Using passive track stations to determine the movement and activity of mid-sized mammalian predators and white-tailed deer in northern Iowa

Abstract

<p>I estimated activity and movement direction of striped skunk (Mephitis mephitis; hereafter, skunk), raccoon (Procyon lotor), red fox (Vulpes vulpes; hereafter, fox) and white-tailed deer (Odocoileus virginianus) as a function of landscape variables using un-baited (passive) track stations along the edges of blocks of grassland (Block Edge) and at various distances away from blocks of grassland (Isolated) in north central Iowa. Specific landscape variables within a 500-m radius for predators and a 1000-m radius for deer of track station locations were quantified using aerial imagery and a geographic information system for inclusion as predictor variables in the general models for isolated and block edge sample units. Logistic regression with repeated measures and Akaike weights were used to determine the influence of landscape composition and configuration variables on predator activity. The importance and relationship between landscape variables and the activity of individual predators differed depending on species and location (Block edge or Isolated). When all predator activity on isolated sample units was assessed, activity was negatively related to the distance to a block of grassland habitat and length of strip habitat, whereas area of pastureland positively influenced predator activity. The distance to woodland, area of woodland, area of pastureland and edge density of grassland habitat each had a negative influence on predator activity along the edges of blocks of grassland. Distance to woodland and area of pastureland had a negative influence on deer activity on isolated sample units, whereas area of grassland positively influence deer activity. More than 74% of predator and deer movement along corners of blocks of grassland was perpendicular to the grassland - row cropland edge, whereas <20% of deer movement direction and <10% of predator movement direction along straight edges was perpendicular. Models that use landscape features to predict wildlife activity could be used to better understand predator-prey interactions as well as reduce detrimental effects of wildlife on other species or property.</p>