Nowadays, monitoring process is simply to ensure about preserving the steadiness of the relationship between the main variable of interest (output) and one or more environmental variables or control variables (inputs) over the time, hence the application of classical SPC is not convenient.
Indeed such an approach was recently developed by Coop et al. [22] to assess evidence for selection at loci showing unusually strong correlations with one or more environmental variables (including subsistence variables), controlling for the effect of population structure [15].
Our results (Table 2) show that there are significant correlations (based on the two standard error limits) with LST at lag 2 and 5, rainfall at lag 3 and relative humidity at lag 0 to 3. Subsequently (S ARIMA models were estimated with one or more environmental variables included.
In contrast, for more exploratory studies where species and habitats are not known, ideally many more environmental variables would have to be taken into account.
On the basis of linear regression alone, we identified 12 loci in the coastal lineage that were significantly correlated with one or more environmental variables (Table 2; Table S3).
In the former, the researcher directly selects on the trait of interest whilet in the latter, the researcher attempts to alter the selective landscape by manipulating one or more environmental variables.
