TABLE 1.

A description of the fitted general linear models. The model number is the reference used in this paper, its form is presented as pseudo code in line with R syntax and the link function denotes the relationship between the linear predictor and the mean of the distribution. Models 1 and 2 were fitted using he glm2 package (Marschner, 2011) and Models 3:5 were fitted with the glm function from R's stats package (R Core Team, 2020)

Model ID	Model form	Link function
M1	$\mathit{\text{Event}}\mathit{Max}.\mathit{\text{Flow}}~\mathit{\text{Total Rainfall}}+\mathit{\text{Beaver Presence}}\times \mathit{\text{Site}}$	Identity
M2	$\mathit{\text{Event}}\mathit{Max}.\mathit{\text{Flow}}~\mathit{\text{Total Rainfall}}\times \mathit{\text{Beaver Presence}}\times \mathit{\text{Site}}$	Identity
M3	$\mathit{\text{Event}}\mathit{Max}.\mathit{\text{Flow}}~\mathit{\text{Total Rainfall}}\times \mathit{\text{Beaver Presence}}\times \mathit{\text{Site}}$	Log
M4	$\mathit{\text{Event}}\mathit{Max}.\mathit{\text{Flow}}~\mathit{\text{Poly}}\left(\mathit{\text{Total Rainfall}},2\right)\times \mathit{\text{Beaver Presence}}\times \mathit{\text{Site}}$	Identity
M5	$\mathit{\text{Event}}\mathit{Max}.\mathit{\text{Flow}}~\mathit{\text{Poly}}\left(\mathit{\text{Total Rainfall}},2\right)\times \mathit{\text{Beaver Presence}}\times \mathit{\text{Site}}$	Log