play nicely with the public.

The issue of how many hunter-gatherers it takes to form a population that can sustain itself over a long period of time is relevant to understanding several issues, including: (1) how hunter-gatherer societies colonized empty landscapes; (2) how/why hunter-gatherer societies take on the forms that they do in different environments; (3) how/why/when those societies change in response to factors such as population growth.

The classic papers on the lower size limits of hunter-gatherer populations were published by Martin Wobst in the 1970's. Like him, I employ model-based approach to address the issue of how big a human population has to be to not be threatened by random fluctuations in mortality, fertility, and the ratio of males to females. Very small populations are more sensitive to those random fluctuations because each person makes up a greater percentage of the population. 

My analysis suggested that, under a range of conditions represented in the model, human populations with more than about 150 people were fairly safe over long periods of time.  That's a smaller lower size limit, I think, than a lot of people conceive of.

Here is the abstract:

"A non-spatial agent-based model is used to explore how marriage behaviors and fertility affect the minimum population size required for hunter-gatherer systems to be demographically viable. The model incorporates representations of person- and household-level constraints and behaviors affecting marriage, reproduction, and mortality. Results suggest that, under a variety of circumstances, a stable population size of about 150 persons is demographically viable in the sense that it is largely immune from extinction through normal stochastic perturbations in mortality, fertility, and sex ratio. Less restrictive marriage rules enhance the viability of small populations by making it possible to capitalize on a greater proportion of the finite female reproductive span and compensate for random fluctuations in the balance of males and females."
  

If you're interested in hunter-gatherer theory stuff, have a look and see what you think. This is probably the first paper of several I'll be writing on the topic.

I'm happy to announce the publication of a new paper of mine in The Journal of Artificial Societies and Social Simulation (JASSS). JASSS is an open access journal focusing on the use of computer simulations as a tool for understanding social systems. I love publishing papers that anyone can get to at any time, without requiring logins, subscriptions, or fees. I wish every paper could be open access, and I wish I had the funds at my disposal to throw money at journals like JASSS to support such efforts. If I ever do have the money, that's the direction I'm going to throw it.

My paper, titled "A Model-Based Analysis of the Minimum Size of Demographically-Viable Hunter-Gatherer Populations" will appear in Volume 20, Issue 4 (due to be released on October 31). You can read the paper online here. It should be available in pdf form soon.

If you really want to get into the nitty gritty, the raw model code and an explanation are available here.
​
The focus of the paper using an agent-based model to revisit the question of how large human groups have to be to be demographically viable (i.e., able to survive over the course of many generations). This is a key question for understanding the size and structure of ethnographically- and archaeologically-known hunter-gatherer social systems as well as fleshing out scenarios of hunter-gatherer groups colonizing empty landscapes. Here is the abstract:

"A non-spatial agent-based model is used to explore how marriage behaviors and fertility affect the minimum population size required for hunter-gatherer systems to be demographically viable. The model incorporates representations of person- and household-level constraints and behaviors affecting marriage, reproduction, and mortality. Results suggest that, under a variety of circumstances, a stable population size of about 150 persons is demographically viable in the sense that it is largely immune from extinction through normal stochastic perturbations in mortality, fertility, and sex ratio. Less restrictive marriage rules enhance the viability of small populations by making it possible to capitalize on a greater proportion of the finite female reproductive span and compensate for random fluctuations in the balance of males and females."

My main finding is that, under the varying conditions I investigate with my model, I find no support for the often-repeated idea that a society of about 500 persons is required to ensure demographic viability.  Students of American anthropological archaeology -- especially, I suspect, those of us that went to the University of Michigan or were taught by U of M alums -- will immediately recognize the "magic number 500" as a concept that emerged from the research of Joseph Birdsell and (later) H. Martin Wobst. As I discuss in the paper, I think neither Birdsell nor Wobst intended the number 500 to take on the meaning that it did -- it became a kind of shorthand gloss for setting a general lower boundary on the size of hunter-gatherer social systems.

My modeling results suggest that the number of people required for demographic viability can be safely pushed down south of 200.  In over 67,000 model runs (under varying conditions of mortality, fertility, and marriage rules) where the mean population exceeded 150 people, the population went extinct only nine times.  I'd take those odds.

All the modeling was done under conditions with no logistical constraints to identifying and obtaining marriage partners:  no spatial component to interaction, no impediments to the flow of information. Logically, putting the model systems in space and dispersing the populations across a social/physical landscape would have the ultimate effect of raising the population size required for demographic viability. Would it double or triple it, though? I highly doubt it.  But the great thing about modeling is that we don't have to be satisfied to simply suppose things -- we can model the problem.  Understanding that less than 200 people are required for demographic viability assuming no interaction issues, we can then unpack the issue to ask why hunter-gatherer societies are often much larger. What role does the structure of mobility play? What about the need to maintain a geographically-extensive social fabric to buffer large-scale environmental variability? Here are a couple of paragraphs from my conclusion:
​

The blog has been on the back burner while I deal with the beginning-of-the-semester crunch. I've got a lot going on this year, so I'll probably have less time to write than I did in years past. Keeping all the parts of my three-headed monster of a research agenda moving is more than a full time job.

I wanted to write a quick post about the presentations I've committed to for the fall (SEAC) and spring (SAA) conferences, as they give you a pretty good idea on what's going on with some of my "big picture" work.  I gave a presentation about my work on understanding the Kirk Horizon to the Augusta Archaeological Society at the end of August, and I'll be giving an informal presentation to SCIAA next week synthesizing what we know so far about the natural/cultural deposits at 38FA608 (site of last spring's Broad River Archaeological Field School). Here's what I'll be doing at the regional and national conferences:

SEAC (November 2017, Tulsa, OK)

David Anderson and I are teaming up to give a paper titled "Structure, Density, and Movement: Large-Scale Datasets and Basic Questions about Early Foraging Societies in the Eastern Woodlands." The paper will part of a symposium organized by Shane Miller, Ashley Smallwood, and Jesse Tune titled The Paleoindian and Early Archaic Southeast: The Last 20 Years, 1996-2016. Here is the abstract of our paper:

"Distributions of diagnostic projectile points show that the Paleoindian and Early Archaic societies of the Eastern Woodlands were spatially-extensive, occupying vast and varied landscapes stretching from the Great Lakes to the Florida Peninsula. The scales of these societies present analytical challenges to understanding both (1) their organization and (2) how and why the densities and distributions of population changed during the Late Pleistocene and Early Holocene. We integrate several large datasets – point distributions, site locations, and radiocarbon dates – to address basic questions about the structure and demography of the Paleoindian and Early Archaic societies of the Eastern Woodlands."  

We'll be integrating data from PIDBA, DINAA, and my ongoing radiocarbon compilation. There will be some significant work involved in meshing all this stuff together in a GIS framework that we can use analytically, so that will be one of the main things on fire for me in the coming month.

SAA Meeting (April 2018, Washington, D.C.)

​At this year's SAA meetings, I'll be contributing to Scott Jones' symposium titled Forager Lifeways at the Pleistocene-Holocene Transition. My paper is titled "Patterns of Artifact Variability and Changes in the Social Networks of Paleoindian and Early Archaic Hunter-Gatherers in the Eastern Woodlands: A Critical Appraisal and Call for a Reboot." Here is the abstract:

"Inferences about the social networks of Paleoindian and Early Archaic hunter-gatherer societies in the Eastern Woodlands are generally underlain by the assumption that there are simple, logical relationships between (1) patterns of social interaction within and between those societies and (2) patterns of variability in their material culture. Formalized bifacial projectile points are certainly the residues of systems of social interaction, and therefore have the potential to tell us something about social networks. The idea that relationships between artifact variability and social networks are simple, however, can be challenged on both theoretical and empirical grounds: complex systems science and ethnographic data strongly suggest that patterns of person-level interaction do not directly correspond to patterns of material culture visible at archaeological scales. A model-based approach can be used to better understand how changes in human-level behaviors “map up” to changes in both the system-level characteristics of social networks and the patterns of artifact variable that we can describe using archaeological data. Such an approach will allow us to more confidently interpret changes in patterns of artifact variability in terms of changes in the characteristics and spatial continuity/discontinuity of social networks during the Pleistocene-Holocene transition in the Eastern Woodlands."  

This is a basket of questions that was the main focus of my dissertation work. My goal is to lay out the case for why we really need to be doing things differently than we are in order to get at questions about social networks and social interaction. With the SAA meetings still months away, I plan to do new modeling work to support my argument. If I'm to do that, I'll have to ramp up my modeling efforts and deal with some issues around adding space back into the main models I've been working with. It needs to be done, so committing to a paper is a way to make sure I prioritize it.

I'll also be participating in a "Lightning Round" about engaging pseudoarchaeology. In this session (organized by Khori Newlander), the participants will each get just three minutes. No abstract is required for this one. As of now, I plan to use my time for "Swordgate: How to Win Friends and Influence People." 

I don't remember who was talking, but I recently heard some TV pundit remark that it was both a surprise and disappointment that the internet has contributed to fostering cultural/social/political divisions rather than promoting unity. 

Disappointment? Yes.

Surprise? No.

Complex systems models have demonstrated over and over again that disunity (i.e., cultural polarization, geographic segregation, etc.) can emerge in systems within which information flows freely. The lesson is that having a high degree of information flow doesn't guarantee homogeneity at the scale at which the information is flowing. Two simple models demonstrate this point nicely.

First, the Schelling Model. If you're a fan of complex systems theory, you've probably heard of the simple simulation model that Thomas Schelling constructed and explored in the early 1970's and published in Micromotives and Macrobehavior (1978).  The original model was implemented using coins and graph paper rather than a computer. It demonstrated how relatively small preferences about the characteristics of ones' neighbors can result in complete segregation of neighborhoods. Actors in the model make decisions about whether to stay put or move based on on information about their immediate surroundings. Through a multitude of individual, localized decisions, large-scale patterns of segregation emerge in the absence of any intent or authoritative control.

Second, Robert Axelrod's (1997) adaptive culture model (see here). If you want a simple complex systems model . . .  it doesn't get much simpler than this one.  The model demonstrates how polarized cultural regions can develop even though the only mechanism for interaction in the model is one of convergence (i.e., the actors can only become more, not less, like their neighbors).

The short version of my argument (which is all I have time for today) is that complex systems approaches will provide an actual chance to figure some of this stuff out. You're not going to be able to write a mathematical formula to do it, as straight math can't handle emergent phenomena (show me a formula that captures this kind of flocking behavior and I'll admit I'm wrong). And you're not going to be able to sort it out by comparing two or three variables at a time ("white males between the ages of 18 and 22 in this county voted for X, which the same demographic in this county voted for Y").

And of course there's no such thing as a "simple" complex systems problem. But complex systems theory has helped us understand a thing or two about human cultural/social/political behavior that we wouldn't be able to understand otherwise. And some of that understanding has come from some relatively simple models.  I'm sure there has already been work done extending models like Schelling's and Axelrod's to represent media influences, complex structures of interaction (i.e., different network topologies, etc.), variable demography, etc. The smart money will pay attention to that work to help identify and understand the characteristics of our system that exacerbate divisions (and can be used to widen those divisions).

It already sounds like a paper title -- just replace what's after the colon with "A Model-Based Approach."

I'm coming up on the end of my second year in South Carolina. I think it takes a few annual cycles before you start to "get" the rhythms and tempos of seasonality in a new environment. Prior to coming here I had lived in the Midwest for most of my life, so there's a lot to learn. 

As an archaeologist, I don't try to understand the environment just so I can give it a round of applause (if I had to pick what to applaud here, however, it probably would be the birds, flowers, and insects). Human societies and natural environments are inter-linked in numerous and  complex ways -- figuring out those linkages and understanding how the "social" and "natural" parts of those coupled systems affect one another is an intrinsically interesting and profoundly important part of understanding how human societies work and how they changed in the past.

My point in writing this isn't to compose a fully-formed, well-researched argument, but rather to jot down a few observations/ideas/questions that have struck me since I transplanted myself into a region of the country with environments that are, in many ways, dissimilar from those of the Midcontinental interior with which I am most familiar (i.e., the Ohio Valley, the Till Plains, the Great Lakes). I don't have time to pull all these strings yet -- I'm just noting them.

First, the Deer . . .

Early on, I commented on what must be differences in the demography and behavior of a key Holocene large game species (white-tailed deer) across the different regions of the Eastern Woodlands. One would expect that those regional differences -- whatever they are -- would have articulated somehow with the behaviors of the human populations that exploited them.  Generally, we presume that periodic (i.e., seasonal) aggregations of hunter-gatherer populations are useful to those societies for a number of demographic and social reasons. Logically, aggregations of large numbers of people have to take place when and where the resource base can support them. I would guess that most archaeologists in the north have a "fall aggregation" model in their heads, based in part on when deer are the fattest and least cautious. Are those conditions different in the Southeast, where the seasonal gradient is much less severe than in the north?  Do deer populations go through boom/bust cycles? If so, are those linked to periodicities in mast production? Do those periodicities differ from region to region in the Eastern Woodlands? Deer hunting isn't everything, but it's surely something.

​Second, the Sea . . .

At some recent conference, I had a conversation with a colleague who has been working in this region for a long time. It was clear he had had a few drinks, so he was probably telling me the truth. He said that the rhythms and tempos of hunting and gathering on the coast are very different than in the interior. I've never done coastal archaeology -- when I go to the beach it's usually to let the kids play, watch birds, and look for shells.

We were at Edisto last year during the time when the loggerhead sea turtles come ashore to lay their eggs. These are big animals, with adults weighing about 300 pounds (up to about 1000 pounds). The females come ashore at night during the summer to lay about 120 eggs in a nest in the sand. 

Watching the Edisto turtle patrol identify and check nests every morning, I became curious about how turtle nesting behavior articulated with prehistoric coastal hunter-gatherers in this region. The nests are easily spotted by the tread-like path that turtles leave as they move across the sand. Caught in the act, the adult turtles are large packages of meat, sitting in the open, defenseless. Presumably a couple of people could flip one on its back and return later for an on-the-spot feast or to butcher the animal.

How much archaeological evidence is there of sea turtle exploitation on the Carolina coast? Does it change through time? Where would sea turtles rank in terms of a seasonally-predictable resource that could be used to support periodic aggregations? Were sea turtles part of coastal Carolina hunter-gatherer cosmology (perhaps in connection with the summer solstice)? I don't know the answers to any of these questions.

Third, the Air . . .

The birds here are beautiful, plentiful, varied, and constant. Of the 914 species of birds documented in the United States, over 400 occur in South Carolina. That's a lot of birds. Some sing all year round. Some even sing at night. It's fabulous.

One bird I have learned about since I moved here is the Mississippi Kite. It is a smallish, grey raptor that winters in South America but breeds in the southeastern United States.

These birds eat mostly flying insects, and you can see them circling over my neighborhood during much of the summer. Their appearance in the region seems to coincide with what I interpret as the "high" insect season -- the cicadas are hatching in force and there are things buzzing around everywhere. They're a signal of a season change here, perhaps much in the same way as the yearly arrival of Turkey Vultures north of the Ohio River. 

However the annual long-distance migration/breeding pattern of the kites evolved, I would guess that the dense insect populations of the Southeast are a key to making it viable. That got me thinking about the effects of longer-term periodicities, particularly the those of the 13-year and 17-year periodical cicadas. The emergence of buhzillions of cicadas at the same time would surely make for easy living for the kites, as well as for game animals with an insect-based diet (e.g., turkeys). The periodical cicadas tend to damage trees, however, which reduces mast production (and hence could have a suppressing effect on deer populations). Did any of this factor into the characteristics (social, behavioral, cosmological, etc.) of the prehistoric human societies of this region? I don't know.

Finally, from the Periodic to the Anomalous . . .

The completion of my second year in Columbia will be marked by a total solar eclipse that I'll be able to experience from my backyard on August 21 at 2:41 p.m. I've never seen a total eclipse before, and I may never see one again. Most people don't see one in their lifetime. I'm really looking forward to it. Thankfully I won't have to stay up late at night to see it.

Obviously, it's now old hat for us to predict these "anomalous" astronomical alignments with a great deal of accuracy (business depends on it). Given how infrequently these things occur and the low probability of any one person accidentally being in the right place at the right time to witness it, it's natural to wonder what prehistoric peoples would have made of this sort of phenomenon. I'm really curious as to what it will feel like to experience it firsthand (I'd also like to know what's it like to be in a hurricane, to break the sound barrier, to be close to a tornado, to fly at the edge of the atmosphere, to experience zero gravity, etc., in case your looking for ideas for my birthday).

So What?

Somewhere in all this mess, there's a question to be crystalized about how human societies "tune" themselves to the predictable and unpredictable fluctuations in their environments. What are the feedbacks? What are the dampers? What are the common denominators? What is the range of risk/variability that societies create cultural rules or behaviors to respond to? What happens when the needle moves outside of that range? Which parts are robust? Which parts break? How do responses scale to the size and predictability of perturbations across time and space?  I have no answers right now, just questions.

And now I've got to move on and do other things.

Following the spring break hiatus, the Broad River Field School will be back in session tomorrow. We'll be shifting gears a bit to carefully work our way into what appears to be a buried Late Archaic/Early Woodland component. I'm also anticipating continued work on the deeper deposits at the site. Hopefully it will be an eventful day. It's supposed to be sunny and in the mid 60's. I'll just leave it at that.

Here are a few quick updates on other things for those playing along at home: a new modeling paper about the minimum size of demographically viable hunter-gatherer populations, new Fake Hercules Swords en route, and an identification of last Friday's whatzit. 

Finally, following up on yesterday's post about Against Me!, I would like to encourage you to listen to the song "Rebecca" if you like the rock'n'roll music. It's on repeat in Andyland. 

I've found that the notes I make on my website are much easier for me to find later than the things I scrawl on pieces of scrap paper. The shoe fits, so I'm going to wear it whether or not it's interesting or useful to anyone else.

I've spend much of my time since Monday working to get the agent-based model I'm currently messing around with (version 3 of ForagerNet3_Demography) to work in version 2.4 of Repast Simphony. Something went awry with my previous installation (2.0) and it made sense to start trying to fix things with the most current version. After many frustrating loops of install/uninstall/reinstall, I finally got a good installation of 2.4 and got my source files moved in and working.

One thing that I wasn't able to make functional in the last version was the batch mode.  Batch mode lets you run jobs in batches, speeding up computation by farming out the computational workload to different processors or machines, etc. (doing runs in parallel rather than serial). It wasn't too troublesome that I couldn't get the embedded batch mode working before, as the model I'm working with isn't terribly demanding. There are bigger things coming down the road, however, so I was happy to see the batch mode appeared to be active and ready to go with the new installation.

I spent most of yesterday trying to figure out how to use the batch mode while preserving the data output format that I've created on my own and that works for me. The way I've written the model, it produces a summary data file at the end of each run. The file (just a simple .txt file that I can open in a spreadsheet) preserves information on all the parameter settings used in the run as well as data on the outcomes. In the Repast/Eclipse batch mode, however, that .txt file wasn't getting produced. I was frustrated to learn that I apparently had to redo all the data capturing in the model so that the software could use something called a "file sink" to collect information. I was then relieved to learn that there was a workaround to using the "embedded system:" you can specify an "Optional Output File Pattern:"

The "Pattern" is the name of the file produced by the model. I changed the file name to "TestBatchMode.txt" in my code for the purposes of experimenting with where and under what circumstances the file would be created. The "Local Path" is what you want the file to be called that the batch mode creates (it will appear in the output directory specified above). Check the "Aggregate" box so that it saves the output from each run and then gloms it all together in a single file. I left "Has Header" unchecked because my data doesn't have a header.

Repast seems to be picky about the settings in the "batch_params.xml" and "parameters.xml" files, wanting them to match. I get null pointer exceptions when I clear everything but the random seed out of the files, so I just left a parameter in to keep the software happy (it does nothing, as it's a constant and doesn't affect any of the settings in the model). 

I'm still testing out the batch mode and the model to make sure everything is working properly. My plan then is to modify version 3 to create version 4, which will include some different mortality settings (which, in turn, will necessitate rewriting some of the code that calculates age-specific mortality outcomes). After all that's done, I'm planning on doing some new modeling work to formally follow up on some preliminary observations I presented at the SAA meetings a few years ago.

I devoted yesterday afternoon and most of today to (finally) producing the updated documentation for Version 3 of the ForagerNet3_Demography model, one of the agent-based models that I've been working with. You can read all about it on this page, and you can even download the raw code if you like. Files and citation information for the model are also available on OpenABM.org.

I used a version of this model (implemented in Repast J rather than Repast Simphony) in a recent paper published in Uncertainty and Sensitivity Analysis in Archaeological Computational Modeling (edited by Marieka Brouwer Burg, Hans Peeters, and William Lovis).

Over the summer I used to model to generate data for a paper on the minimum viable population (MVP) size of human groups. That's in the editing stages now -- hopefully I'll be able to get it done and submitted somewhere before the Christmas break.

​Onward.

We're now into the fourth week of the semester here at the University of South Carolina. As usual I've been writing for this blog less than I'd like (I have several unfinished draft posts and ideas for several more, and there's currently a backlog of Fake Hercules Swords). A good chunk of my time/energy is going into the Forbidden Archaeology class (you can follow along on the course website if you like -- I've been writing short synopses, and student-produced content will begin to appear a few weeks from now). Much of the remainder has gone into pushing forward the inter-locking components of my research agenda. This is a brief update about those pieces.

Small-Scale Archaeological Data

At the beginning of the summer I spent a little time in the field doing some preliminary excavation work at a site that contains (minimally) an intact Archaic component buried about 1.9 meters below the surface (see this quick summary).  Based on the general pattern here in the Carolina Piedmont and a couple of projectile points recovered from the slump at the base of the profile, my guess is that buried cultural zone dates to the Middle Archaic period (i.e., about 8000-5000 years ago).​

My daughter washed some of the artifacts from the site over the summer, and I've now got an undergraduate student working on finishing up the washing before moving on to cataloging and labeling. Once the lithics are labeled we'll be able to spread everything out and start fitting the quartz chipping debris back together. Because I piece-plotted the large majority of the lithic debris, fitting it back together will help us understand how the deposit was created. I'm hoping we can get some good insights into the very small-scale behaviors that created the lithic deposit (i.e.,perhaps the excavated portion of the deposit was created by just one or two people over the course of less than an hour).

When the archaeology faculty met to discuss the classes we'd be offering in the spring semester, I pitched the idea of running a one-day-per-week field school at the site. Assuming I can get sufficient enrollment numbers, that looks like it's going to happen. The site is within driving distance of Columbia, so we'll be commuting every Friday (leaving campus at 8:00 and returning by 4:00). The course will be listed as ANTH 322/722. It's sand, it's three dimensional, and it's pretty complicated -- it's going to be a fun excavation. I'll be looking to hire a graduate student to assist me on Fridays, and I'll be applying for grant monies to cover the costs of the field assistant's wages, transportation, and other costs associated with putting a crew in the field. 

Large-Scale Archaeological Data

Some parts of my quest to assemble several different large-scale datasets are creeping along, some are moving forward nicely, and some are still on pause.

Complex Systems Theory and Computer Modeling

Complex systems theory is what will make it possible to bridge the small and large scales of data that I'm collecting. Last year, I invested some effort into transferring my latest computer model (FN3_D_V3) into Repast Simphony and getting it working. I also started building a brand new, simpler model to look at equifinality issues associated with interpreting patterns of lithic transport (specifically to address the question of whether or not we can differentiate patterns of transport produced via group mobility, personal mobility between groups, and exchange).  

As it currently sits, the FN3_D_V3 model is mainly demographic, lacking a spatial component. Over the summer I used it to produce data relevant to understanding the minimum viable population (MVP) size of human groups. Those data, which I'm currently in the process of analyzing, suggest to me that the "magic number of 500" is probably much too large: I have yet to find evidence in my data that human populations limited to about 150 people are not demographically viable over spans of several hundred years even under constrained marriage rules. But I've just started the analysis, so we'll see. I submitted a paper on this topic years ago with a much cruder model and didn't have the stomach to attempt to use that model to address the reviewers' comments. I'm hoping to utilize much of the background and structure of that earlier paper and produce a new draft for submission quickly. I also plan to put the FN3_D_V3 code online here and at OpenABM.org once I get it cleaned up a bit. I also discuss this model in a paper in a new edited volume titled Uncertainty and Sensitivity Analysis in Archaeological Computational Modeling (edited by Marieka Brouwer Burg, Hans Peeters, and William Lovis). 

It will be a relatively simple thing to use the FN3_D_V3 model in its non-spatial configuration to produce new data relevant to the Middle Paleolithic mortality issue I discussed at the SAA meetings a couple of years ago. I'm also going to be working toward putting the guts of the demographic model into a spatial context. That's going to take some time.

I've been spending most of my discretionary work time over the last couple of weeks grinding through the process of getting my computer models up and running again. The main challenges have involved converting my models from Repast J to Repast Simphony. While the Java components are the same, the "world" of the models is structured differently in Simphony. So I've had to try to figure out how to re-connect the various parts of the models using what Simphony calls "context."  I can't yet say I fully understand how "context" works, but I got my ForagerNet3_Demography_V3 (FN3_D_V3) model up and running by trial and error and looking at examples of code from other models.   

Since I haven't figured out to configure the model to use the batch run GUI (I haven't even been able to find it yet, although it apparently exists somewhere in Eclipse), I've been using a primitive parameters file to do batch runs. As I wrote on Friday, these batches would throw an "Out of Memory" error and freeze up around the fortieth run. That suggested some kind of memory leak where object produced during a run were not being deleted before the next run. The gradual accumulation of unused objects eats up the memory until there isn't any left to use to run the model, then it dies.

After going through the code several times and trying a bunch of options, I think I finally found the culprit(s) and made the corrections.  I set the model to run 500 times a couple of days ago, and things seemed to be chugging along just fine (when I got to my office this morning it was at run number 720-something, so clearly I didn't have the "stop" command implemented correctly).  Anyway, I've  now got a batch of new data that I can compare with data produced by the model when it was implemented in Repast J.  The figure below compares old model data (left) with new model data (right).  It is gratifying to see the model is behaving the same.

The data on the left are from a paper of mine ("The Sensitivity of Demographic Characteristics to the Strength of the Population Stabilizing Mechanism in a Model Hunter-Gatherer System") that will be published in an upcoming volume titled Uncertainty and Sensitivity Analysis in Archaeological Computational Modeling (edited by Marieka Brouwer Burg, J. H. M. Peeters, and William A. Lovis; Springer).

I'm happy that this model is back in business. I'll do some more testing to make certain everything is working, then I'll clean up the code and make it available on this website and under my profile at OpenABM. I plan to use this model for some work on the demographic viability of small populations and, perhaps, to push ahead with exploring demography, mortality, and fertility during the Middle Paleolithic.