Book contents
5 - Dealing with spatial autocorrelation
Published online by Cambridge University Press: 29 July 2009
Summary
Introduction
The familiar procedures of parametric statistics are based on the assumption of independence of the individual observations in the data under scrutiny, but in ecological data the assumption of independence is often violated and we need to understand the effects of a lack of independence. A lack of independence can arise because, in the natural world, things (samples, observations, etc.) that are closer together sometimes have a tendency to be more similar than those that are further apart, a phenomenon known to geographers as ‘Tobler's Law’ (Tobler 1970, and see Chapters 1 and 3). We refer to this lack of independence as ‘spatial dependence’ in the data (see Chapter 1), whatever the cause. One source of this phenomenon is autocorrelation in the data due to causal interactions within the measured variable itself; for example, in studying species distribution and abundance, the abundance of a single species may be spatially autocorrelated because of constraints on the organisms' mobility and dispersal. This kind of autocorrelation is sometimes called ‘true autocorrelation’ (see Chapter 1), but it might be more accurate to refer to it as ‘inherent autocorrelation’ (‘autogenic’ might be even more accurate, perhaps, but more unwieldy). The descriptor is to distinguish this phenomenon from ‘induced spatial dependence’ (see Chapter 1), where the observed variable (e.g. species abundance) has a functional dependence on an underlying variable (e.g. soil moisture or nutrient content), which is itself autocorrelated (cf. Legendre et al. 2002).
- Type
- Chapter
- Information
- Spatial AnalysisA Guide for Ecologists, pp. 212 - 255Publisher: Cambridge University PressPrint publication year: 2005
- 1
- Cited by