Landscape changes during the first 3 decades of the Breeding Bird Survey (BBS) may account for the Painted Bunting's declining population trend. In the southeastern U.S., it is estimated that this bunting has declined 3.5 % per year since 1966. I collected landscape data centered on identical 5-stop areas (n = 33, 306 ha each) of the BBS during early (1960s - 1970s) and late decades (1980s - 1990s). Peak 30-yr counts for Painted Buntings were found at the center of the 5-stop areas. I used stepwise multiple regression analysis to model the mean number of Painted Buntings (in the area during 3 yr, dependent variable) associated with landscape metrics (independent variables). During the early decades the average amount of edge on developed land (p = 0.10), average patch size of agriculture land (p = 0.01), average size of shrub-scrub and young forest (p = 0.09), and average amount of edge for emergent wetlands (p = 0.03) explained 40% of the variation in Painted Buntings counts. In the late decades average amount of edge on developed land (p = 0.04) and average amount of edge on emergent wetlands (p = 0.005) explained 35% of the variation in Painted Bunting counts. Large losses of agricultural land (proportion = 0.177 to 0.094), which was developed or converted to intensively managed pin plantations, may have reduced potential bunting breeding habitat. Shrub-scrub and young forest habitat was constant (proportion = 0.136 to 0.134) but did not affect mean counts of buntings in the late decades. Protected emergent wetlands remained constnat also from the early to late decades (proportion = 0.056 to 0.06) and may provide habitat to maintain a smaller Painted Bunting population. At this time, it's unclear how develped land, which is increasing (proportion = 0.036 to 0.088), may be affecting the Painted Bunting population in GS and SC.
