<b>Wildlife Monitoring and Response to Silviculture Within Forested Ecosystems</b>

<p dir="ltr">Silvicultural treatments change forest structure and impact wildlife populations. Wildlife response to forest management can, in turn, impact forested ecosystems. Sciurid response to silviculture, for example, can affect forest regeneration because their diet is heavily reliant on mast-producing species. Response to silviculture can also be an important indicator of population health, making it desirable to monitor wildlife populations. However, monitoring wildlife poses risks both to wildlife and personnel. As such, it is important to use field techniques that minimize harm to wildlife and produce reliable population estimates. One way in which wildlife can be monitored is through assessing their behavioral response to silviculture. Wildlife, such as <i>Sciurus carolinensis, Odocoileus virginianus, </i>and <i>Procyon lotor,</i> may adjust their behavior in response to forest management. In this investigation, our objectives were to (1) summarize the current knowledge relevant to Sciuridae populations within managed forests, (2) compare the efficacy of abundance indices as alternatives to mark-release-recapture (MRR) abundance estimates, and (3) assess the impact of silviculture on wildlife behavior. To achieve our first objective, we systematically reviewed the literature and conducted a meta-analysis of sciurid population response (ln(SPR)) to silviculture, mast production, and study design. Key words in the systematic review included terms related to genera within the Sciuridae, forest management, and mast. Of the papers returned, 13 could be included in the meta-analysis and an additional paper was added that was relevant but not returned by the systematic search. The search results were biased towards North American sciurid species and study design explained most of the observed variation in the ln(SPR). For our second objective, we live and camera trapped <i>Tamias striatus </i>and <i>Peromyscus</i> within 6 x 6 trapping grids during the summer of 2022 in central Indiana. <i>T. striatus </i>and <i>Peromyscus</i> were ear tagged during live trapping, but individual identification was not possible for camera trapping. We used linear regressions to compare abundance estimated from simple, closed population models and minimum number known alive (MNKA) / 100 trap nights (TN) with a series of abundance indices. Captures / 100 TN, calculated from unmarked, live trapping data provided the best alternative to MRR abundance estimated for both species. Of the indices calculated from camera trap data, instantaneous sampling models and occupancy models performed the best for <i>T. striatus</i>, but the estimates did not explain enough variation to be considered a viable alternative. No indices calculated from camera trap data performed well as alternatives to MRR abundance estimates for <i>Peromyscus. </i>To achieve our third objective, we deployed 297 camera traps in three-camera arrays within central Indiana. Where possible, all images were tagged to the species levels and focal species (<i>S. carolinensis,</i> <i>O. virginianus </i>and <i>P. lotor</i>) were also tagged with behaviors using species-specific ethograms. We used homogeneity of proportions of the relative frequency of behaviors across silvicultural treatments, on singular cameras vs pooled within the array, and before and after baiting cameras. Expected proportions of behaviors deviated from that expected for <i>S. carolinensis </i>and <i>O. virginianus </i>among management types, and before and after baiting for all focal taxa. Pooling records across cameras increased sample size, but did not change the observed behavior profiles. Our research provides insight into the challenges to summarizing literature on sciurid response to forest management and to using indices of abundance calculated from camera trap data as an alternative to MRR abundance estimation. We also provide evidence that wildlife demonstrates and behavioral response to forest management and camera baiting. Reliable estimation of population parameters is important for monitoring populations, and contributing to the knowledge base surrounding wildlife response to silviculture.</p>