plr

A function that undertakes a Power Law Residual analysis. This method compares two time periods but can take multiple time periods and will complete all pairwise comparisons. The analysis returns the standardised residuals of a linear model. The model has the log number of cells occupied in the second time period as the dependent variable and the log number of cells occupied in the first time period as the independent variable. The results are therefore a relative measure of change.

# Load the library
library(sparta)

#load example dataset
data(ex_dat)

# Passing data as an R object
plr_out <- plr(Data=ex_dat,
               time_periods = data.frame(start=c(1980,1990,2000),end=c(1989,1999,2009)),
               min_sq = 5,
               site_col = 'hectad',
               sp_col = 'CONCEPT',
               start_col = 'TO_STARTDATE',
               end_col = 'Date')

The argument (min_sq) specifies the minimum number of squares occupied in the first time period in order for a trend to be calculated for a species. This ensures that potentially inaccurate large changes are not generated.

head(plr_out)

     CONCEPT     plr_1_2     plr_1_3    plr_2_3
1  Species 1 -0.96250478 -0.05908147  0.5422159
2 Species 10 -0.10893608  0.38260573  0.6172385
3 Species 11 -0.06959541 -0.69617627         NA
4 Species 12          NA          NA         NA
5 Species 13 -0.29504221  0.47243163  0.6205836
6 Species 14  0.44982084 -0.78746346 -1.0797575