01_device Bird diversity and abundance in the UP Diliman 1 *Corresponding author Science Diliman 20:1, 1-10 Spatial Patterns of Bird Diversity and Abundance in an Urban Tropical Landscape: The University of the Philippines (UP) Diliman Campus Benjamin Vallejo Jr.a*, Alexander Aloyab, Perry Ongc, Annette Taminod, Jonathan Villaspere aInstitute of Environmental Science and Meteorology, College of Science University of the Philippines, Diliman, Quezon City bNatural Science Research Institute, College of Science University of the Philippines, Diliman, Quezon City c Institute of Biology, College of Science, University of the Philippines Diliman, Quezon City dCollege of Science, University of the Philippines, Baguio City eDepartment of Geography, College of Social Science and Philosophy University of the Philippines, Diliman, Quezon City E-mail: bmvallejo@up.edu.ph Date submitted: February 3, 2006; Date accepted: May 7, 2008 ABSTRACT An analytical biogeographic analysis of urban bird diversity and abundance was conducted in the University of the Philippines Diliman campus from February to April 2005. Using the Jokimäki urban bird census technique on four different land use subplots we observed 36 species of birds and 4036 cumulative counts individuals of these species. The open field area had the highest number of species (23) while the residential area had the lowest species number (14). The residential area has the highest bird counts and the College of Science complex had the lowest number of counts. Linear regression analysis of landscape features on bird abundances reveals that the number of trees and buildings are most significant predictors of abundances. Spatial features and the trophic characteristics of the bird species are not significant factors to account for abundance. Trees and buildings affect the distribution and abundance of urban exploiting species Passer montanus and Pycnonotus goiavier with the former favoring built spaces and the latter favoring trees. Analysis of species area curves suggests that the bird community of the campus is not homogenized and that certain habitats allow uncommon species to persist. We recommend that in order to increase bird biodiversity existing green spaces must be preserved, fruit trees planted and in areas reserved for building development, pocket gardens and rooftop gardens be included in the landscape architectural design. Key words: birds, biogeography, urban ecosystems, University of the Philippines, Metro Manila Vallejo, et al. 2 Science Diliman 20:1, 1-10 INTRODUCTION Urban environments are generally areas characterized by a growing human population, pollution, and conversion of natural habitat into ‘built-up’ areas. Like any major Asian metropolis where at least 10% of the national population resides, Metro Manila has about 12% of the entire population of the Philippines (Ohmachi 2002). Metro Manila’s population growth rate is 3% per annum largely brought upon by migration from the provinces. Based on landscape statistics by the Philippines National Mapping and Resource Information (NAMRIA 2003), built-up areas in Metro Manila was estimated to be around 37.8% of the total land cover. Since urban areas are characterized as centers of human habitation, Philippine cities have not been considered as important areas for biodiversity conservation. Many studies have shown that metropolitan areas still support significant levels of animal and plant biodiversity which includes the occurrence of rare species in wooded, green spaces (Gyllin 1999; Szacki 1999; Ong et al. 1999). Landscape developers and architects have long realized the importance of green spaces in mitigating microclimatological effects of overheating (Ohmachi 2002) and in helping maintain humidity levels (Szacki 1999) in built-up areas. Green areas are generally locally fragmented and patchy in cities (Clemants & Moore 2003; Gilber 1992; Hobbs 1988) and this fragmentation may result in localized extinctions (Hengeveld 1993). In a recent study in a university campus in Brazil, habitat fragmentation resulted in a failure to sustain avian biodiversity (Manhães & Loures-Ribeiro 2005) Species richness decreased with increasing urbanization in Vancouver, Canada (Melles et al 2003). However, bird biodiversity loss in Metro Manila has been not well documented with the exception of possible extinctions recorded at the University of the Philippines Diliman campus in Quezon City. Four forest dwelling birds (mostly raptors) were recorded in the 1960s but have never been recorded since, making them locally extinct from the Diliman district ( Ong et al. 1999). Localized extinctions may be a result of species invasions by non-native species (Kowarik 1995; Blair 2004). Urban bird communities are usually characterized by the dominance of a few species (Beissinger & Osborne 1982; Marzluff 2001) and most of the species making up the communities are introduced. A low spatial variation of urban bird communities is expected and may probably result to more similar bird communities all over the world (Jokimäki et al. 1996). Urban birds are generally classified in three general categories: urban avoiders, suburban adaptable and urban exploiters (Blair 1996; McKinney 2002). A growing number of amateur bird watchers and clubs (e.g. Wild Bird Society of the Philippines) in the last 10 years have recorded about 117 bird species in Metro Manila. Some of the accounted species are listed as endangered (Collar et al. 1999) and were mostly seen in wooded areas of universities, memorial parks, reclamation sites and in watersheds (e.g. La Mesa Watershed). The numerous observations by amateur and professional birdwatchers may support the idea of the conservation value of Metro Manila’s green spaces. Nonetheless, there is not much scientific information on the distribution and habitat association (e.g. landscape ecological characteristics) of bird biodiversity. In this research we investigate the possible effect of landscape features on the patterns of diversity and abundance of bird communities in the 493 hectare University of the Philippines (UP) Diliman campus. Several researchers have recommended local scale investigations in studying bird-habitat relationships aimed at the development of effective management decisions for the conservation of birds in urban habitats (Jokimäki & Jokimäki 2003; Clergeau et al. 2001). METHODS Study area Our study area is the University of the Philippines Diliman campus in Quezon City (140 N, 1210 E), Metro Manila (Fig. 1). The campus is located 15 kilometers from downtown Manila. It is a fully functional community and a local government unit. The campus has faculty and student residential areas, academic use areas, parks and commercial areas. The campus Bird diversity and abundance in the UP Diliman 3Science Diliman 20:1, 1-10 daytime population is 40,000 composing of students, staff, employees and informal settlers. We use the campus as a model for Metro Manila’s urbanization. Geophysical studies by Midorikawa and Bautista (2002) categorized the area to be part of the ‘central plateau’ or Guadalupe Tuff region of Metro Manila with an elevation of at least 10 meters above sea level. Rapid urban development began as early as 1947 when the campus was transferred from Ermita district in downtown Manila after the Second World War. The climate type is tropical monsoonal (Type I) with a pronounced dry season from November to April, and a pronounced wet season for the remainder of the year. The area is protected from the northeast monsoon, but open to the southwest monsoon and tropical cyclones. Four survey sub-plots (20-25 hectares) were established within a 2 km × 2.5 km area. These plots were defined by existing university road networks and characterized by varying levels of land use and degree of infrastructure developments. Plots were spatially delineated as UP Academic Oval (OV), open space along University Avenue (OP), College of Science complex (CS), and the UP faculty and staff residential site (RE). Bird census technique We employed a bird survey designed for urban communities developed by Finnish biogeographers Jokimäki and Jokimäki (2003). We initially established four sub-plots (20-25 has.) within the UP Diliman campus which was normally surveyed during early mornings (0530-0800 H) and before sunset (1630-1800 H) at a pace of 10 ha/20 min walk. We surveyed each plot six times during the dry and hot season from February to April 2005. We conducted the census by walking a random zigzag route pattern in wooded and open spaces while following road pavements in the residential site. Careful bird counts were observed continuously through a fast-paced walk, taking note of bird movement across the route. This criterion is intended to avoid double or multiple counts of an individual. Birds which merely flew high overhead from Figure 1. The University of the Philippines (UP) Diliman campus in northeast Quezon City, Metro Manila. Bird survey subplots are indicated. Vallejo, et al. 4 Science Diliman 20:1, 1-10 the upper canopies or buildings were not included in the counts. Observations on bird species richness and abundance including microhabitat associations were noted on standard data sheets. All corresponding statistical analyses were based on cumulative counts of species recorded for the entire survey. The Jokimäki method is designed to estimate the abundances of mainly resident perching birds (Passeriformes) that are usually associated with tree canopies and urban landscape features. The method cannot be used to estimate the abundance of birds with large foraging areas such as raptors and to detect extremely rare or cryptic species. While raptors have been observed by other researchers in campus, their sighting frequencies are rare and it is likely that these birds are migrants passing over the campus. Ecological statistical analyses We used the non-parametric statistic Spearman’s coefficient (R s ) to determine the degree of correlation between species occurrence and abundance in all plots. To analyze bird community diversity, we used Shannon- Wiener’s (H’) and Simpson’s (D) Diversity Indices. Shannon-Wiener index was generally used in ecological studies concerned with the number and abundance of rare species while Simpson’s index for more abundant or common species (Peet, 1974). Hill’s modified ratio (E 5 ) was used to compute for bird community evenness, where a value approaching 0 indicates single species dominance. Thus, E 5 = N 2 - 1/ N 1 – 1 (1) where: N 1 = eH’ (exponential function of Shannon-Wiener index) N 2 = 1 / D (reciprocal of Simpson’s index) To measure the general trend in spatial species diversity among sub-plots, we obtained a species-area curve from nested plots. The derived curve describes the linear association of species richness with increasing sample size (Rosenzweig 1995). If a non-linear association emerges, then a case of homogeneity of bird fauna might exist between plots (Rosenzweig 1995). In order to estimate effects of landscape features on bird abundances, we log-transformed abundance data before using stepwise linear regression analysis with Statistical Package for the Social Sciences (SPSS Version 13.0). Log-transformations were done to stabilize variances and fulfill parametric statistical assumptions. Ecologists have long accepted that biotic communities generally follow a lognormal distribution as species abundance values are affected by a range of complex processes and interactions in a community (Waiten 1999). Species abundance was also subjected to simple linear regression analysis with UP-Diliman landscape variables (Ballad et al. 2002, unpublished data). Data on species richness and occurrence were used to compute qualitative community similarities among sub-plots (Sörensen index): C N = 2 jN /(aN + bN) (2) where aN is the total number of species recorded in Site A, bN, the total number of species recorded at Site B, while jN is the number of joint species occurrences for all sites. RESULTS Species richness and abundance A total of 36 species were accounted in four sub-plots in UP Diliman (this includes 3 species accounted through opportunistic observations, see Table 1). Twenty or more species occurred in 3 sites: College of Science, UP Oval, and the Open field site. Only 14 species were detected in the residential area. Philippine endemics comprise 14% of the total species list including the Lowland White-eye Zosterops meyeni, which only occurs in the Luzon Faunal Region (LFR). Other endemics were Pink-bellied Imperial Pigeon Ducula poliocephala, Philippine Hanging-parrot Loriculus philippensis, Philippine Coucal Centropus viridis, and Philippine Pygmy-woodpecker Dendrocopus maculatus. Migrants make up 19% while introduced species 8%. Three out of the 4 known introduced breeding bird species in the Philippines were present in the study sites. Bird diversity and abundance in the UP Diliman 5Science Diliman 20:1, 1-10 Bird density A total of 4,056 cumulative counts of individual birds were recorded for 33 species (Appendix 1). Positive correlation exists between abundance of birds and the number of site occurrences (R s = 0.902, n = 33, P < 0.01). The residential site (although with only 14 species) appeared to have the highest absolute bird density with of 1,685 individuals (Table 2). Dominant species were Eurasian Tree-sparrow Passer montanus, Yellow-vented Bulbul Pycnonotus goiavier, Brown Shrike Lanius cristatus, and Pied Fantail Rhipidura javanica comprising >5% of the total number of occurring birds in all areas. It can be noted that P. montanus (x = 520.8, SD = 374.4, n = 4) and P. goiavier (x = 179.5, SD = 56.9, n = 4) composes majority of avian populations in all study plots. Bird community diversity and evenness Shannon-Wiener’s index which provides a quantitative description of the intensity of species richness was highest at the OP site (H’ = 1.8704) followed by CS (1.8536), OV (1.7951), and very least at RE site (1.3196, Table 3). Simpson’s index was marginally highest at RE (D = 0.4337) with the other 3 sites having a range value of 0.24 to 0.26. Hill’s index of diversity evenness was highest at OV (E 5 = 0.6048), followed by CS (0.5472), OP (0.5172) and RE (0.4763). Linear regression analysis A low R2 value of 0.175 suggests the effects of scale in the regression model. The significant F-statistic from the ANOVA table (Table 5) shows the effectiveness of the regression model in explaining abundance as affected by various landscape variables (i.e. no. of trees, % cover of buildings and green spaces, feeding guilds). Coefficient values generated by the linear model indicates the number of buildings (á=0.014) and trees (á=0.004) as significant landscape predictors of bird abundance (Table 6). Other spatial attributes such as building and green-space cover appeared to be non- significant. Food preferences of bird species as indicated by feeding-guild values were also not significant. (1) CS (2) OV (3) OP (4) RE P. montanus 43.9% P. montanus 36.9% P. montanus 46.7% P. montanus 63.7% P. goiavier 20.6% P. goiavier 30.7% P. goiavier 14.6% P. goiavier 12.1% L. cristatus 9.4% L. cristatus 7.9% L. cristatus 9.0% R. javanica 9.3% R. javanica 6.1% R. javanica 7.7% G. striata 8.5% G. sulphurea 5.6% L. schach 4.3% G. sulphurea 6.0% R. javanica 5.1% L. cristatus 3.2% H. tahitica 3.8% C. livia 1.9% G. sulphurea 3.9% H. tahitica 2.3% G. gallus 2.5% H. chloris 1.5% G. gallus 2.6% G. gallus 1.2% G. sulphurea 2.5% Z. meyeni 1.5% H. tahitica 2.4% M. palustris 1.7% H. tahitica 1.4% L. schach 2.1% n= 1,685 individuals G. striata 1.3% L. schach 1.2% C. livia 1.6% S. chinensis 1.0% n = 635 individuals n= 810 individuals n = 926 individuals CS, College of Science; OV, Academic oval; OP, Open field site; RE, Residential site Table 2 Summary of dominant bird species expressed as percentage of the cumulative number of birds counted in surveys (n=6) of four sub-plots in UP Diliman Philippines Luzon This survey (UP Diliman) Total CS OV OP RE Total 572 391 36 22 20 23 14 Residents 198 137 19 14 11 14 8 Philippine endemic 176 100 4 2 3 1 1 LFRb endemic 25 25 1 0 1 0 0 Migrants 162 154 7 3 3 4 2 Introduced 4 4 3 3 2 3 2 aResidency data is taken from bird listings of Kennedy et al. (2000) bLuzon Faunal Region CS, College of Science; OV, Academic oval; OP, Open field site; RE, Residential site Table 1 Summary of species richness and residency status of birds detected from February to April 2005 in UP Dilimana. Vallejo, et al. 6 Community similarity Average bird community similarity among all sites is 61.4% (SD = 5.97, n = 6) of which sites CS and OP have the highest pairwise comparison with a similarity value of 71.1% (Table 7). Pair plots CS-RE and OV- OP were found out to have more different species occurrences with only about 55% similarity. The degree of species heterogeneity among the study plots can also be evaluated in the species-area curve derived from nested plot data (Fig. 2). The 0.94r2 indiates strong correlation of species richness with increased sampling area. It implies further that some species may have special preferences on certain areas over the others. This could be attributed to habitat type and landscape factors (Jokimäki & Jokimäki, 2003), food availability, or competition from other types of species. Figure 2. Species-area curve for bird nested data over six subplots in UP Diliman. Adjusted Std. Error Model R R Square R Square of the Estimate 1 .934(a) .871 .813 .39781(b) a Predictors: feeding guilds, number of trees, number of buildings, spatial area of green spaces, spatial area of buildings b Dependent Variable: LOG abundance Table 4. Summary statistics of the regression model. Table 3. Indices of bird community diversity and evenness, also shown are landscape data* in the 4 study plots in UP Diliman No. of No. of Building Road Area Green Plot H’ Simpson’s Evenness Buildings Trees Area (m2) (m2) Spaces (m2) CS 1.8536 0.2535 0.5472 24 3818 36,647.90 5,430 254,853.44 OV 1.7951 0.2478 0.6048 3 3737 6,546.83 2,760 144,771.09 OP 1.8704 0.2604 0.5172 13 1225 11,787.35 32,400 269,731.86 RE 1.3196 0.4337 0.4763 93 1662 33,169.73 17,100 114,171.21 *Data courtesy of Geodetic Engineering Department, and Training Center for Applied Geodesy and Photogrammetry, College of Engineering, UP-Diliman 2002 H’ = Shannon-Wiener index of diversity Table 5 ANOVA table of linear regression analysis between log-transformed data on bird abundance with feeding guilds (biotic factor) and landscape variables (abiotic factors). Sum of Mean Model Squares df Square F Sig. 1 Regression 23.602 10 2.360 14.914 .000(a) Residual 3.481 22 .158 Total 27.083 32 a Predictors: feeding guilds, number of trees, number of buildings, spatial area of green spaces, spatial area of buildings b Dependent variable: LOG abundance Unstandardized Standardized Coefficients Coefficients Model B Std. Beta t Sig. Error 1 No. of buildings .005 .002 .500 2.670 .014 No. of trees .000 .000 .666 3.228 .004 Building area -1.58E-005 .000 -.534 -1.326 .198 Greenspace area 1.44E-006 .000 .391 1.295 .209 Insectivores .110 .230 .049 .476 .639 Frugivores -.178 .355 -.047 -.502 .621 Omnivores .381 .204 .198 1.873 .074 Nectarivores .291 .461 .055 .631 .534 Seed eaters .553 .333 .146 1.659 .111 Carnivores .106 .282 .038 .377 .710 a Dependent Variable: LOG abundance Table 6. Correlation coefficients of various landscape variables and foraging habits with bird abundance over spatial scales. Science Diliman 20:1, 1-10 y = 0.4352x + 0.6858 R 2 = 0.9424 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 1.2 1.4 1.6 1.8 2 2.2 Log area (has) Lo g # spe cies Bird diversity and abundance in the UP Diliman 7 DISCUSSION This study detected about 9.2% of birds that can be found in Luzon with 5 species endemic to the Philippines. Nineteen species of the birds recorded by Ong et al. (1999) during October 1997 to January 1998 and June 1998 were detected in this survey. Difference in species accounts may be due to the time of survey (this one was conducted late February to April 2005), the method used, or some previously recorded species may have been locally extinct, escapees or were accidentally recorded. Nevertheless, results of this survey indicate persistence of a relatively diverse avifauna in UP Diliman campus since 1997. It is also a fact that no major habitat conversions were made in the study sites from year 1997 to the first half months of year 2005. The presence of a number of Philippine endemic species D. maculatus, D. poliocephala, Z. meyeni, L. philippensis, and C. viridis makes the campus an important urban bird area. This study indicates that although endemics were present, their population represents only a mere 0.59% of the total bird community. This figure may have been affected by successful colonization by other introduced or resident species as a result of native faunal extinction (Blair 2004) or as a direct result of habitat conversion from its original state. It is unknown whether this small fragment of endemic birds found in the study sites were remnant population, re-introduced or ‘regular foragers’ from adjacent disturbed forest blocks. What is certain is the vulnerability of the endemic population even to slight habitat disturbance or modifications. Our data on bird densities showed two species dominating bird population in all study plots, P. montanus making up 51.4% and P. goiavier 17.7% of the cumulative bird count. P. montanus is a communal seed-eater that was introduced in the Philippines in the 1930s (Ong et al. 1999) while P. goiavier is resident breeding an omnivore, present in all terrestrial habitat type except in mature and secondary forests. Another species which has also been found abundant in all four plots is the pied fantail R. javanica (7.5%). The ability of these species to dominate bird populations in the area categorizes them as ‘urban exploiters’ (Blair 1996; McKinney 2002). Thus, these species may have directly benefited from earliest land conversion and development in UP Diliman and in all other areas in Metro Manila. The community diversity and evenness indices showed how various land-use schemes determine bird species richness. Species evenness was highest at the OV site with a value of 0.6048 indicating higher community diversity wherein large numbers of species are all similarly abundant (Table 3). This may be attributed to the continuous block of wooded areas in the park and minimal number of buildings (n = 3). The low Shannon- Wiener (H’) and evenness value (E 5 ) in the RE site is the function of low species richness and diversity. RE had the most number of recorded individuals with 1,685 but composed only of 14 species. It should be noted that this plot is residential with the most number of infrastructures (number of single-family houses = 93). Accordingly, bird community structure is primarily composed of species that are commensal with humans such as P. montanus (64%). Our data indicates contrary result from the findings of Jokimäki & Jokimäki (2003) in urban centers in Finland wherein bird diversity was observed to be higher in single-family house areas. This could be the case since their data indicated minimal ‘built-up’ spaces in residential areas that are located in sub-urban localities in developed countries (i.e. Finland). However there are lesser areas of green spaces in highly urbanized cities like Metro Manila. This was the case in RE site wherein it has the lowest green space area of approximately 114,000 m2 among other plots. Our findings suggest the influence of landscape- gradient variables in determining local abundances structure of birds in a tropical urban environment. This Plots % Similarity CS-OV 61.9% CS-OP 71.1% CS-RE 55.6% OV-OP 55.8% OV-RE 58.8% OP-RE 64.9% CS, College of Science; OV, Academic oval; OP, Open field site; RE, Residential site Table 7 Percent similarities (Sorensen index) of paired plots based on qualitative data (species occurrences). Science Diliman 20:1, 1-10 Vallejo, et al. 8 Science Diliman 20:1, 1-10 is a result that supports Melles et al (2003). It appears in our analysis that abundance was determined by the number of spatial entities (i.e. number of buildings and trees) and not by spatial area (i.e. area in m2 of built- up and green spaces) or food preference (i.e. feeding guilds). This finding contrasts with that of Lim and Sodhi (2004) who found that increasing urbanization adversely impacted on insectivorous and carnivorous feeding guilds. In our study, insectivorous birds seem to have adapted to the urbanizing landscape. We did not observe carnivores that were probably extirpated in the 1960s as suggested by Ong et al (1999). This finding supports the general idea that typical urban birds primarily respond to architectural features and secondarily to natural features of the environment (Savard & Falls 2001). Bird species following this scheme were the most abundant especially for the highly adaptable and communal species, P. montanus. The more numerous the number of buildings, the more places these species find suitable foraging and nesting grounds. This phenomenon was further observed in our data as the distribution patterns on bird abundance were highly maximal towards ‘urban exploiter’ species like R. javanica, L. cristatus, Zebra Dove Geopelia striata, and P. goiavier. Apparently, landscape homogeneity and increased proportion of built-up spaces (e.g. RE site) lowers bird community diversity and evenness. Without much habitat modifications since the wildlife inventory in 1999, the UP Diliman campus still harbors a viable complex ecosystem for bird communities. Despite pressures by urban sprawl inherent in highly urbanized areas (Blair 2004), a considerable number of endemic species still persist. However, populations of these birds were very marginal compared with other highly urban-adaptable species. Our results indicate that increased number of infrastructure and homogenizing landscape (e.g. residential site) greatly contributes to lower biodiversity patterns and domination of commensal, urban exploiting species. However, various land-use types in the campus appear to contribute to the heterogeneity of bird communities. This was evident the fact that there was moderate similarity in terms of birds species composition among sites. In order to preserve and promote diversity of bird species in the area, complex habitats should be sustained by preserving green spaces such as gardens and parks. Recent research on tropical urban bird communities in Brazil indicate that isolated forest patches within urban areas fail to sustain high bird biodiversity (Manhães & Loures-Ribeiro 2005). A possible solution to mitigate biodiversity loss is to ensure that these patches be preserved as to provide refuges and resource areas for birds (Melles et al 2003). Artificial habitats may also mitigate biodiversity loss. Thus we recommend that in areas reserved for building development, pocket gardens, rooftop gardens and parks must be integral to the landscape design. The planting of fruiting trees (not necessarily of economic value to humans) should also be encouraged. In Singapore, low density housing developments with suitable gardens planted to fruiting trees favored the persistence of frugivores (Lim & Sodhi) Endemics such as the lowland white-eye Z. meyeni were only seen once during this survey in a flock composed of 10-12 individuals in the upper canopies of fruiting mango trees in the OV site. A number of urban bird researchers in developed countries recommend increasing biodiversity by attracting minority species in the community but are capable of utilizing resources available in urban landscapes (Blair 1996; Jokimäki & Suhonen 1998). A good example would be the endemic pygmy woodpecker D. maculatus, an insectivore which is locally uncommon in UP Diliman. The bird’s population has the potential to increase the fact that it appears to mimic more common birds (e.g. Eurasian tree sparrows) and it does not depend on fruiting or flowering trees for food. It however is a cavity nesting species that probably existed in greater abundances before rapid urbanization. The planting of suitable nesting trees may help in its population recovery. We suggest that further research be conducted on the inability of rainforest birds to persist in an urbanizing landscape. This is a finding noted by Lim and Sodhi (2004) that is likely applicable to the Philippine situation. As most of the escapee pet birds we have observed in campus were of rainforest origin, the question of whether they can persist in an urban landscape has important conservation implications. Anecdotal evidence from the WBCP suggests that at least one parrot species has established itself in Metro Manila. Bird diversity and abundance in the UP Diliman 9Science Diliman 20:1, 1-10 ACKNOWLEDGEMENTS This work is funded by University of the Philippines NSRI grant ESM-05-2-01. We would like to thank M. Lu and A. Jensen of the Wild Bird Club of the Philippines and Prof. A. de Villa, CSSP Department of Philosophy for helping in bird identification, the Delaware Museum of Natural History for sending us species and identification catalogues and all volunteers who participated in the surveys. We are also very thankful to Dr. R. M. Gonzales, Chair of the UP Department Geodetic Engineering for providing us remotely sensed data and GIS databases of the UP Diliman campus. We also would like to thank the institutional support of academic and research units of the University of the Philippines; the Institute of Environmental Science and Meteorology, the Institute of Biology, the Marine Science Institute, Department of Geography and the Natural Science Research Institute. REFERENCES Beissinger, S. R. & Osborne, D. R. 1982. Effects of urbanization on avian community organization. Condor 84: 75-83. Blair, R. 1996. Land use and avian species diversity along an urban gradient. Ecological Applications 6: 506-519. Blair, R. 2004. The effects of urban sprawl on birds at multiple levels of biological organization. Ecology and Society 9(5): 2. Available fromhttp://www.ecologyandsociety.org/vol9/ iss5/art2 (Accessed January 2005) Clemants, S. & Moore, G. 2003. Patterns of species richness in eight northeastern United States cities. Urban Habitats 1: 4-16. Clergeau, P., Jokimäki, J. & Savard, J-P.L. 2001. Are urban bird communities influenced by the diversity of adjacent landscapes? Journal of Applied Ecology 38: 1122-1134. Collar, N., Mallari, N. A. D. & Tabaranza, B. R. J. 1999. Threatened Birds of the Philippines. Bookmark, Manila. Gilber, O. 1992. The ecology of urban habitats. Chapman and Hall, London. Gyllin, M. 1999. Integrating biodiversity in urban planning. Urban density and green structure. University of Gotheburg. Hengeveld, R. 1993. Biogeographical ecology. Journal of Biogeography 21:341-351. Hobbs, E. R. 1988. Species richness of urban forest patches and implications for landscape diversity. Landscape Ecology 1: 141-152. Jokimäki, J. & Jokimäki, M-L. K. 2003. Spatial similarity of urban bird communities: a multiscale approach. Journal of Biogeography 30: 1183-1193. Jokimäki, J. & Suhonen, J. 1998. Distribution and habitat selection of wintering birds in urban environments. Landscape and Urban Planning 39: 253-263. Jokimäki, J., Suhonen, J., Inki, K., & Jokinen, S. 1996. Biogeographical comparison of winter bird assemblages in urban environments in Finland. Journal of Biogeography 23: 379-386. Kennedy, R.S., Gonzales, P.C., Dickinson, E.C., Miranda, H.C. & Fisher, T.H. 2000. A Guide to the Birds of the Philippines. Oxford University Press Inc., NY, USA. Kowarik, I. 1995. On the role of alien species in urban flora and vegetation. Pages 85-103 in P.K. Pysek, M. Prach, M. Rejmanek, and P.M. Wade, editors. Plant Invasions: general aspects and special problems. SPB Academic, Amsterdam, The Netherlands. Lim, HC and Sodhi, NS (2004) Responses of avian guilds to urbanization in a tropical city. Landscape and Urban Planning. 66:199-215. Manhães, M.A. and Loures-Ribeiro, A. 2005. Spatial distribution and diversity of bird community in an urban area of Southeast Brazil. Brazilian Archives of Biology and Technology. 48 (2): 285-294 Vallejo, et al. 10 Marzluff, J. M. 2001. Worldwide urbanization and its effects on birds. in J. M. Marzluff, R. Bowman and R. Donelly, editors. Avian ecology and conservation in an urbanizing world. Kluwer Academic Publishers, Boston, MA, USA. Melles, S., Glenn, S., and Martin, K. 2003. Urban bird diversity and landscape complexity: species-environment associations along a multiscale habitat gradient. Conservation Ecology 7(1)5. [online] URL http:// www.consecol.org/vol7/iss1/art5 McKinney, M. L. 2002. Urbanization, biodiversity, and conservation. BioScience 52(10): 883-890. Midorikawa, S. and Bautista, B.C. 2002. Expansion of Damage Potential Area Due to Urban Sprawling. in T. Ohmachi & E. R. Roman, editors. pp 301-307 Metro Manila: In Search of a Sustainable Future. University of the Philippines Press, Quezon City, Philippines. National Mapping and Resource Information (NAMRIA). 2003. NAMRIA, Fort Bonifacio, Makati City, Philippines. Available fromhttp://www.namria.gov.ph/downloads/ lcstat.zip (accessed March 2005) Ohmachi, T. 2002. Ending the Cycle of Environmental Deterioration. in T. Ohmachi & E. R. Roman, editors. pp 3-9 Metro Manila: In Search of a Sustainable Future. University of the Philippines Press, Quezon City, Philippines Ong, P., Pedregosa, M., & de Guia, M. 1999. Wildlife inventory of the UP Diliman and Ateneo de Manila University campuses, Diliman, Quezon City, Luzon, Philippines. Science Diliman 11: 6-20. Peet, R.K. 1974. The measurement of species diversity. Annual Review of Ecology and Systematics 5: 285-307. Rosenzweig, M.L. 1995. Species diversity in space and time. Cambridge University Press, UK. Savard, J-P.L. & Falls, B. 2001. Survey techniques and habitat relationships of breeding birds in residential areas of Toronto, Canada. in J.M. Marzluff, R. Bowman and R. Donelly, editors. Avian ecology and conservation in an urbanizing world. Kluwer Academic Publishers, Boston, MA, USA. Szacki, J. 1999. Ecological corridors and greenspace in the city- A Polish perspective. Urban density and green structure. University of Gotheburg, Sweden. Statistical Package for the Social Sciences, Version 13.0 (SPSS, V. 13.0) 2005. SPSS Inc. Headquarters, 233 S. Wacker Drive, 11th Floor, Chicago, Illinois 60606 Waite, S. 1999. Statistical Ecology in Practice: A guide to analysing environmental and ecological field data. Prentice Hall, UK. Science Diliman 20:1, 1-10 Species CS OV OP RE Acridotheres cristatellus 0 0 4 0 Alcedo atthis 0 1 0 0 Anthus novaeseelandiae 2 0 1 0 Cacomantis variolosus 4 0 0 0 Centropus viridis 2 2 0 0 Chalcochaps indica 0 0 1 0 Columba livia 3 15 15 5 Dendrocopus maculatus 0 5 1 5 Dicaeum sp. 0 1 0 0 Ducula poliocephala 0 2 0 0 Galirallus sp. 0 0 0 1 Gallus gallus 16 0 24 20 Geopelia striata 8 4 79 13 Gerygone sulphurea 16 49 36 95 Halcyon chloris 5 12 6 0 Hirundo tahitica 24 11 22 38 Ixobrychus cinnamomeus 0 0 2 0 Lanius cristatus 60 64 83 53 Lanius schach 27 10 19 0 Loriculus philippensis 2 0 0 0 Megalurus palustris 11 3 5 0 Megalurus timoriensis 3 2 0 0 Motacilla cinerea 1 0 2 0 Muscicapa griseistica 0 3 0 0 Nectarinia jugularis 0 4 0 6 Passer montanus 279 299 432 1073 Phylloscopus borealis 0 0 1 10 Pycnonotus goiavier 131 249 135 203 Rhipidura javanica 39 62 47 157 Sterna/Chlidonia sp. 0 0 1 0 Streptopelia chinensis 2 0 9 6 Turdus sp. 0 0 1 0 Zosterops meyeni 0 12 0 0 CS, College of Science; OV, Academic oval; OP, Open field site; RE, Residential site Appendix 1 Abundance matrix of regularly occurring birds in 4 survey plots in UP Diliman (CS, College of Science complex; OV, Academic oval-park; OP, Open field site; RE, University residential area)