Insect-mediated apparent competition between mammals in a boreal food web

<p>Datasets generated and analyzed within the study area located in the Côte-Nord region of Québec, Canada. "DataFinal_Vegetation.csv" was used to evaluate the availability of deciduous vegetation (column <em>Cover_Deciduous</em>) in stands impacted by the spruce budworm (SBW, severity index values presented in column <em>SBW_SevCum</em>) outbreak and test our prediction that the reduction in canopy cover caused by SBW in coniferous stands would result in greater deciduous vegetation. We determined the percentage cover of deciduous vegetation for two height strata ([0-1m] and [1-3m] classes, column <em>Stratum</em>).</p> <p>"DataFinal_Moose_density.csv" was used to assess the influence of SBW outbreak (<em>SBW_SevCum</em>) on moose abundance by comparing moose densities (<em>Density</em>) before (<em>year </em>= 2006) and during SBW outbreak (<em>year </em>= 2018). We also determined the effect of the percentage cover of deciduous vegetation (<em>Broadleaf</em>) and the percentage cover of salvage-logged forest stands (<em>SBW_cut_2018Factor</em>) on moose densities.</p> <p>"DataFinal_RSF_Caribou.csv" and "DataFinal_RSF_Wolves.csv" were used to develop Resource Selection Functions for caribou and wolves to assess habitat selection. RSFs compare resource characteristics of observed (scored 1) and random (scored 0) locations presented in column <em>case</em>. Habitat characteristics (<em>columns conif_dense, conif_open, mixed, open, other, burned0_5, burned6_20, burned21_, cut0_5, cut6_20, cut21_</em>) was extracted from the Canadian National Forest Inventory (NFI) forest cover maps. Land cover maps were updated every year by adding roads, recent (&lt;5 years), regenerating (6–20 years) and old (21–50 years) cutblocks/fires based on information provided annually by local forestry companies and from the Canadian National Fire Database (CNFDB). SBW outbreak was characterized by its cumulative severity (<em>SBW_SevCum</em>), and also if infested stands were logged (<em>SBW_Cut</em>). As our study focused on the survival of boreal caribou, we added the mortality status for each individual  (<em>DeadStatus, </em>remained alive throughout the study = 0, died = 1<em>) </em>and interaction terms (i.e. the multiplication of two columns) <em>Dead_SBW_SevCum </em>and <em>Dead_SBW_Cut. </em></p> <p>Finally, we used "DataFinal_SurvivalAnalysis.csv" to evaluate whether the proportion of natural (<em>Prop_burned</em>) and anthropogenic (<em>Prop_Cut0_5, Prop_Cut6_20, Prop_SBW_cut</em>) disturbances within winter home range had an effect on the mortality risk of caribou (determined with the mortality status column <em>Dead : </em>remained alive throughout the study = 0, died = 1). For this analysis, we determined the individual ID (<em>ID</em>), the years during they were monitored (<em>Year</em>), and the time was defined as the total period of monitoring for that individual (<em>Time</em>). </p>