Figure 4From: Determinants of the geographic distribution of Puumala virus and Lyme borreliosis infections in BelgiumEmpirical variograms and envelopes of residuals from non-spatial models. (a) Residuals from non-spatial negative binomial regression on PUUV infections. (b) Residuals from non-spatial negative binomial regression on Lyme borreliosis infections. Envelopes were computed by permutation of the data values on the spatial locations [43].Back to article page