Summary: | 碩士 === 國立臺灣大學 === 生態學與演化生物學研究所 === 101 === The habitats of shorebirds can be divided to foraging habitat and roosts, the latter is the area in which shorebirds rest in flocks when the intertidal flats are inundated during high tide period. The coastal area of Chang-Hua, Taiwan is an important wintering site and also a stopover site on the East Asian – Australasian Flyway. The nutrient rich mudflat in the central west-coast has been the major feeding habitat for thousands of shorebirds. However the neighboring land area is now occupied by man-made landscapes, forcing the shorebirds to roost on aquacultural ponds’ dikes during high-tide. Previous studies conducted in natural coasts indicated that predation risk and energy limitation are the major factors affecting roosts selection. However few studies has been made in aquacultural environment. Thus, the aim of this study is: (1) To verify whether the usage of roosting habitat of shorebirds is selective, and to compare the habitat preferences between different species. (2) To compare the relative importance of environmental variants.
Bird surveys were conducted monthly from October, 2011 to May, 2012. The sample area is the aquacultural zone located in Fangyuan Township, Changhua County. Six environment variants of dikes were collected through fieldwork and remote-sensing images. Length, vegetation coverage and sky visibility are the variants related to predation risk. Distance to feeding site, substrate type and shelter are related to energy limitation. Cluster analysis is used upon species of small sample size, to cluster them into groups based on the similarity of the dikes they used. Univariate analysis is used to determine whether there is environment preference of roosts selection. Model analysis is used to compare the relative importance of variants.
In the period, 18 species, 52922 individuals were recorded, which totally used 147 dikes as their roosts. Kentish Plover (Charadrius alexandrinus) had the highest abundance and number of dikes used, followed by Dunlin (Calidris alpina). Except these two species, two groups were made by cluster analysis. Group 1 includes Pacific Golden Plover (Pluvialis fulva), Greenshank (Tringa nebularia), Marsh Sandpiper (Tringa stagnatilis) and Redshank (Tringa totanus). Group 2 includes Grey Plover (Pluvialis squatarola), Terek Sandpiper (Xenus cinereus), Eurasian Curlew (Numenius arquata), Great Knot (Calidris tenuirostris), Greater Sand Plover (Charadrius leschenaultii), Lesser Sand Plover (Charadrius mongolus), Ruddy Turnstone (Arenaria interpres), Red-necked Stint (Calidris ruficollis), Sanderling (Calidris alba), Red Knot (Calidris canutus), Curlew Sandpiper (Calidris ferruginea) and Tattler (Tringa brevipes). The result of univariate analysis indicated that all species prefer to roost on dikes with longer length, higher visibility, shorter distance to feeding site and with sheltered side. The preference for vegetation coverage of group 1 is opposite to other species. The result of model analysis indicated that length, sky visibility, vegetation coverage and substrate are more important than other factors.
This study verifies that like in natural coasts, roosts selection of shorebirds in aquacultural environment is also affected by predation risk and energy limitation. Different preference for vegetation coverage among species shows that different anti-predation strategies might be used. Besides flocking, shorebirds of group 1 possibly use camouflage simultaneously as their anti-predation strategy when roosting. Predation risk is the major factor affecting roosts selection of shorebirds. This might be attributed to the characteristic of aquacultural environment, which usually has fewer large roosting area and more visual obstruction than natural coasts due to the existence of complicated artificial buildings.
|