few small edits wrt figure and citing

.Renviron

@@ -1 +1,2 @@
 R_LIBS=~/Dropbox/R_library
+

vignettes/Introduction.Rmd

@@ -41,7 +41,7 @@ Environmental response curves, for example, illustrate how the probabilty of spe
 
 These species-environment relationships can be used to map the distribution of species in *geographic space*. Figure 1b shows a predicted distribtion for the Ebola virus [@pigott14]. Mapping the current distribution of a species can be useful for planning responses to public health emergencies or prioritising habitat conservation, for example.
 
-We can also use SDMs to predict the future distribution of species. For example, we might need to know how changing climate will shift the distribution of a species. By using future climate scenarios to create environmental covariates we can predict the future distribution of species. We could also identify the potential future distribution of an invasive species, in current or future climates. Elith et al [@elith10] did just this for the invasive cane toad, *Bufo marinus*, in Australia. Figure 1c illustrates the current distribution of the cane toad (black line) and the predicted future distribution (grey area).
+We can also use SDMs to predict the future distribution of species. For example, we might need to know how changing climate will shift the distribution of a species. By using future climate scenarios to create environmental covariates we can predict the future distribution of species. We could also identify the potential future distribution of an invasive species, in current or future climates. @elith10 did just this for the invasive cane toad, *Bufo marinus*, in Australia. Figure 1c illustrates the current distribution of the cane toad (black line) and the predicted future distribution (grey area).
 
 ```{r echo = F, out.width= '650px', fig.align = "center", fig.cap="*Figure 1. Species distribution modelling aims: ecological inference (a; figure from @redfern17, photo from @cetus), predicting current species distributions (b; figure from @pigott14), or future distributions (c; figure from @elith10, photo from @fraser-smith).*"}
 knitr::include_graphics("../vignettes/Images/Figure_1_litSDM.png")
@@ -56,7 +56,7 @@ A species distribution is typically considered the result of three factors: the
 
 We have illustrated the SDM fitting process in a flowchart below (Figure 2). We correlated presence data for the Carolina wren in the United States with a suite of environmental covariates, such as the cover of different forest types. We've mapped species presence records onto geographic space (Figure 2a; black dots) and a random selection of background points (brown dots), which we used as pseudo-absences in our model. We've also plotted these same data in environmental space against two covariates (Figure 2b): percentage of deciduous and mixed forests. By correlating presence-background data with environmental covariates, we've predicted the Carolina wren's probability of occurrence in environmental space (Figure 2c). In the final plot (Figure 2d), we've mapped these probabilities back onto geographic space. The Carolina wren is most likely to occur in southeast USA (purple) and least likely to occur in the northwest (yellow). By modelling species distributions, we can begin to understand where species are likely to occur and why.
 
-```{r Figure_4, echo = FALSE, out.width= '650px', fig.align = "center", fig.cap="*Figure 2. Species distribution modelling process. Presence-background points in geographic (a) and environmental space (b; P = presence, B = background), and predicted probability of occurrence in environmental (c) and geographic space (d; colour scale represents probability of occurrence).*"}
+```{r Figure_4, echo = FALSE, out.width= '550px', fig.align = "center", fig.cap="*Figure 2. Species distribution modelling process. Presence-background points in geographic (a) and environmental space (b; P = presence, B = background), and predicted probability of occurrence in environmental (c) and geographic space (d; colour scale represents probability of occurrence).*"}
 knitr::include_graphics("../vignettes/Images/Figure_2_SDMtheory.png")
 ```