BIRD COMMUNITIES IN TWO FRAGMENTS OF SEMIDECIDUOUS FOREST IN RURAL SÃO PAULO STATE POZZA

A quali-quantitative survey was done in two fragments (75 and 100 ha) of semideciduous forest in rural São Paulo State. The aim was to characterize the bird community according to richness, abundance, and occurrence frequency in these areas. The qualitative survey showed 145 species in the Estação Ecológica de São Carlos – EESCar (Brotas) – and 173 in the Fazenda Santa Cecília – FSC (Patrocínio Paulista), while the quantitative survey showed the presence of 60 and 72 species in EESCar and FSC respectively. The isolation and the lower environmental quality of the EESCar fragment may be responsible for the lower number of species in this area compared to that of FSC. Abundance index value analysis (IPA) showed that both areas have a large number of species with low IPA and few species with intermediate or high IPA compared to the pattern observed in other surveys. At FSC, a larger number of occurrences of species in danger of extinction in São Paulo State was also observed. Apparently, the FSC fragment had better environmental quality for sheltering a larger number of species, including the endangered ones. The study of the community of birds is important in planning management and conservation of natural areas.

The remaining areas can, however, provide information about the original ecosystems of the State (Rodrigues & Shepherd, 1992), and should be preserved as genetic banks and opportunities for reforestation (Araújo & Teixeira, in press).
Reduction and isolation of fragmented areas (Andrén, 1994) result in decreases of species and bird populations (Laurance et al., 2000;Stouffer & Bierregaard, 1995).Many authors have emphasized the importance of studies of bird communities as indicators of ecosystem quality (Ramos, 1997;Moraes, 1997;Simon & Ribon, 1997), as birds constitute one of the most suitable animal groups for this purpose use as indicators environmental quality (Machado, 1995).
The present study characterizes (qualiquantitatively) the structure of bird communities as to richness, abundance, and frequency of occurrence of the species in two fragments of semideciduous forest in rural São Paulo State.
EESCar is a conservation unit situated in the middle east of São Paulo State, in Brotas, between geographical coordinates 22º05' and 22º7'S and 48º00' and 48º05'W.An isolated fragment of 75.26 ha, it borders Santana Reservoir to the north; the other sides are bordered by sugarcane monoculture.This area was chosen because it is a preservation area close to UFSCar (30 km), where this study was developed.
The FSC is situated in Patrocínio Paulista, in northeastern of São Paulo State, between coordinates 20º46'2''S and 47º14'24''W.A remnant, it consists of 100 ha privately owned and preserved.The forest is bordered by 300 ha of cerradão, in addition to pastures and humid areas.This area was chosen because it is a major preserved area of northeastern São Paulo State.

Methodology
Quali-quantitative surveys were based on the sampling method by points (Vielliard & Silva, 1990).
In the period from September 2000 to August 2001, each studied area was visited 24 times, an average of 6 hours per visit, totaling 150 observation hours for each area (Table 1).

Quantitative survey
To carry out this survey, two tracks 600 m long each in each were used fragment.Five points were chosen along the tracks, 150 m distant from each other.The order of the sampling was established by raffle and the observation time in each point was 15 minutes, according to methodology adapted from Vielliard & Silva (1990).The species and respective number of individuals of each were identified auditorily and visually and noted.When doubts existed about the number of individuals, only one of each species was recorded to prevent overestimates.A mini tape recorder was used to record unknown vocalizations, for later identification.This survey was restricted to the points chosen along the tracks; it began at about 6.00 a.m.

Qualitative survey
The survey was carried out employing the same tracks used in the previous survey.The surrounding forest environment, including humid areas, pastures, cerrado and plantation areas were considered.All species observed, their locations, contact type (visual and/or auditory), and number of individuals were noted.All species observed in the quantitative survey were included in this survey.

Qualitative survey
A total of 145 species were recorded in or near EESCar and 173 in FSC.The number of species recorded is significant when compared to that of other surveys using the same methodology and sampling time equal to or higher than ours (Table 2).In both areas of São Paulo State were identified endangered species (SMA, 1998).In FSC the following species were identified: Crypturellus undulatus, Sarcoramphus papa, Propyrrhura maracana, Aratinga auricapilla, Amazona aestiva, Pteroglossus aracari, Antilophia galeata and Saltator atricollis.In EESCar Oryzoborus angolensis was identified.The frequency of occurrence of these species as well as the maximum number of individuals recorded are in Table 3.
The occurrence of a large number of endangered species in FSC may indicate that this fragment presents better environmental quality than does EESCar and is an important area for feeding, reproduction, and resting of these species.

Frequency of occurrence
The frequency of occurrence (FO) relates the proportion of the days on which the species were found with the total number of survey days, allowing us to conclude if a given species can be regularly found or not (Vielliard & Silva, 1990).
Species occurrence frequency was classified as shown in Table 4.
The FO above 75% includes resident species representing 4.8% of the species in EESCar and 20.35% in FSC.
A great part of the bird community (56.55% of EESCar and 41.86% of FSC) is composed of species with an FO inferior to 25%, appearing in less than six out of 24 visits.The lower FO or single occurrence of some species can be explained by their brief stay (rovers), such species may inhabit other environments and only occasionally exploit forest resources (occasional) or be they may migratory species.
The species recorded and their respective FO are in Table 5.

Number of species
A total of 60 species was recorded for EESCar and 72 for FSC, with a total of 115 samples collected in each area.

Number of contacts
In the115 sample total from each area, 630 contacts were recorded for EESCar with an average 5.4 contacts per sample, and 774 contacts in FSC for an average 6.7 contacts per sample.

Abundance index values (IPA) per species
The IPA per sampled species relates the average number of contacts of these species with the total number of samples.The IPA is a relative number but, like FO, it is comparable between measures of the same species for different dates, places, and communities (Aleixo & Vielliard, 1995).At EESCar the specific IPA varied from 0.008 (1 contact) to 0.408 (47 contacts); at FSC the IPA varied from 0.008 to 0.660 (76 contacts) (Table 5).
Both in EESCar and FSC few species had a high IPA and a great number of species showed intermediate and low numbers, compared to the pattern observed in other surveys (Vielliard & Silva, 1990;Aleixo & Vielliard, 1995;Almeida, 1997).Among species with a high IPA, some are conspicuous, like Herpsilochmus atricapillus, Basileuterus culicivorus, and Vireo olivaceus for far-reaching chant or constant vocalizations.Some species presented a high IPA in one area and a low one in the other.This was the case with Leptotila verreauxi, Cyclarhis gujanensis, and Pitylus fuliginosus, which showed a high IPA in EESCar and a low one in FSC.On the other hand, Chiroxiphia caudata and Synallaxis ruficapilla presented a high IPA in FSC and a low one in EESCar.This could be because those species have larger populations in the areas where they presented a higher IPA, making their identification easier.
Other species were found in just one area, but with an intermediate/high IPA number.This was true of Amazona aestiva in FSC and Baryphthengus ruficapillus in EESCar.A possible explanation is that due to their non-homogenous distribution in the original environment, their territories were not included during the fragmentation process (Bierregaard et al., 1992).This absence of given species in certain areas can also occur as a result of non-adaptation to a simplified environment or intensification of interspecific competition for limited or fragmented resources in their occurrence areas.Local extinction could also explain species lack in the studied areas, but real clarification depends on long-term studies on the ecology of these species.) ) It should be remembered that FSC is an area whose owners are always on the alert for harmful activity in this area, while EESCar lacks protection completely.
Even so, both areas are important for the conservation of bird communities and other animal groups.The current study can serve as a subsidy for management planning in these areas.
Fazenda Santa Cecília, which is environmentally protected, could be modified in order to become a conservation unit, governmental or private ("Reserva Particular do Patrimônio Natural") where as EESCar requires better management to qualify as a conservation unit.
Studies like this provide preliminary knowledge of fauna communities and a partial diagnosis of the environment quality of fragments studied.Further research can identify species and the fragments where they are found.
Localization of study areas.

TABLE TABLE TABLE TABLE TABLE 1 1 1 1 1 Months and days of data assessment. Months and days of data assessment. Months and days of data assessment. Months and days of data assessment. Months and days of data assessment. Place Area (ha) Number of species Sampling time (months)
Almeida (1997)a (1997).