How verticalization of cities affects bird. A case study of Magpies in the urban area of Beijing, China

Transcription

1 How verticalization of cities affects bird distribution and behaviour? A case study of Magpies in the urban area of Beijing, China Hiroshi HASHIMOTO 1 Mirai DOMON 1 Koji OOHATA 2 Shuhua LI 3 Hiroshi HASHIMOTO 1, Mirai DOMON 1, Koji OOHATA 2, Shuhua LI 3 1 Meijo University, Japan; 2 Wild Bird Society of Japan, Japan; 3 Tsinghua University, China

2 Introduction ti Verticalization of urban landscapes caused by urbanization and urban redevelopment of downtown areas is rapidly increasing in the East Asian countries.

3 Introduction(2) ti Birds have wings, but tall buildings will cause birds to flyover and will block the bird s view. To evaluate the influence of tall buildings on the distribution ib ti of urban birds, we studied the habitat preferences of Blackbilled Magpie (Pica pica) in the urban area of Beijing, China.

4 The Black-billed magpie belongs the family of Crows, and breeds widely in urban and suburban areas of Asia and Europe. A pair of Magpie protects their own breeding territory around their nest, especially top of the trees in their territory. Their nest is big and like a ball, and easy to find.

5 Recent urbanization with tree plantation programs in France led to an increase in the probability of magpie occupancy (François et al. 2008). Do recent urbanization and urban redevelopment with tall buildings in the East Asian cities also affect positively on the magpie occupancy? We studied nest site preference and home range of Magpies in residential i area(r) and urban center(u). Nest Nest

6 (1) Nest site preference We surveyed the distribution of magpie nests in the March 2008, and compared areas of tree cover and buildings around between nest trees and random trees. We also created a habitat model of the magpie using logistic regression model.

7 Creating a 3 D GIS map The vegetation map was created from an ALOS AVNIR2 satellite image, and trees were classified based on the NDVI value. The 10m resolution of digital surface model (DSM) of this area was created from a stereo pair of ALOS PRISM satellite images by RESTEC in Japan.

8 = Tree distribution DSM 3-D GIS map We classified trees and buildings using these two maps. The spatial resolution of these maps was 10m x 10m, and the vertical resolution of DSM was 1m.

9 Mapping the Magpie s nest distribution ib ti We surveyed the distribution of the Magpie s nest in March, 2008, and found 189 nests (including old nests) in the study area. 33 nests of them were thought to be active 33 nests of them were thought to be active nests in that year.

10 Environmental variables We measured the areas of 1) trees, 2) buildings taller than the nest ttree, >nest ttree+10m, >nest+20m, and >nest+30m, 3) visible trees which can be seen from the top of the nest tree, in the four different radii (100m, 150m, 200m and 250m) of circles from their nest and 60 trees randomly selected.

11 How to measure the area of visible trees which can be seen from the nest (by Arc GIS). 1 30m 2 DSM - 30m 3 +30m at the target point 4 Observe (View shed) the target from trees 30m in m (set RADIUS2) with 30m OFFSETA..

12 Nest ttrees and random trees 33 nests of them were thought to be active nests in that year. We compared the environmental variables between the nest trees and random trees by t test.

13 The results of comparison of mean area (m 2 ) of environmental variables.

14 Habitat model of the magpie using logistic regression model The area of visible trees in the radius of 200m was the only variable in the best model. dl Area of Visible trees (ha) in 200m

15 (2) Home range We observed the breeding behaviours of magpies for 23.5 hours in two different urban areas in Beijing in the May 2008, and compared the areas of home range, tree cover in their home range. We also compared the area of visible trees from perch points and random points in each home range.

16 Comparison of home range of the Magpie in a residential area (R) and in an urban center (U). Residential area of east of Gulou (R) Urban center along Jinpo Jie (U)

17 Comparison of home range of the Magpie in a residential area (R) and in an urban center (U). Area of home range (Minimum Convex Polygon) Gulou (R) << Jinpo Jie (U) Maximum distance of home range from their nest Gulou (R) < Jimpo Jie (U) Area of tree cover in their home range Gulou (R) < Jinpo Jie (U) The area of home range and tree cover in the urban center was larger than that in the residential area. But, magpies in the urban center did not use all trees in their home range.

18 Comparison of home range of the Magpie by the Kernel method in a residential area (R) and in an urban center (U). Area of home range (by Kernel) 50% Gulou (R) < Jinpo Jie (U) 90% Gulou (R) < Jinpo Jie (U) 95% Gulou (R) < Jinpo Jie (U) Area of tree cover in their home range (by Kernel) 50% Gulou (R) > Jinpo Jie (U) 90% Gulou (R) = Jinpo Jie (U) 95% Gulou (R) > Jinpo Jie (U) Although the areas of home range in the urban center was larger than that in the residential area, magpies used almost same area of trees in each habitat.

19 Comparison of mean area (m 2 ) of visible trees which can be seen from the perch points or random points. Gulou (Residential area) Perch points n.s. Random Jinpo Jie (Urban Center) Perch points > Random The pair of magpies breeding in urban center have several good perch points (including the roof of buildings) where they can watch many trees in their territory.

20 Conclusion (1) The results of habitat modeling and observations about tbreeding behaviors bh at two different habitats in Beijing shows that Magpies can breed in highly urbanized area, if they have enough visible trees from their nest or several good perch points (including buildings) near the nest where they can watch many trees in their territory.

21 Conclusion (2) The results from this study, we recommend to make spaces at the corners of crossroads, to make a setback in front of buildings and plant trees in these spaces (and along the streets) for planning birds friendly cities.