TWEL Jennifer Smith-Castro Thesis

Impacts of recreational trails on the nesting success of passerines in a riparian forest system
Jennifer R. Smith-Castro, MS
Advisor: Amanda Rodewald
Most public lands, particularly those in urban areas, are designed to meet both social and ecological needs. Although recreation has traditionally been viewed as relatively harmless to animal communities, trails within urban parks may influence the behavior of wildlife through human disturbance and through changes in the distribution of vegetation. Because park managers must balance the competing interests of conservation and recreational uses of parks in urban areas, consideration must be given to the potential consequences of human disturbance to breeding birds.
The central purpose of this study was to determine the extent to which trails affect nest predation and evaluate possible causes of those effects. Specifically, I tested the following three hypotheses about the impacts of human use of trails on breeding birds: 1) Human use of trails in urban forests reduces nest survival by reducing parental attendance rates, thereby leaving nests more vulnerable to predation; 2) Relationships between nest survival and trails derive from modified vegetation surrounding a nest, and these changes in habitat drive trail-related impacts on breeding birds; 3) Human use of trails in urban forests influences the sensitivity of breeding birds to disturbance and induces changes in nest placement through habituation and self-sorting behavior. From April – August 2006 and 2007, I monitored the fate of 263 Northern Cardinal nests and quantified vegetation structure and composition around nests and at random plots within riparian forests in central Ohio. Sites were located in urbanizing landscapes and contained paved and unpaved recreational trails. Trail cameras were utilized at 6 sites to estimate the amount of human recreational activity. One hour parental attendance observations were conducted at 125 nests to estimate variation in nest attendance. Two experimental trials were conducted on 63 nests such that Flight Initiation Distance (FID), as an index of sensitivity, was recorded as each nest was approached either directly or along a trail. In addition, FID was collected during routine nest checks where the nest was approached directly (n = 160).
Results showed that birds were 6x more likely to flush when the nest was approached directly than when an observer passed along a trail. Nest height mediated the tendency to flush somewhat, as higher nests were less likely to flush, but flush tendency was not related to distance to trail. Interestingly, the distance at which a bird flushed (FID) was not significantly related to either distance to trail or nest height. I used an information theoretic approach, incorporating a set of models into a logistic-exposure method to model daily nest survival. Estimated daily survival rates for Northern Cardinals were similar across sites, and variation in daily survival rate of Northern Cardinal nests was not well explained by FID, parental attendance, trail usage, or site. Rather the best explanatory model contained only the variable of nest height, though several alternate models, including one containing distance to trail as well as a null model, were similarly ranked. I used canonical correlation analysis to examine the extent to which vegetation variables were related to distance to trail separately for nest and randomly-located plots. Results suggest that birds selected nest locations surrounded by greater amounts of native vegetation than expected when farther from trails. In addition, birds selected nest sites that were lower to the ground and more interior in the nest plant when farther from trails. Thus, recreational trails appeared to indirectly influence reproductive behavior by altering nest site selection. Relatively few studies have measured the impact of recreational disturbance to nesting passerines, and this study illustrates that recreational use has the potential to impact, even those species thought to be urban-adapted. Nevertheless, because nest attendance was not related to daily nest survival rates, this study fails to provide evidence of negative reproductive and, hence, potential population level consequences of behavioral responses to human disturbance. Thus, my work suggests that, at least for certain synanthropic species, recreation may indeed be compatible with conservation.