Date of Award


Degree Type


Degree Name

Master of Science (MS)


Biological Sciences


Grassland bird conservation should be a priority in the northeastern United States because many grassland species have declined since 1966. During 2004 and 2005, I examined rates of depredation on artificial grassland bird nests in two cool season grasslands, I (98 ha= large) and C (8.5 ha= small), at Iroquois National Wildlife Refuge (INWR). I established 50 m by 50 m gridded study plots in each grassland. One artificial nest was randomly placed in each grid section, and baited with a House Sparrow (Passerdomesticus) egg; half of the nests also received a clay egg to aid in identification of nest predators. Concurrently, I searched for and monitored real nests of Savannah Sparrows (Passercu/us sandwichensis) and Bobolinks (Dolichonyx oryzivorus). Vegetation measurements included percent cover, total hits of vegetation on a narrow aluminum rod and Robel pole scores. I trapped small mammals to determine small populations of species that potentially depredate nests, and I used characteristic bite marks left in the clay eggs to identify nest predators. The larger I field consistently exhibited a greater proportion of successful nests than the smaller C field. Daily survival rates of artificial nests averaged 0.914. The most frequently identified small mammal nest predator, based on clay egg bite mark identification, was Microtus pennsylvanicus, which was also the most frequently caught small mammal. In I field successful nests had significantly greater cover, higher Robel scores, and higher total hits of vegetation. In C field, there was a trend towards greater cover being related to greater nest success. Binomial logistic regression suggested that for nests in I field, year and the year*Robel interaction significantly affected nest success: 2004 nests, and those with higher Robel scores, were more successful: Year*distance to forest edge and year*cover interactions approached significance; nests placed in 2004 that were further from the forest edge and with more cover tended to be more successful. ln C field, year and distance to nearest hedgerow were the most important variables, although neither was significant and only year approached significance; nests in 2004 tended to be more successful than in 2005. Results from my study support much published data, and indicate that grassland bird nest success is greater in larger fields. Although artificial nest success increased with vegetative cover and density, these results should not be taken to indicate that managers should attempt to increase grassland density and height too much; common grassland birds in the Northeast nest preferentially in relatively low vegetation ( <1 m high). My data is consistent with results from other studies that suggest that when managing for grassland birds, one should focus on not mowing too frequently, and not mowing until after nestlings have left the nest by mid July. Ideally, grassland habitat patches should be greater than 50 ha (125 acres) in size, with shapes that minimize edge effects. Leaving some dry vegetation litter that can subsequently serve as cover may be beneficial as well (Swanson 1998).