Accounting for biological variability and sampling scale: a multi-scale approach to building epidemic models

When one considers the fine-scale spread of an epidemic, one usually knows the sources of biological variability and their qualitative effect on the epidemic process. The force of infection on a susceptible unit depends on the locations and the strengths of the infectious units, and on the environmental and intrinsic factors affecting infectivity and/or susceptibility. The infection probability for the susceptible unit can then be modelled as a function of these factors. Thus, one can build a conceptual model at the fine scale. However, the epidemic is generally observed at a larger scale and one has to build a model adapted to this larger scale. But how can the sources of variation identified at the fine scale be integrated into the model at the larger scale? To answer this question, we present, in the context of plant epidemiology, a multi-scale approach which consists of defining a base model built at the fine scale and upscaling it to match the scale of the sampling and the data. This approach will enable comparing experiments involving different observational processes.

Testing mutual independence between two discrete-valued spatial processes: a correction to pearson chi-squared

A common feature of data collected in environmental and earth sciences is that they typically exhibit spatial autocorrelation. Violating the assumption of independent observations can have dramatic effects on inferences derived from standard statistical methods. In this article, we examine the consequences of spatial autocorrelation on Pearson's chi-squared test of mutual independence between two categorical responses with a general number of classes. Correspondingly, we suggest a simple modification to the standard test statistic that allows for spatial autocorrelation. Our modified statistic is based on a first-order correction factor and thus provides only an approximate test. However, we show by Monte Carlo simulation that this approximation results in satisfactory inferences in several situations of practical interest. The usefulness of the method is displayed through an application to categorical data arising in the study of the relationship between the distribution pattern of plant species and woodland age in a forest in northern Belgium.