Investigation of land-use strategies and economic sustainability in the Iron Age Central Europe Workshop Ancient Socio-environmental Modelling Kiel 28th January 2014 Alž ěta Da ielisová Institute of Archaeology CAS, Prague, v.v.i. Socio-economic „transformations of the La Tè e period burial evidence (CE) fürste oppida collapse fürste sitze money elites warrior burials/events no burials at all 400 500 Ha D LT A LT B1 300 LT B2 Punctuated equilibria theory (Eldredge – Gould 1972). 100 200 LT C1 LT C2 LT D1 0 LT D2 Stradonice Ně či e Manching Roseldorf Bratislava so called „Boian territory emporia x oppida 3rd – 1st cent. BC dynamics of their chronology spectrum of their functions  main focus Late Iron Age - Bohemia oppida Oppida = Complex systems/societies with multiple functions written sources (Caesar) organised layout centre of politics aristocratic farms cult social hiearchy production cult politics fortification commerce oppidum commerce production Galinié 2009 production aspects: • • • • Land-use strategies, ecology Economics and subsistence Interactions, organisation of production Surplus management rural settlements Oppida (CE) - dynamics of their occupation material collections coins 100 Staré Hradisko 50 0 ltc2 ltd1 ltd2 )á ist 100 40 50 20 4000 3000 Hrazany 0 ceramics 0 LT C2 LT D1 LT D2 LT C2 LT D1 Lt D2 2000 0 II 4000 3000 Stradonice Třísov 1000 III IV V animal bones 100 400 50 200 0 0 2000 LT C2 LT D1 LT C2 LT D2 LT D1 LT D2 1000 0 II III IV 15 10 Závist V brooches Manching 60 60 40 40 20 20 0 5 0 LT C2 LT D1 LT D2 LT C2 LT D1 LT D2 0 0 I II III IV V Evidence of (probably massive?) emigration in the 2nd half of 1st cent. BC? Scarce material of late (LT D2) phase at both the oppida and the open settlements. Dynamics of the occupation 150 BC „up ard s eeps „ risis of the e tral pla es climax ? 90-70 BC change time 50/30 BC burden of the economic growth  absorbing all the production from their regions ... „house of cards“ model decline collapse population CENTRE x CENTRE 1 2 Complexity of the process of change causes? political/historic organizational economical/long distance contacts ecological/subsistential Revised iteration model for upsweep development in complex societies (Chase-Dunn et al. 2007) Principles: - supply crisis - reaching the limit (carrying capacity) - deterioration of welfare subject for testing Idea  impacts of demographic / production cycles population x resources integrative phase specialization disintegrative phase complex elites carrying capacity long distance trade market relations money economics 250 200 collapse 100 0 BC/AD „expansion“ „stagflation“ „crisis “ Turchin – Nefedov 2009 „Secular cycles „collapse “ Relationship between population, production and resources Effects of population growth on total production, subsistence needs Production of the surplus Turchin – Nefedov 2009 „Secular cycles Methodology  consistent and systematic analysis of the complex structures  materialistic point of view complemented by: behavioral analysis  hu a s as „age ts  Simulations with agent based modelling  perceptions and interactions with an actual environment based on independent decisions „Complex simulation of the sustainability of the oppida subsistence strategies  „Czech Science Foundation“ project no. P 9 Principles:      exploratory models ask proper / simple questions work on them until they are stable add details, work on complexity revision, explanation BehaviorSpace BehaviorSearch SystemDynamics Bohemia and Moravia – 3rd – 1st cent. BC Lovosice Stradonice Prague Závist Nowa Cerekwia Čs.Lhoti e St. Hradisko Ně či e Třísov Roseldorf e poriu „Ně či e t pe rd agglomeration 2nd-1st cent. BC oppidum – 2nd cent. BC Case study – central Moravia – oppidum of Staré Hradisko spatial and chronological shift of the central place Hnevotin LT B-C LT (C2)D LT B-D Staré Hradisko 150 – 50/30 BC Pte í Ohrozim Klentnice Dětko i e 290/250 – 150 BC Ně či e Theoretical framework • Modelling the crucial resources: - hinterland  size, land-use units • Modelling the population and its subsistence needs: ⁻ population size, demographic structure, energy (=nutrient) needs • Modelling the resource exploitation and setting the limits: - Energy potential of the key resources rop ields, fodder, oodla ds … - Outlining the possible exploitation and production strategies. - Organization of the working process in relation to the land-use patterns and social structure - Determining limit factors and their impacts • Modelling the dynamics of the production: - Fluctuations of the harvests  actual harvest levels, surplus and potential storage reserves - Stability and sustainability of production  number of stress situations - Limit thresholds  carrying capacity in relation to the population pressure and potential scarcity or depletion of resources,  work force - External factors, if there were any (climate change, weather events ... ) Da ieliso á et al. Data - Population density and evidence of husbandry economics • • • • • • settlement structure number and size of households site chronology (stratigraphy, material collections) number of livestock crop cultivation agricultural tools Staré Hradisko 100 50 0 LT C2 LT D1 LT D2 Data – environment  topography  soils, geology  vegetation composition  historical mapping grasslands forest fields Data – archaeobotanical, archaeozoological Staré Hradisko crops animals Modelling of the hinterland Da ieliso á – Pokor ý 9 lack of data  „La d use suita ilit Multi-criteria evaluation method ap (GIS) Quantitative GIS model (Eastman 2006) Environmental variable Relief (digital elevation model DEM) Landforms, topographical features, topographic wetness index Hydrology Geology and Soils Soil quality Potential vegetation Climate Weather Source modelled in GIS from 1:5 000 topographical maps (ArcGIS, resolution 5x5m) modelled in GIS from DEM (ArcGIS, IDRISI, Whitebox, Landserf) modelled in GIS from DEM and complemented by fluvial sediments in geological maps and historic mapping (ArcGIS, manual correction) digitalised from geological and soil maps 1:50 000 BPEJ soil evaluation Neuhäuslo á et al. Macrophysical Climate Model (MCM) created from local meteorological data (Bryson – DeWall 2007, Da ieliso á – Haj alo á i pri t recorded historic frequency of the e e ts hailstor s, late frosts, hea rai s … Brázdil et al. 2006) Da ieliso á et al. landforms+soils+wettness ind.+geology - settlement area theory - site catchment theory T. Bayliss – Smith 1978 R. Ebersbach 2002 R. Schreg 2011 Strategies of land use „E os ste theor refers directly to the point of the community´s demography, subsistence, and organisation of labour in relation to the environment settlement arable land pastures woodland ased o ala e et ee three „L Land – Livestock - Labour „Ope “ s ste „Ma i u “ s ste „Closed“ s ste 0.15 ha/person 0.5 - 1 cattle/person 0.39 ha/person 0.28 cattle/person unlimited 0.8 – 2 cattle/person Strategies of land cultivation • Intensive „closed system high labour input > high productivity > limited scale • Extensive „open system low labour input > low productivity > extended scale (fallows) • Historical, ethnographic data: • Ploughing, management, harvest, crop processing rates • Animal, forest management rates Combination Intensity of land-use Experimental data (long term agricultural experiments in similar environment)       yields under different regimes harvest fluctuations good/bad years arable land extent necessary labour input climate determinants Hejcman-Ku zo á 2010 Ku zo á-Hejcman 2009 Rothamstead Research 2006 Climate reconstruction Macrophysical Climate Model -no proxy data -orbital forcing (Milankovich) -changes in transparency of atmosphere -calculates position of Centres of Action Protivanov -applies principles of synoptic climatology -input of local climate normals 1961-1990 => local climate models Ivanovice Precipitation – Evaporation SH_Precip2 Protivanov SH_Evap2 700 Precipitation – Evaporation Temperature history Ivanovice 650 9.00 IVA_Precip IVA_Evap 700.0 8.50 600 650.0 8.00 7.50 550 600.0 500 550.0 5.50 -950 -750 -550 -350 -150 50 250 450 6.00 650 1050 6.50 850 7.00 500.0 1050 850 650 450 250 50 -150 -350 -550 -750 -950 SH_Ptemp IVA_Ptemp 1050 850 650 450 250 50 -150 -350 -550 -750 -950 450.0 5.00 Da ieliso á et al. i prep Temperature Climate reconstruction 250 BC 150 BC 50 BC Precipitation 250 BC 150 BC 50 BC Climate reconstruction Distribution of precipitation, evapotranspiration Agricultural year Distribution of precipitation - Sh 120 100 1950 80 50 60 -50 40 -150 20 -250 0 Potential Evaporation - Sh 60 40 20 0 -20 -40 1950 50 -50 -60 -150 -80 -250 -100 Da ieliso á et al. i prep Population Dynamics model (ABM) each agent: • gender (2 categories) • age (7 categories)  total population Outputs: 1) population growth  dynamics of the population increment 2) necessary energy input  consumption (caloric input value extrapolated from the actual oppidum population in all sex/age groups) 3) available workforce (actual number of people (15-49 years) in particular age/sex categories) - „stro g for e - „ eak for e for each year of simulation 100 – 120 years (ticks = 120) • Cal ulatio of „working availability hours per month, per year for strong and weak workforce Population Dynamics model (ABM) natality/mortality rate  based on lifeexpectancy tables (Saller 1994)  Model Life Table Level 3 2% pop.growth in suitable circumstances Initial and final age distribution Wo a ’s probability of having children ! q(x) = probability of dying before the next exact age (production treshold) Population Dynamics model workforce labour availability during the agricultural year need to cover the necessary production tasks ? non-producers 15.0 15.0 10.0 meadows woodland 10.0 domestic woodland other 5.0 animals 5.0 field males females Dec No … Oct Sept Aug… July Mai… June Apr Mar Ja … Fe … 0.0 Dec No … Oct Aug… Sept July June Mai… Apr Mar Fe … 0.0 Ja … field Husbandry Economy Model overview – simulation sequence Livestock management Model overview Simulation – time management  each month/120 years (= 1440 ticks) Inputs: • • • • • • average crop yield and its standard deviation (INTxEXP 700-3000 and 500-2000 kg/ha) agricultural strategy workforce data (availability) % ratio of consumption workers / area unit seed corn, losses • Model structure: – – – – – – – – – agriculture.nls animals.nls food.nls forest.nls GISload.nls output.nls population.nls simulation.nls visual.nls Interface buttons distance penalty • • • • • GISload.nls – loads data sets, applies rasters Distance from oppidum Distance from streams Wetness index Slope percent Woodland taxa Agricultural.nls • • Setup Animals.nls • • • • • • • herd population pasture requirements (energy inputs) nutrient yields (proteins) labour for individual tasks • Forest.nls • • • initial state – management area wood (fuel, construction) – acquisition renewal field requirements management strategy (INT x EXT) labour for individual tasks annual harvest (fluctuations) Events – small/big probability seed corn Food.nls Output.nls • for every month • - field area, yields, consumption, critical situations • • consumption data – per mont/year STORAGE – crops, milk, meat (kcal) Crop storage (1 – 3 years) Losses Agricultural model – results Land Use - Area statistics Intensive Year: 1 Population: 800 Year: 60 Population: 1450 Year: 100 Population: 2300 Year: 60 Population: 1450 Year: 100 Population: 2300 Extensive Year: 1 Population: 800 Agricultural model – results Production sustainability NoEvents SelfSustainable Intensive NotSustainable Extensive Simulations show decreased net returns or problems with availability of the arable land after reaching certain point of population density and particular farming strategy employed. Agricultural model – results Labour input  oppida were sustainable by their own production ... to a certain limit  communities could house non-producers (15-20%)  from certain labour input (50 – 80% Sf+Wf) production is stable Even in case when detailed data is limited, these models could point to the constraints of the particular agricultural strategy and population density in relation to the specific environment Wurzer, G. – Kowarik, K. – Reschreiter, H. (Eds.) 2013 Rural settlements  oppidum Eigenvalues of correlation matrix Active variables only 6 52.29% 5 Husbandry data Eigenvalue 4 3 23.36% 2 1 6.89% 6.21% 4.08% 3.22% 2.11% 1.26% .36% .21% 0 -1 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 Eigenvalue number PCA Analysis LT B-C LT (C2)D LT B-D Chronology System dynamics – villages and oppidum model overview System dynamics – villages and oppidum interface example current questions Oppida and rural settlements Correlation factors food-supported part of society x producers  nobility  specialists: how many people in productive categories are actually working in agriculture? as it at all „ur a - ased ? what sites can offer in reciprocal relationship ? nature of goods, clientelism, dependence bad years surplus carry overs stability of market network, transport monetary economics – how common? acquiring new data Network Analysis of late Iron Age Society • • Nodes = oppida, settle e ts, illages,… Li ks = roads, o u i atio , produ t e ha ge, lo al trade e ha ge,… • More o ple pi ture of fu tio i g of the so iet … Where Agent Based Modelling should be headed? • odels should tr to e ulate the real orld as closely as possible with as much detail and data as possible • Or should odels e as si ple as possi le, maybe even stylized and mainly be used for exploration and experimentation? Lake 2010, Premo 2010, Kowarik 2012 Thank you for your attention