04_Kumar_01_22.indd UDC 598.26:504.06(540.23) LANDUSE PATTERNS, AIR QUALITY AND BIRD DIVERSITY IN URBAN LANDSCAPES OF DELHI V. Kumar1*, V. Jolli1, C. R. Babu2 1Shivaji College (University of Delhi), Raja Garden, New Delhi- 110 027, India 2Centre for Environmental Management of Degraded Ecosystems, University of Delhi, Delhi-110 007, India *Corresponding author E-mail: vijaycemde@yahoo.co.in Landuse Patterns, Air Quality and Bird Diversity in Urban Landscapes of Delhi. Kumar, V., Jolli, V., Babu, C. R. — In the present paper we attempted to explain the relationships among the landuse pattern, levels of air pollutants and bird diversity based on data from 5 sampling sites in Delhi. Five landuse categories — percent built up area, tree cover, park area and barren area were recognized in the study area. Th e objective of this study is to fi nd out the eff ects of landuse changes on air pollution and bird diversity and whether birds can serve as indicator of landuse changes and air pollutants. Th e levels of six air pollutants (PM10, PM2.5, NOX, SO2, Ozone and Benzene) from the monitoring stations were used. Th e bird diversity was assessed using conventional measures. All the sites showed remarkable diff erences with respect to each of the fi ve landuse categories, air pollution levels, and bird diversity. Th e results suggest that landuse changes infl uence air pollution and bird diversity and some bird species can be used as indicator of landuse change and air pollution. K e y w o r d s : Delhi, Urban ecosystem, Landuse pattern, Air Pollution, Bird diversity. Introduction Urbanization is a global phenomenon and brings out major alterations in landuse, particularly the area under green spaces and their structure. Spillover of zoonotic diseases to human population leading to pan- demics is also attributed to large scale changes in landuse and habitat degradation (UNEP, 2016). Delhi — the second most populous city of the world and National Capital Territory of India — is one of the most polluted cities in the world (Marlier et al., 2016; Landrigan, 2017). Th e city has been witnessing rapid changes in green infrastructure and high levels of air pollution (poor to severe as per the Central Pollution Control Board, Gov- ernment of India). Although Delhi is considered as one of the greenest capitals in the world, but the largest greenspace of Delhi, the ridge spreading over an area of 7,700 hectares is largely a monoculture of Prosopis julifl ora — an invasive alien species introduced by then British Government of India. Data on the extent of green spaces in diff erent localities are known but their structural diff erences are least known. Kumar et al. (2019) demonstrated that the air quality diff ers at diff erent locations and is infl uenced by the structure of the green cover. Further, the kind of tree species, abundance of tree species, and leaf size have signifi cant impacts on air quality. Zoodiversity, 56(1): 39–50, 2022 DOI 10.15407/zoo2022.01.039 40 V. Kumar, V. Jolli, C. R. Babu Bird diversity has been investigated in urban landscapes (Urfi , 2010; Filloy et al., 2019) and birds have been used as indicator species for air quality (Eeva et al., 2000; Birdlife International, 2004) and also for landuse changes (Lawton et al., 1998). However, studies on how the landuse patterns infl uence the air quality and bird diversity are limited. In the present paper we explain how the changes in land cover impact the air quality and the bird diversity in the worst polluted city of Delhi. Methods S t u d y a r e a Delhi is located between 28°24´ and 28°53´ N latitude and 76°50´ and 77°0´E longitude, with an aver- age altitude of 216 m above m. s. l. Th e urban landscape of Delhi includes the Ridge Forest spreading over 7,700 hectare and is predominantly composed of an invasive alien species — Prosopis julifl ora, avenue planta- tion, green belts and patches of mixed wood lots, besides 52 km stretch of the Yamuna River. Th e avenue planta- tions show diff erences in species compositions among diff erent locations. Th e avifauna of Delhi, at one time, was very rich and as many as 400 species have been reported from ter- restrial and aquatic habitats. With the loss and/or degradation of native vegetation and structural changes in the green spaces, there has been marked reduction not only in bird diversity but also in their abundance (Khera et al., 2009). Th e common birds found in urban landscapes of Delhi include the House Crow (Corvus splen- dens), Common Myna (Acridotheres tristis), Red-vented Bulbul (Pycnonotus cafer), Rock Pigeon (Columba livia), Eurasian Collared Dove (Streptopelia decaocto), Asian Koel (Eudynamys scolopacea), Coppersmith Bar- bet (Megalaima haemacephala), Indian Grey Hornbill (Ocyceros birostris), Jungle Babbler (Turdoides striatus), Red Wattled Lapwing (Vanellus indicus), Black-winged Stilt (Himantopus Himantopus), Cattle Egret (Bubulcus ibis) etc. Th e wetlands of Delhi also receive sizeable populations of migratory birds from Siberia, Central Asia, Europe and Ladakh in the months of November–March. S e l e c t i o n o f s a m p l i n g s i t e s a n d v e g e t a t i o n Five sampling sites were selected based on the location of Delhi Pollution Control Committee (DPCC) air pollution monitoring stations. Th e fi ve sites are: (i) R.K Puram (RK) site dominated with Alstonia scholaris and Polyalthia longifolia as avenue plantations; (ii) Mandir Marg (MM) site dominated with Ficus religiosa and Polyalthia longifolia as avenue trees; (iii) Punjabi bagh (PB) site dominated with Polyalthia longifolia and Syzy- gium cumini as avenue trees; (iv) Civil lines (CL) site dominated with Acacia leucocephala and Prosopis julifl ora forest; and (v) Anand Vihar (AV) site dominated with Albizia lebbeck and Azadirachta indica as avenue. Th e map of Delhi showing sampling sites is given in fi g. 1. Fig. 1. Map of Delhi showing location of sampling sites. 41Landuse Patterns, Air Quality and Bird Diversity in Urban Landscapes of Delhi A i r q u a l i t y d a t a Secondary data from DPCC monitoring stations were collected and the levels of 6 air pollutants (PM10, PM2.5, NOX, SO2, Ozone and Benzene) at the fi ve sites were analysed as per the procedures outlined earlier (Ku- mar et al., 2019) and correlated with bird diversity, richness and abundance, and with landuse categories. Mean value (based on 11months during 2015–2016) were used for assessing the pattern of variation in air pollutants. L a n d u s e m a p s o f s t u d y s i t e s Landuse maps were generated using Google Earth ver 10 for quantifying the area under fi ve categories- tree cover, park area, built up area and barren area and expressed as percent of total area sampled in each locality. For each site a square plot of (1 × 1) km was demarcated by keeping transect in the middle, using most recent satellite images. Th e visual image interpretation method was used for characterising land features based on true colour (red, green and blue colour wavelengths) high resolution Google Satellite images (NASA, 2013). Land features in polygon were saved as KML fi le format and were later converted to shape fi les using online Mygeodata-GIS Converter (http://converter.mygeodata.eu/). Th e fi les were geometrically corrected by UTM WGS84. Th ese fi les were later added, overlayed and processed using ArcGIS 9.3 soft ware and landuse maps of study sites were generated. B i r d s u r v e y s Surveys were made for monitoring the birds on selected sites during winter and monsoon seasons using line transect method (Bibby et al., 2000). Sample size (n) of RK, MM and PB was nine, whereas for CL and AV it was ten. Birds sighted were identifi ed using a fi eld guide (Kazmierczak & Perlo, 2015). A square plot of (1 × 1) km was demarcated in each site (DPCC monitoring station in its middle position). On each plot a line tran- sect of 1 km length was laid and all bird individuals heard or seen were counted on both sides of the transect in 30 minutes duration. To adjust detection eff ect, birds were counted up to 50 m of both sides of the transect line using binocular (Nikon Action Ex 10 x 50 mm). Bird counting was carried out only on a clear day from 09:00–11:00 A.M. during 2015–2016. Th e species richness was expressed as number of species recorded at each site; the diversity was estimating using Shannon–Weiner Index (Hammer et al., 2001). Th e percent relative abundance of each bird species recorded was estimated. Th e birds recorded were grouped under foraging guilds based on their food habits known from the published data (Ali & Ripley, 1983). D a t a A n a l y s e s To fi nd out the relationships of bird abundance, richness and diversity with landuse categories and air pollutants levels, RMA (Reduced Major Axis) linear regression analyses were carried out and Pearson’s cor- relations (r) were computed. Further, Canonical Correspondence Analysis (CCA) was also performed. PAST Version 4.09 statistical soft ware was used to analyse the data (Hammer et al., 2001). Results L a n d u s e p a t t e r n s a n d a i r p o l l u t i o n Th e fi ve diff erent sites selected in the study area showed marked diff erence in all the four parameters — area under tree cover, park area, built up area and barren area (table 1). For example, PB site showed maximum (61.17 %) built up area as compared to all other four sites where the built up area was 22.33 (CL) to 44.39 % (RK); on the other hand, the tree cover was maximum (59.32%) at MM and values at other sites varied from to 18.31 (AV) to 58.93 % (CL). Th e area under park was maximum (16.82 %) at CL but in other sites the values ranged from 6.15 (AV) to 15.66 % (RK). No barren land was observed at PB and CL, but more than one fourth of the area i. e. 27.99 % area was barren at AV. RK and MM sites also showed markedly less barren area than other sites (table 1). The five sampling sites showed characteristic landuse patterns specific to each site. For example, the MM site showed highest tree cover, lesser built up area, mod- erate area under park and very small barren area; on the other hand, the maximum barren area was observed at AV site, where the values for all other categories were also low. The range of variation in the park area among the sampling sites was rather narrow (i. e. 10.15–16.82 %). Th e fi ve diff erent sampling sites also showed marked diff erences in the levels of all the six air pollutants analysed. Although the variation in the levels of each of the six pollutants among diff erent sites has already been published (Kumar et al., 2019), for the purpose of assessing the impact of landuse patterns on the air quality, air pollution data was also pre- sented in the table 1. 42 V. Kumar, V. Jolli, C. R. Babu Th e PB sampling site, which showed the maximum built up area, low tree cover and moderate area under park, also showed higher concentration of PM10, PM2.5, NOX and Ozone; AV site, which had the maximum barren area, the least tree cover, relatively low built up area and also least park area, showed highest levels of all air pollutants except Ozone. Moderate levels of air pollutants were observed at CL site which was characterised by maximum park area, relatively high tree cover but relatively low built up area. MM site with maximum tree cover, moderate park area and relatively low built up area showed low- est levels of air pollutants (table 1). T a b l e 1 . Landuse categories and levels of air pollutants at diff erent sampling sites RK Puram, RK Mandir Marg, MM Punjabi Bagh, PB Civil Lines, CL Anand Vihar, AV Landuse Categories, % (1 x 1) km Tree cover 35.71 59.32 24.13 58.93 18.31 Park area 15.66 14.28 10.15 16.82 6.15 Built up area 44.39 23.36 61.17 22.23 36.88 Barren area 0.85 0.28 0 0 27.59 Air Pollutants, μg/m3 PM10 278.85 ± 37.87 237.01 ± 31.67 286.44 ± 38.95 319.9 ± 18.89 418.02 ± 61.55 PM2.5 143.89 ± 26.05 113.63 ± 24.19 133.8 ± 27.28 191.78 ± 14.03 169.65 ± 30.13 SO2 26.22 ± 4.59 18.39 ± 2.55 21.6 ± 2.78 21.44 ± 1.26 21.05 ± 3.04 NOX 74.42 ± 6.32 57.96 ± 5.57 80.02 ± 6.08 78.28 ± 7.04 78.74 ± 7.29 Benzene 6.33 ± 1.37 3.27 ± 50.52 0.87 ± 0.14 10.55 ± 2.92 13.78 ± 2.33 Ozone 56.41 ± 7.03 34.19 ± 4 58.33 ± 7.51 78.05 ± 11.09 30.08 ± 3.07 T a b l e 2 . Mean relative abundance (%) of bird species recorded at diff erent sampling sites (2015–2016) English Name Scientifi c Name R K Puram, RK (n = 9) Mandir Marg, MM (n = 9) Punjabi Bagh, PB (n = 9) Civil Lines, CL (n = 10) Anand Vihar, AV (n = 10) Cattle Egret Bubulcus ibis 0 0 0 0 8.78 Red Wattled Lapwing Vanellus indicus 0 0 0 0 1.25 Black-winged Stilt Himantopus himantopus 0 0 0 0 4.08 Black Kite Milvus migrans 6.19 4.14 12.34 4.5 15.67 Shikra Accipiter badius 0 0 0.11 0 0 Egyptian Vulture Neophron percnopterus 0 0 0 0 0.31 Eurasian Collared Dove Streptopelia decaocto 0.61 7.56 2.08 2.14 1.25 Laughing Dove Streptopelia senegalensis 0.31 0 0 0 0 Rock Pigeon Columba livia 21.37 28.23 55.24 34.26 36.99 Asian Koel Eudynamys scolopaceus 0 0 0.11 0 0 Rose-ringed Parakeet Psittacula krameri 1.56 5.39 0.92 7.28 0 Asian Palm Swift Cypsiurus balasiensis 0.61 0 0 0 0.31 Brown-headed Barbet Megalaima zeylanica 0.31 0.53 0 0 0 Coppersmith Barbet Megalaima haemacephala 0 0.18 0 0 0 Indian Grey Hornbill Ocyceros birostris 0 0.53 0 0 0 Rufous Treepie Dendrocitta vagabunda 0.31 0 0.11 0 0 Common Myna Acridotheres tristis 16.1 16.01 11.42 21.84 9.72 House Crow Corvus splendens 23.52 20.87 13.04 19.49 21 Large-billed Crow Corvus macrorhynchos 0 0.18 0 0 0 Red-vented Bulbul Pycnonotus cafer 13.32 7.56 3.11 3.21 0 Red-whiskered Bulbul Pycnonotus jocosus 0.31 0 0 1.71 0 Jungle Babbler Turdoides striata 15.49 3.42 0.34 3.64 0 Common Tailorbird Orthotomus sutorius 0 0 0.58 0 0.63 Purple Sunbird Nectariniaasiaticus 0 1.08 0.58 1.07 0 Oriental White-eye  Zosterops palpebrosus 0 3.77 0 0 0 House Sparrow Passer domesticus 0 0.53 0 0.86 0 43Landuse Patterns, Air Quality and Bird Diversity in Urban Landscapes of Delhi B i r d s u r v e y s A total of 26 bird species were recorded from all the sampling sites (table 2). Species rich- ness and the diversity of birds varied signifi cantly among diff erent sampling sites (fi g. 2). Th e MM site, which had the highest tree cover, showed the maximum number of species (8 ± 0.53) and highest diversity (1.82 ± 0.05); on the other hand, AV site which showed least tree cover, highest barren area and low park area, had lowest number of bird species (5.8 ± 0.36) and relatively low bird species diversity (1.34 ± 0.06); sites like PB with highest built up area, low tree cover and no barren area and relatively low park area had least species diversity of birds (1.26 ± 0.07) but species richness was moderate and inhabited by Rock Pigeon, House Crow, Common Myna and Black Kite; the RK site, which had moderate tree cover and park area, rela- tively high built up area and low barren area, also showed relatively higher bird species diversity (1.65 ± 0.05) and moder- ate species richness (fi g. 2; table 1). Th e number of species (7.3 ± 0.47) and diversity of birds (1.66 ± 0.06) was relatively low at CL site as compared to MM which had highest species diversity (1.82 ± 0.05) and species rich- ness (8  ± 0.53), although both sites had more area under tree cover. Th e percent relative abundance of diff erent bird species not only var- ied within the site but also among sites (table 2). For example, percent relative abundance of Rock Pi- geon varied from 21.37 % to 55.24 % across the sam- pling sites with maximum value at PB and lowest value at RK; on the other hand, the variation in per- cent relative abundance of House Crow was 13.04 % to 23.52 % across the sam- pling sites. Some birds such as Large-billed Crow, House Sparrow and Cop- persmith Barbet were ex- tremely rare at MM site; RK M ea n bi rd a bu nd an ce /K m /3 0 m in s 0 20 40 60 80 100 120 MM PB CL AV RK M ea n bi rd sp ec ie s r ic hn es s/ K m /3 0 m in s M ea n bi rd sp ec ie s r di ve rs ity /K m /3 0 m in s 0 2 1 3 4 6 5 8 7 9 MM PB CL AV c b a RK 0 0,6 0,4 0,2 0,8 1 1,4 1,2 1,6 1,8 2 MM PB CL AV Fig. 2. Variation in mean bird abundance (a); mean bird species richness (b); and mean bird species diversity (c) among diff erent sampling sites. N o t e .Y bar indicates standard error. 44 V. Kumar, V. Jolli, C. R. Babu birds such as Indian Grey Hornbill, Oriental White- eye, Coppersmith Barbet were absent at all other sampling sites except MM site. Th e most dominant birds in the study area were Rock Pigeon, House Crow and Common Myna and these were in higher abundances at the sampling sites (table 2). Mean abundance of birds was maximum at PB (96.33 ± 8.13) and mini- mum at AV (32.4 ± 3.29) (fi g. 2, a). F o r a g i n g g u i l d s To understand the distribution patterns of foraging guilds among the sampling sites, the birds recorded in the study area were grouped under six diff erent foraging guilds, based upon their food preferences. Th e highest percentage of birds (45.14  %) belonged to om- nivorous guild, whereas 38.05  % of birds were granivorous; about 10.66  % of birds were carnivorous; and 3.7 % of birds were frugivorous. Th e insectivorous and nectivorous birds constituted 1.78 % and 0.7 %, respectively (fi g. 3). Th e MM Site showed relatively high percentage of frugivorous, insectivorous and nec- tivorous as compared to other sites; whereas, AV site showed high percentage granivorous and omnivorous birds but frugivorous and nectivorous birds were absent (fi g. 4). C o r r e l a t i v e a n a l y s e s R e l a t i o n s h i p b e t w e e n l a n d u s e c a t e g o r i e s a n d a i r p o l l u t i o n Th e relationships between landuse categories and air pollution levels are given in the table 3. Th e levels of PM10 and NOx showed negative and statistically signifi cant (p < 0.05) relationships with percentage of tree cover, but their relationships with percent built up area and barren area were positive and statistically signifi cant (p < 0.05). Th e ‘r’ values for Om niv oro us Gr an ivo rou s Ca rn ivo rou s Fr ug ivo rou s In sec tiv oro us Ne cti vo rou s A bu nd an ce o f F or ag in g gu ild (% ) 0 5 15 10 20 25 30 35 40 45 50 45,14 38,05 10,66 1,78 3,7 0,7 Omnivorous Frugivorous Carnivorous Insectivorous Granivorous Nectivorous 0 10 30 20 40 50 60 70 80 90 100 RK MM PB CL AV % Fig. 3. Variation in abundance of foraging guilds (%) of birds of all the sam- pling sites together. Fig. 4. Variation in abundance of foraging guilds (%) of birds among diff erent sampling sites. 45Landuse Patterns, Air Quality and Bird Diversity in Urban Landscapes of Delhi all the combinations involving the percent tree cover with air pollutants showed statisti- cally insignifi cant correlations. Similar patterns of relationships were observed for combi- nations involving PM10, NOx with percent area under park (table 3). Th e percent of built up area did not show statistically signifi cant relationships with all the air pollutants except for SO2 for which the relationship was positive and statistically sig- nifi cant (p < 0.05). Similarly, percent barren area showed statistically signifi cant (p < 0.05) positive relationship with PM10 and Benzene, but its relationship with Ozone was negative and statistically signifi cant (p < 0.05). Th e ‘r’ values for all other combinations involving barren area and levels of air pollutants were low and statistically non-signifi cant (table 3). Th e regression analysis of the data also showed similar relationship patterns among the variable tested. T a b l e 3 . ‘ r’ and ‘r2’ values between logarithms of percentage of landuse categories (measured as in percent area) and levels of air pollutants (PM10, PM2.5, NOX, SO2, Benzene and Ozone measured as μg/m 3) Landuse Categories, % PM10 PM2.5 SO2 NOx Ozone Benzene r r2 r r2 R r2 r r2 r r2 r r2 Tree cover –0.66* 0.43 –0.13 0.016 –0.25 0.06 –0.59* 0.35 0.37 0.13 0.07 0.005 Park area –0.72* 0.52 –0.12 0.014 0.18 0.03 –0.35* 0.12 0.65 0.42 –0.08 0.006 Built up area 0.011 0.012 –0.19 0.03 0.51* 0.26 0.52* 0.27 0.06 0.003 –0.54 0.29 Barren area 0.89* 0.8 0.34 0.12 –0.11 0.01 0.29 0.08 -0.62* 0.38 0.73* 0.53 *Highly signifi cant at p < 0.05. T a b l e 4 . ‘r’ and ‘ r2 ‘values between logarithms of mean bird species diversity and levels of air pollutants (PM10, PM2.5, NOX, SO2, Benzene and Ozone as μg/m 3) and landuse categories Bird species diversity versus air pollutants r r2 Particulate Matter*, PM10 –0.59 0.34 Particulate Matter, PM2.5 –0.20 0.04 Oxides of Nitrogen*, NOX –0.72 0.52 Sulphur Dioxide, SO2 –0.14 0.02 Benzene 0.29 0.09 Ozone, O3 0.08 0.005 Bird species diversity versus landuse categories r r2 Tree cover* 0.90 0.81 Parkland* 0.77 0.60 Builtcover* –0.76 0.58 *Highly signifi cant at p < 0.05 T a b l e 5 . ‘r’ and ‘ r2’ values between logarithms of mean bird species richness and levels of air pollutants (PM10, PM2.5, NOX, SO2, Benzene and Ozone as μg/m 3) and diff erent landuse categories Bird species richness versus air pollutants r r2 Particulate Matter, PM10) 0.32 0.32 Particulate Matter, PM2.5 –0.20 0.10 Oxides of Nitrogen, NOX –0.29 0.08 Sulphur Dioxide, SO2 –0.42 0.17 Benzene 0.11 0.01 Ozone*, O3 –0.98 0.96 Bird species richness versus landuse categories r r2 Tree cover* 0.91 0.83 Parkland* 0.72 0.52 Builtcover* –0.66 0.43 *Highly signifi cant at p < 0.05. 46 V. Kumar, V. Jolli, C. R. Babu R e l a t i o n s h i p b e t w e e n b i r d d i v e r s i t y a n d l a n d u s e c a t e g o r i e s Th e relationships of landuse categories with bird diversity were analysed (table 4). Th e percent tree cover (r = 0.89, p = 0.008) and park area (r = 0.77, p = 0.02) showed statistically signifi cant positive relationships with bird diversity; and the bird diversity showed negative and signifi cant relationship (r = –0.76, p = 0.002) with built up area. R e l a t i o n s h i p b e t w e e n b i r d d i v e r s i t y a n d a i r p o l l u t a n t s Th e relationship of air pollutants with bird diversity were also analysed (table 4). Bird diversity showed statistically signifi cant negative relationship with NOx (r  = –0.72, p = 0.02) and PM10 (r =  –0.58, p = 0.01), but its relationship with other air pollutants were negative and weak. Th e relationship of percent relative abundance of individual bird species with PM10, NOX and Ozone pollutants were analyzed. Th e percent relative abundances of the House T a b l e 6 . ‘r’ and ‘r2’ values between logarithms of mean bird species abundance and levels of PM10, NOX and Ozone Name of Bird Species rPM10 r 2 PM10 rNOX r 2 NOX rOzone r 2 Ozone Common Myna –0.73074* 0.53398 –0.26519 0.070326 0.57945 0.33577 Rock Pigeon -0.20878 0.043589 0.14066 0.019785 0.202633 0.041057 House Crow –0.83474* 0.69679 –0.51645* 0.26672 0.15851 0.025124 Red-vented Bulbul –0.95875* 0.9192 –0.52102 0.27146 0.31653 0.10019 Common Tailorbird 0.21184 0.044877 0.48887 0.239 –0.00956 0.0001 Eurasian Collared Dove –0.671* 0.45024 –0.72418 0.52443 –0.19355 0.03746 Black Kite 0.86016 0.73987 0.42329 0.17917 –0.09105 0.00829 Shikra –0.62355 0.38882 0.35972 0.1294 0.26387* 0.06963 Cattle Egret –0.64827* 0.42025 0.2928 0.085735 –0.66648* 0.44419 Red Wattled Lapwing –0.62938 0.39612 0.2928 0.085735 –0.66648* 0.44419 Black-winged Stilt 0.86016 0.73987 0.2928 0.085735 0.66648* 0.44419 Jungle Babbler 0.19499 0.038021 –0.38062 0.14487 0.36053 0.12998 Oriental White-eye –0.14477 0.020958 –0.97949* 0.95941 –0.48664* 0.23682 Rose-ringed Parakeet 0.86016 0.73987 –0.53697 0.28833 0.37566 0.14112 Brown-headed Barbet –0.74589* 0.55635 –0.98245* 0.96521 –0.41775* 0.17452 Red-whiskered Bulbul 0.11395 0.012985 0.28712 0.082438 0.73237 0.53637 Rufous Treepie –0.29467 0.086833 0.34155 0.11666 0.39231* 0.1539 Purple Sunbird –0.5975* 0.357 –0.41478 0.17204 0.24957 0.062284 Grey Hornbill –0.64827* 0.42025 –0.97949* 0.95941 –0.48664* 0.23682 Coppersmith Barbet –0.64827* 0.42025 –0.97949* 0.95941 –0.48664* 0.23682 Jungle Crow –0.64827* 0.42025 –0.97949* 0.95941 0.36053* 0.12998 House Sparrow –0.35194 0.12386 –0.49376* 0.2438 0.22868 0.052293 Asian Koel –0.14477 0.020958 0.35972 0.1294 0.26387 0.06963 Egyptian Vulture 0.86016* 0.73987 0.2928 0.085735 –0.6648* 0.44419 Laughing Dove –0.21613 0.046714 0.058592 0.003433 0.2166 0.046916 Asian Palm Swift 0.19712 0.038858 0.19993 0.039972 0.10336 0.010683 *Highly signifi cant at p < 0.05. 47Landuse Patterns, Air Quality and Bird Diversity in Urban Landscapes of Delhi Crow, Red-vented Bulbul, Common Myna, Brown-head barbet, Coppersmith Barbet, Cattle Egret, Jungle Crow, Black-winged Stilt and Purple Sunbird showed negative and statistically signifi cant relationship with PM10 but r-value was low. House Sparrow showed negative relationship with PM10; however, Egyptian Vulture showed positive relationship with PM10 (table 6). Similarly, the percent relative abundance of the Oriental White-eye, Grey Hornbill, Brown-headed Barbet, Coppersmith Barbet, Jungle Crow, House Crow and House Spar- row showed statistically signifi cant (p < 0.05) negative relationships with NOX; the percent relative abundance of Asian Koel and Shikra showed negative relationship with NOX but ‘r’ values were low (table 6). Th e relative abundances of bird species like Indian Grey Hornbill, Coppersmith Barbet, Egyptian Vulture, Black-winged Stilt, Red Wattled Lapwing showed negative relationship with Ozone, whereas Rufous Treepie and Jungle Crow showed low positive r values (table 6). Similar patterns of relationships were observed when regression analysis was carried out. R e l a t i o n s h i p s o f l a n d u s e p a t t e r n s , a i r p o l l u t a n t s a n d b i r d s a b u n d a n c e CCA was performed to assess the relationships of landuse patterns, air pollutants and bird species abundance (fi g. 5). CCA plot with axis 1 and 2 were plotted as they ex- plained 75 % variance in the data. Among the landuse categories, tree cover was found to be the most dominant variable aff ecting the bird species abundance. Bird species such as the Brown-headed Barbet, Rose-ringed Parakeet, Purple Sunbird and House Sparrow were associated with increase in percentage of tree cover (fi g. 5). Whereas, among the air pol- lutants PM10 was aff ecting the bird species abundance. Few bird species such as Black Kite, Rock Pigeon were found in sites having high concentration of PM10 (fi g. 5). Discussion Human dominated landscapes are rapidly undergoing changes in landuse pattern and quality of environment, both of which have been impacting other forms of life (Grimm et al., 2008). Delhi is one of the worst polluted cities in the world and rapid urbanization has been resulting in marked changes in landuse and enhanced levels of air pollutants (Marlier et al., 2016). A number of studies have been carried out on the impacts of landuse changes on the avian biodiversity (Newbold et al., 2016; Herrando et al., 2016). But studies on the impact Fig. 5. CCA plot showing the relationships of landuse patterns, air pollutants and bird species abundance. 48 V. Kumar, V. Jolli, C. R. Babu of landuse on air quality and bird diversity are limited. It has been demonstrated that birds serve as indicators of habitat change (Lawton et al. 1998; Gregory & Strien, 2010), air qual- ity (Eeva et al., 2000; Eeva et al., 2003) and also water quality (Ormerod & Tyler, 1993; Sor- ace et al., 1999). We attempted to discuss our results on the impact of changes in landuse patterns on the air quality and bird diversity, particularly to fi nd out whether bird diversity can be used as an indicator of air pollution and landuse changes. Our earlier investigations indicated that kinds of tree species, relative abundance of tree species and leaf size infl uence the effi cacy of the avenue plantations as fi lter in mitigat- ing air pollution (Kumar et al., 2019). We have taken the same secondary data on the levels of air pollutants (PM10, PM2.5, NOx SO2, Ozone and Benzene) from the same sampling sites used in our earlier investigation (Kumar et al., 2019); in the present investigation we also examined here the impact of landuse changes on air quality and bird diversity. Th e AV site is remarkably diff erent from all other sites in having the lowest tree cover (18.31 %), moderately high built up area (36.88 %), low park area (6.15 %) and very high barren area (27.59 %), in contrast to highest tree cover (59.32 %) high park area (10.15 %), low built up area (23.36 %) with very low barren areas (0.28 %) of MM site. Th is indicates site specifi c landuse pattern in the study area resulting from urbanization. Th e four diff erent categories of landuse (tree cover, park area, built up area and barren area) showed marked diff erences among the fi ve sampling sites (table 1). Th e site-specifi c landuse patterns did show site specifi c air quality profi les, indicating that landuse patterns infl uence the air quality and bird diversity, relative abundance of birds and diversity in for- aging guilds. For example, AV site has highest level of air pollutants, least bird species rich- ness, low bird diversity and very low relative abundance of birds with high percent of gra- nivorous, omnivorous and carnivorous birds and absence of nectivorous and frugivorous birds, all of which are negatively correlated to built up area and barren area but positively correlated with percent tree cover (fi g. 2 and 3; table 4 and 5). On the other hand, MM site (dominated with broad-leaved tree species such as Ficus religiosa) showed highest species diversity and species richness of birds with moderately high bird abundance and presence of all the foraging guilds at higher percentage, all of which are positively correlated with tree cover and negatively correlated with air pollutant levels (fi g. 2 and 3; table 4 and 5). In other words, birds not only serve as indicators of habitat change but also air quality. Th ese results substantiate that landuse pattern not only infl uence air quality but also impact on bird richness, diversity and their relative abundance and foraging guilds. Similar observa- tion was made by other workers (Aronson et al., 2014). CL site, had large area under tree cover which is predominantly composed of micro- phyllus tree species like Prosopis julifl ora, but the site had higher values of air pollutants and relatively low bird richness, diversity and relative abundance (fi g. 2) suggesting that structure of green spaces also infl uence air quality and bird diversity. In fact, Kumar et al. (2019) showed that P. julifl ora and other microphyllous trees are ineff ective as fi lters for air pollutants; Khera et al. (2009) mentioned that abundance of tree cover of P. julifl ora was negatively correlated with bird diversity. Th e PB site had a maximum built up area and high bird abundance but the lowest bird species diversity (fi g. 2; table 4) suggesting that built up area has adverse impacts on bird diversity and air pollution. Th is is also evident from the negative relationship of built up area with bird diversity whereas positive relationship with air pollutants (table 4). Further, our results also showed avian homogenization with increase in percentage of built up area in Delhi. For example, two bird species (Rock Pigeon and House Crow) constitute 58 % of the total relative abundance of birds at PB site. Similar observations were made by McKin- ney (2006); Devictor et al. (2007); Aronson et al. (2014) on the impact of urbanization on bird communities. It may be noted that some of the birds such as Oriental White-eye, Brown-headed Bar- bet and Asian Koel, Laughing Dove, Coppersmith Barbet, Indian Grey Hornbill were ab- 49Landuse Patterns, Air Quality and Bird Diversity in Urban Landscapes of Delhi sent at AV and PB sites, both of which have low tree cover, and high levels of air pollutants. Th ese sites have less number of foraging guilds (4) and relatively higher area under parks as compared to tree cover. Th ese observations suggest that landuse change has marked eff ect on the air quality as well as on the bird diversity, and tree cover is more eff ective in miti- gating air pollution and also support habitat specifi c bird species. However, the park area has more generalist birds and less eff ective in mitigating air pollution. Th is is evident from the fact that the sites such as RK and PB sites, had higher park area, lesser tree cover and relatively low barren area, showed higher abundance of generalist birds such as Common Myna, Rock Pigeon and House Crow, and relatively moderate levels of air pollutants except SO2 and Ozone. Of the diff erent species analysed for the foraging behaviour, we found the nectrivorous birds are more sensitive to air pollutants, as these are either rare or absent in highly polluted sites. For example, nectrivorous birds were recorded only at MM and CL sites which showed relatively low levels of air pollutants and higher area under tree cover (fi g. 3; table 1) suggesting that nectrivorous birds likely to be served as sentinal species for monitoring air quality. In fact, other workers also observed reduction in nectivorous birds due to urbanization (Pauw & Louw, 2012). It is interesting to note that the relationships between species richness and relative abundance of birds with Ozone was negative and statistically highly signifi cant (table 5 & 6). For example, Egyptian Vulture, Coppersmith Barbet, Indian Grey Hornbill, Brown- headed Barbet, Oriental White-eye abundance showed decline in abundance with increase in concentration of Ozone (table 6). Th is suggests that Ozone might be impacting bird species richness. In fact, bird species like Coppersmith barbet, Indian Grey Hornbill, Asian Palm Swift and Oriental White-eye were not observed in the PB site which has highest Ozone level (table 2). CCA analysis revealed decrease in bird species abundance with increase in concentra- tion of PM10 whereas bird species abundance increased with increase in percentage of tree cover (fi g. 5). Th us PM10 and tree cover likely to aff ect the bird species abundance in the current study. Th ese observations further confi rmed the relationships derived from Pear- son’s correlation analysis. Conclusions Our study suggests that landuse changes, particularly area under tree cover and struc- ture of greenspaces do infl uence the air quality and the bird diversity. For example, bird diversity is higher in areas where tree cover is high and low levels of air pollutants and vice versa. Some of the bird species may serve as indicator of air pollutants. For example, birds such as Oriental White-eye, Coppersmith Barbet, Indian Grey Hornbill, Brown-headed Barbet are sensitive to Ozone and are absent in areas with high level of Ozone. Th e species composition of greenspaces is critical in mitigating air pollution and also to sustain avian diversity in urban landscape. Detailed studies involving larger number of sampling sites may further substantiate the conclusions drawn from the present study. We like to thank University of Delhi for providing research grant (SHC 307) under Innovation Project Scheme. We are thankful to the Principal Dr. Shashi Nijhawan for supporting and encouraging us throughout this project. We acknowledge Delhi Pollution Control Committee, Government of NCT Delhi for access to the data on air pollution. Th e contributions of Arunender, Aakash, Hitakshi, Shruti and Tarun for their help during fi eld data collection are also duly acknowledged. References Ali, S, Ripley, S. D. 1983. Hand Book of the Birds of India and Pakistan. Compact Edition. Oxford University Press, New Delhi. 50 V. Kumar, V. Jolli, C. R. Babu Aronson, M. F., La Sorte, F. A., Nilon, C. H., Katti, M., Goddard, M. A., Lepczyk, C. A., Warren, P. S., Wil- liams, N. S., Cilliers, S., Clarkson, B., Dobbs, C. 2014. A global analysis of the impacts of urbanization on bird and plant diversity reveals key anthropogenic drivers. Proceedings of the Royal Society B: Biological Sciences, 281 (1780), 20133330. Bibby, C. J., Burgess, N. D., Hill, D. A., Mustoe, S. H. 2000. Bird Census Techniques, Academic Press, CA. BirdLife, International. 2004. Birds as a ‘Quality of Life’ indicator in the United Kingdom. Downloaded from http://www.birdlife.org (accessed 28/07/2019). Devictor, V., Julliard, R., Couvet, D., Lee, A., Jiguet, F. 2007. Functional homogenization eff ect of urbanization on bird communities. Conservation Biology, 21 (3), 741–751. Eeva, T., Lehikoinen, E., Nikinmaa, M. 2003. Pollution-induced nutritional stress in birds: an experimental study of direct and indirect eff ects. Ecological Applications, 13 (5), 1242–1249. Eeva, T., Tanhuanpää, S., Råbergh, C., Airaksinen, S., Nikinmaa, M., Lehikoinen, E. 2000. Biomarkers and fl uc- tuating asymmetry as indicators of pollution-induced stress in two hole-nesting passerines.  Functional Ecology, 14 (2), 235–243. Filloy, J., Zurita, G. A., Bellocq, M. I. 2019. Bird diversity in urban ecosystems: the role of the biome and land use along urbanization gradients. Ecosystems, 22 (1), 213–227. Gregory, R. D., van Strien, A. 2010. Wild bird indicators: using composite population trends of birds as mea- sures of environmental health. Ornithological Science, 9 (1), 3–22. Grimm, N. B., Faeth, S. H., Golubiewski, N. E., Redman, C. L., Wu, J., Bai, X. and Briggs, J. M. 2008. Global change and the ecology of cities. Science, 319 (5864), 756–760. Hammer, Ø., Harper, D. A., Ryan, P. D. 2001. PAST: paleontological statistics soft ware package for education and data analysis. Palaeontologia electronica, 4 (1), 9. Herrando, S., Brotons, L., Anton, M., Paramo, F., Villero, D., Titeux, N., Quesada, J., Stefanescu, C., 2016. As- sessing impacts of land abandonment on Mediterranean biodiversity using indicators based on bird and butterfl y monitoring data. Environmental Conservation, 43 (1), 69–78. Kazmierczak, K., Perlo, V. B., 2015. Om fi eld guide birds of India. Om Book International, New Delhi, India. Khera, N., Mehta, V., Sabata, B. C. 2009. Interrelationship of birds and habitat features in urban greenspaces in Delhi, India. Urban Forestry & Urban Greening, 8 (3), 187–196. Kumar, V., Jolli, V., Babu, C. R. 2019. Avenue plantations in Delhi and their effi cacy in mitigating air pollu- tion. Arboricultural Journal, 41 (1), 35–47. Landrigan, P. J. 2017. Air pollution and health. Th e Lancet Public Health,2 (1), e4–5. Lawton, J. H., Bignell, D. E., Bolton, B., Bloemers, G. F., Eggleton, P., Hammond, P. M., Hodda, M., Holt, R. D., Larsen, T. B., Mawdsley, N. A., Stork, N. E. 1998. Biodiversity inventories, indicator taxa and eff ects of habitat modifi cation in tropical forest. Nature, 391 (6662), 72. Marlier, M. E., Jina, A. S., Kinney, P. L., DeFries, R. S., 2016. Extreme air pollution in global megacities. Current Climate Change Reports, 2 (1), 15–27. McKinney, M. L. 2006. Urbanization as a major cause of biotic homogenization. Biological conservation, 127 (3), 247–260. NASA. 2013. https://earthobservatory.nasa.gov/features/ColorImage (accessed 15/02/2020). Newbold, T., Hudson, L. N., Arnell, A. P., Contu, S., De Palma, A., Ferrier, S., Hill, S. L., Hoskins, A. J., Lysen- ko, I., Phillips, H. R., Burton, V. J. 2016. Has land use pushed terrestrial biodiversity beyond the planetary boundary? A global assessment. Science, 353 (6296), 288–291. Ormerod, S. J. and Tyler, S. J. 1993. Birds as indicators of changes in water quality. In: Birds as monitors of en- vironmental change.  Springer, Dordrecht, 79–216. Pauw, A., Louw, K. 2012. Urbanization drives a reduction in functional diversity in a guild of nectar-feeding birds. Ecology and Society, 17 (2), 27. Sorace, A., Colombari, P., Cordiner, E. 1999. Bird communities and extended biotic index (EBI) in some tribu- taries of the Tiber river. Aquatic Conservation: Marine and Freshwater Ecosystems, 9 (3), 279–290. UNEP Frontiers. 2016. Report: Emerging Issues of Environmental Concern. United Nations Environment Pro- gramme, Nairobi. Urfi , A. J., 2010. Using heronry birds to monitor urbanization impacts: a case study of Painted Stork Mycteria leucocephala nesting in the Delhi Zoo, India. Ambio, 39 (2), 190–193. Received 28 June 2021 Accepted