Efeitos da fragmentação florestal sobre a riqueza de espécies de aves no Centro de Endemismo Pernambuco, Nordeste do Brasil

Abstract

The literature on habitat fragmentation is broad and diverse, reflecting the complexity of their ecological impacts. These impacts depend on spatial scale, the degree of habitat loss, isolation and size of the fragments, as well as the nature of the landscape matrix. The Theory of Island Biogeography has been cautiously applied to habitat fragments by assuming that fragments behave as islands in a sea of uninhabitable landscape. The Atlantic Forest of Brazil is an example of a highly fragmented habitat, remaining only 11.4 to 16% of the original forest. The forests of the Northeast, and particularly the area around Murici, contain a large number of rare and endangered species. The objective of this study is to measure the influence of the physical characteristics of the fragments (e.g. size, shape and isolation) on the bird communities of the Atlantic Forest in an area of 116,100 ha, the APA Murici (09°15'14''S, 35º47'59''W), Pernambuco Endemism Center, Northeast Brazil. An avifauna survey was conducted in 15 remaining (18 sites) divided into three size categories: six small (15 ha < SFR < 100 ha), six medium (100 ha ≤ MFR < 1,000 ha) and three large (LFR ≥ 1,000 ha). In the absence of patches larger than 5,000 ha, the three large fragments was considered as "control area" with two sampling sities each. Were demarcated 228 points counts of ilimited radius and at least 150 m from all other points. Data collection occurred from February to November 2012, in the first three hours of the day. Each site was visited three times and just eight points was randomly chosen for sampling visit. The length of stay in each point count was ten minutes. The total effort was 54 days in the field, where we sampled 432 point counts (72 hours). We used Ecotone 8 x 28 mm binoculars and a recorded Marantz PMD 661 with Sennheiser ME66 microphone. Rarefaction curves were constructed to verify the sampling sufficiency, and Chao 1 estimates were obtained using the program ESTIMATE S to estimate species richness for each fragment. The influence of the characteristics of the fragments in species richness was analyzed by multiple regression analysis. The metrics were calculated for the fragments in FRAGSTATS 3.3. We recorded 4,806 specimens belongiong to 130 species (12 Orders and 33 Families) in 18 areas sampled. Were observed 37-64 species and predicted 45.3-94.2 species per fragment. Respectively for large, medium and small fragments, the observed richness was 103, 88 and 89 species, while the predicted richness was 117.2, 96.0 and 101.1 species. Regarding the use of habitat, 88 species (26 taxa threatened) were classified as forest dependent, 31 as semi-dependent and 11 as forest independent. There was a strong positive correlation between (log) fragment size and species richness (r = 0.570, n = 18, p = 0.014) and number of species that depend on forest (r = 0.818, n = 18, p < 0.001). The bird species richness of forest independent and semi-dependent were higher in the fragments with more edge effect, usually small, isolated and/or of the more complex shape. Moreover, not all species classified as "forest dependent" be demonstrated highly sensitive to fragmentation. Fragment shape was the fragment metric that best explained the species richness of forest-dependent birds in each fragment.