South China Sea 



Fig.l. Geography of the southern half of the South China Sea showing bird 

 survey locations. Numbered segments are bird location survey 

 locations for R/V Akademik Korolev, 23-310ctober 1988. 



northern two-thirds of the South China Sea, demonstrating the 

 potential for long-distance (ca. 1,000 km) migration, they did 

 not demonstrate that a corridor for land-bird migrants exists 

 further south between Vietnam and Borneo. In this paper, we 

 present data from the South China Sea that strongly support the 

 presence of such a migration route. 



Study Area and Methods 



We encountered migrant land birds during our 

 23-31 October 1988 indirect passage (Fig. 1) from Balabac 

 Strait to Singapore on the Soviet research vessel Akademik 

 Korolev. While in transit, we observed birds during dawn-to- 

 dusk seabird surveys from the flying bridge (12 m above sea 

 le\ el ). During a 3-day period while the ship was anchored or 

 drifting without power to conduct oceanographic research 

 (Fig. 1, Station 13;06°01'N, 106°55'E), we conducted periodic 

 v. alking inspections of the ship (usually at half-hour intervals) 

 and searched the ship each night by flashlight to count roosting 

 birds. Five raptors anil several barn swallows were captured by 

 hand (primarily at night) and examined for physical condition. 



Results and Discussion 



During our 9-day passage, we encountered about 1 50 land 

 buds 1 121 by conservative count, 84 minimum count, Table 1 1 

 representing 14 families. Almost all were migrants that winter 

 (at least in part) south of the South China Sea. Most of these 

 birds (9ft by conservative count ) arrived on the ship during the 

 ! da) period while we were stationary (Fig. 1, Station 13) 

 around 350 km southeast of the southern tip of [ndo-China. 

 The presence oi land buds at (Ins location suggests that they 

 were in passage across the South China Sea from Indo-Chin.i 

 to the ( ireater Sunda Islands. The low bird counts seen before 



arriving at and after leaving this location (Table 2) suggest that 

 this spot lies on a rather narrow migratory pathway although, 

 alternately, birds may have been reluctant to approach a moving 

 vessel. Simpson's 1982 observations (Wells, in prep.), made 

 in the Terengganu oil field (Fig. 1 ) very near our cruise track, 

 suggested that he was sampling the same corridor we visited; 

 if so, the pathway may be somewhat wider than we detected. 



The number of birds we observed (Tables 1,2) is small 

 when compared with record counts for well-known migration 

 pathways. However, our visit was brief and probably too late 

 for detecting the bulk of migrating land birds. Simpson's 

 ( 1983a) dates for six of nine frequently encountered land birds 

 near northeastern Borneo fell before the time of our visit, and. 

 just as important, migrating land birds most often aggregate 

 where land and water configurations encourage them to collect 

 (e.g.. on north or south projecting peninsulas). By contrast, we 

 were on the open sea where birds are much less likely to 

 concentrate. Considering these factors, it seems likely that 

 adequate spatial and temporal sampling will reveal many 

 thousands of land birds moving south from Indo-China across 

 the South China Sea. 



Although our records and those of Simpson's (Wells, in 

 prep.) demonstrate that a sizable migration is probably normal 

 across the South China Sea. we should mention an alternate 

 hypothesis that may help explain the presence of these birds 

 where and when we observed them. First, our passage occurred 

 when Typhoon Ruby was ravaging the Philippine Islands 

 (Anon.. 1989). Although we did not encounter heavy seas or 

 strong winds, some of the birds we observed may have been 

 forced out to sea, if nonmigratory. or shunted away from their 

 normal migration route, if migratory, by the storm. However, 

 most of the birds we observed far from land (Table 3) are 

 known to be strong migrants. The four hawks tentatively 

 identified as shikras (Accipiter badius) and crested goshaw ks 

 (A. trivirgatus), and the dove (Streptopelia sp. ) are the only real 

 surprises, although a few others in Table 3 would not be 

 expected this far from land. 



Flight direction may give some indication of the likelihood 

 of either hypothesis. If the birds were displaced migrants, they 

 would probably have been heading southwest (i.e.. away from 

 the storm). If in passage from a concentration zone on the Indo- 

 China Peninsula to Borneo, they should have been heading 

 southeast to encounter our vessel. If. as we observed, the 

 raptors (33'f of all land birds) were foraging at sea (Ellis etal., 

 1 990 ) rather than migrating, there would likely be no consistent 

 trend in their flight direction. In Fig. 2. there is no clear east- 

 west trend in arriving or departing flights. However, although 

 the data are very few strong southward and westward 

 components are e\ idem. In constructing Fig. 2. we eliminated 

 directional readings for birds seen on cruise track segments 

 1-7 and 20-30 because these segments were near enough to 

 land (i.e., within 1 00 km ) that the birds' flight directions could 

 have been influenced by sight or sign of nearby land. In 

 addition, all Right bearings could have been influenced h\ the 

 presence of the ship. 



Physical condition of the birds w e obsen ed mas also be an 

 indicator of the regularity with which this migration route is 

 used. If a high proportion of the know n o\ erseas migrants were 



248 



