Modelling settlement futures: Techniques and challenges

Limitations of secondary data collected by external agencies for examining demographic change in sparsely populated areas (SPAs) are well documented in this volume (especially Chapter 7) and elsewhere (for example, Taylor, 2011). Even robust data collections specifically designed to provide settlement level analysis, such as population censuses, present with a diversity of issues. Broadly, these pertain to enumeration issues, conceptual issues, collection issues, changes to collection methods over time, or simply unexplained events at individual settlements (Koch and Carson, 2012; Taylor et al., 2011). Without local knowledge of specific issues under these themes (should they exist), downstream analysis and the dissection of demographic change for settlements is obstructed by a lack of distinction between ‘real’ demographic shifts and those which simply represent the outcome of one or more of these influences. Alternatively, ‘black swan’ events (where the event – like a major shift in the sex ratio for a settlement over a short period of time) may be neither predicted nor traceable to known factors. Most often it is a combination of these, and often the precedent cause is relatively unclear, making the task of modelling time series and projecting future settlement level demographics a hefty challenge. There is value therefore in sourcing and utilising tacit knowledge and internally compiled (by settlement residents or organisations) datasets to inform modelling on the causes and consequences of demographic change for individual settlements, both past and future. Such knowledge is built up over...