The field sparrow breeds in a variety of vegetation types including brushy pastures, second growth scrub, forest openings and edges, Christmas tree farms, orchards, nurseries, and roadsides and railroads near open fields (Carey and others 1994). Field sparrow abundance increases in forested landscapes managed for early successional habitat (Yahner 2003), and birds commonly occur on reclaimed mines (DeVault and others 2002) and savanna restoration sites as well (Davis and others 2000). Field sparrow abundance is positively related to the size of old fields in Arkansas (Bay 1994). Field sparrows nest on or near the ground in early spring but may nest in saplings or shrubs later in the year. Brood parasitism rates vary geographically, but the field sparrow is generally a poor cowbird host. Parasitism rates are higher in thinned forest stands than in regenerating plantations (Barber and others 2001).
Field sparrows also use grasslands, although at lower densities than in shrub-scrub habitats (Horn and others 2002). Grass type affects habitat suitability, with warm-season grasses supporting higher abundance (Giuliano and Daves 2002, Walk and Warner 2000), nest density (Farrand 2005), and productivity than cool-season grasses (Giuliano and Daves 2002). Conservation Reserve Program fields serve as source habitat for field sparrows in Missouri (McCoy and others 1999).
The model predicting habitat suitability for field sparrows contains six parameters: landform, land cover, successional age class, canopy cover, density of small (<2.5 cm d.b.h.) stems, and presence of grassy landcover.
The first suitability function of the field sparrow habitat suitability model combines landform, landcover, and successional age class into a single matrix (SI1) defining unique combinations of these classes (
. We used the field sparrow habitat associations reported in Hamel (1992) to directly assign suitability index scores to these combinations.
We included canopy cover (SI2) and small stem density (SI3) as suitability indices in our model to account for the absence of field sparrows from closed canopy forests or forested sites with an open understory. We used data from Annand and Thompson (1997;
, respectively) to fit a quadratic function to canopy cover and a Gaussian function to small stem density for predicting suitability index scores for field sparrow habitat (
Field sparrows are often associated with grasslands where sufficient perches exist (Carey and others 1994, Kahl and others 1985); therefore, we included a habitat suitability function related to grasslands (SI4) in the model. Many useable grassland sites may not have sufficient woody cover to be classified as shrublands in the NLCD. We assumed natural grasslands were more likely to contain perches and dense grass nesting sites than pastures and hayfields. Thus, we assigned natural grasslands a suitability index score of 1.000 and pasture-hayfields a 0.500 provided they were ≤170 m of a wooded edge – a distance approximating a large field sparrow territory (Best 1977;
To calculate the suitability index score for field sparrow habitat in forested landcovers, we calculated the geometric mean of the suitability index scores relating to forest structure (SI1, SI2, and SI3). The suitability index score for grasslands (SI4) was then added to this value to determine the overall SI score.
Overall SI = ((SI1 * SI2 * SI3)0.333 + SI4)