KinModel: an agent-based model of rangeland kinship networks

McAllister, R.R.J., Gordon, I.J., and Stokes, C.J. (2005) KinModel: an agent-based model of rangeland kinship networks. In: Proceedings of MODSIM05 - International Conference on Modelling and Simulation. pp. 1624-1630. From: MODSIM05 - International Conference on Modelling and Simulation, 12-15 December 2005, Melbourne, VIC, Australia.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website:


What role does kinship and family structure have in the adaptive capacity of Australian rangelands? We have been considering this question as part of the CSIRO's Complex Systems Science funded project "Social Adaptation and Ecological Uncertainty" and in this paper we present our concept on how to treat the problem as an agent-based model.

Human-based networks disseminate information, help enforce social norms, facilitate trade, and so forth. An additional role of networks in many rural settings is linking otherwise fragmented ecological nodes (Janssen et al. in press). In rangelands, ecological units which have been fragmented and disconnected through conversion to pastoral land are linked through human networks (McAllister et al. in press). This is significant because rangeland resources are highly variable in time and space, and some networks that allow the movement of stock around the landscape have the potential to reconnect fragmentation units of land and hence buffer the effects of resource variation. Kinship based networks may help in this buffering role, and these networks are the focus of this paper.

Here we present 'KinModel', an agent-based model exploring kinship networks in rangeland landscapes. The model is constructed using the agent-based modelling platform RePast (version 3, Java). We devote this paper to outlining our concept, preliminary implementation, and data sources for a case-study rangeland system, the Dalrymple Shire.

The key elements of the model are:

a rangeland landscape, with variation of resources in time and space (Figure 1); and

pastoralists, who can marry, produce offspring and bequest their properties.

While our methods are still developmental, details are starting to emerge. One hypothesis we wish to consider is that because labour was initially in short supply in many Australian rangeland systems, long surviving families should be associated with families with many sons. Our model shows this argument could be flawed in two ways. First, brides were is shorter supply than labour, so daughters had a much greater chance of marring into successful pastoral families and building a larger kin-based network. Second, if we assume that all sons (and daughters for that matter) shared inherited land (i.e. not just the eldest son), then large families could ultimately split the enterprise into parcels too small to individually make a living. Small enterprises are limited in their capacity to buffer resource variation. Even if a small enterprise returns enough profit on average to support itself, it is at risk of collapsing economically in poor years if its networks are not strong enough to help in the buffering process.

Item ID: 42653
Item Type: Conference Item (Research - E1)
ISBN: 0-9758400-0-2
Keywords: kinship, rangelands, complex adaptive systems, agent-based modelling
Funders: CSIRO Complex Systems Science
Date Deposited: 10 Feb 2016 07:43
FoR Codes: 05 ENVIRONMENTAL SCIENCES > 0502 Environmental Science and Management > 050205 Environmental Management @ 100%
05 ENVIRONMENTAL SCIENCES > 0502 Environmental Science and Management > 050206 Environmental Monitoring @ 0%
SEO Codes: 96 ENVIRONMENT > 9605 Ecosystem Assessment and Management > 960510 Ecosystem Assessment and Management of Sparseland, Permanent Grassland and Arid Zone Environments @ 100%
Downloads: Total: 4
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page