Some notes on the communal behaviour of Socotra Cormorant, Phalacrocorax nigrogularis

Socotra cormorants Phalacrocorax nigrogularis are the conservation flagship species for the thirty or more Bahraini islands of the Hawar archipelago * (QB28) that lie close to the Qatar peninsula. Endemic to the Arabian Gulf and Arabian Sea the status and biology of the Socotra cormorant still requires much investigation. In the following notes details are given of observations of communal behaviour outside of the breeding season. Once the cormorants move into the breeding colony in late September the daily routine changes dramatically, the nature of that behaviour will be the subject of further reports.

Despite full species protection in Bahrain, numbers at the largest documented breeding site, on the island of Suwad al Janubiyah (Suwad) in Hawar, show evidence of a decline in the long term. The concern at this decline is compounded by the fact that over the past decade many breeding sites around the gulf have been lost as sites have been encroached upon by development or subjected to prolonged human disturbance. Also as a ground breeding species this cormorant is known to be vulnerable to natural disasters, such as that recorded at Suwad in late November 1997 when heavy rain flooded the entire colony, filling nests with water, drowning incubated eggs and causing tens of thousands of developing chicks to die from hypothermia. In the last breeding year (2002/2003), in early April an isolated thunderstorm accompanied by strong winds (strong enough to remove the roof a building on the main island of Hawar) resulted in the sudden termination of the breeding season on Suwad with the death of thousands of chicks. The carcasses of dead chicks were found to cover an area of 19,750 M' with an average density of 43 per 100 M2 . This particular disaster is thought to have been minor in comparison to what happened in 1997.

Roosting

Outside of the breeding season (the breeding season is October to April) the birds that constitute the Suwad breeding colony disperse around the Gulf of Salwa, establishing large roosts at isolated locations. On the exposed western shores of offshore islands * (e.g. Umm Nassan, QA29) that lie between Bahrain and Saudi Arabia, a days flight to the north west of Hawar, over fifty thousand birds congregate daily to form one of the largest roosts known. The largest congregations occur at these locations throughout July and August. Roosts are well defined and occupy the smallest possible ground footprint, generally with a forward edge on the shore. The birds press tightly together (shoulder to shoulder) even though there is plenty of space available. Just why the birds form such dense roosts is not yet understood; roosts sites monitored all provide exposure to seasonal winds and cooling breezes but the birds are so tightly packed that the benefits of any breeze would seem to be negated for the majority. One explanation could be that such dense congregations provide the maximum shade area for the most feet, important when summer ground or surface temperatures are known to exceed 60 'C. Estimates of the numbers of individual birds that constitute a particular roost were difficult to ascertain but video evidence taken during August 2003 indicates that at the largest roost monitored, on the island of Umm Nassan, exceeded 30,000 birds. A fact that is reflective of the highly sociable nature of the species, evident in many aspects of the daily and seasonal life cycle. Other than a small numbers of birds observed throughout the day moving in and out of roosting sites, the main feature of summer roosting behaviour were the prolonged periods of almost total inactivity. Roosts were observed to remain in situ regularly from dawn until the middle of the afternoon, when the days fishing foray would start. Almost without exception other activities when initiated were undertaken collectively by the entire roosting flock, irrespective of its size. Other than the afternoon fishing forays the most frequently observed activity involved mass fly outs into the adjacent shallow bays close offshore, where the entire flock would bathe and subsequently raft. Rafting behaviour was particularly evident during periods of hot calm and humid weather but would also occur if a roost was disturbed. Once formed out at sea a raft, which also has the smallest possible footprint and is best described as a floating roost, could remain intact for many hours with the birds simply drifting or floating along the edge of tidal steams or channel currents.

Bathing

On several occasions afternoon fishing forays were initiated by the whole flock from the rafted up location. Otherwise daily afternoon fishing forays at the roost were announced around 15.00 hrs by an almost predictable surge of activity. Often initiated by birds at the back and acting as if on command, large groups were observed to fly out and drop into the sea in front of the mass of birds on the shore to initiate their individual bathing routine. Almost instantly, as if this were a trigger, the birds on the shore would swim out rather than fly the short distance to bathe, joining the constantly growing mass of bathing birds. The birds instinctively concentrate themselves during this manoeuvre into a small well-defined section of the bay. Caught in strong currents close inshore, the bathing birds quickly drift out into deeper water, with the space left closer to the shore instantly occupied by joining birds. When the rate of departure from the roost exceeded the available space close inshore, birds on the wing would over fly any on the water and drop down at the leading edge, out in the bay, creating a processional line of activity both in the air and on the sea. This movement of birds some flying, some swimming, some bathing would continue until the whole roost were offshore, a mass of wing flapping and splashing water that eventually would form up into a long loose raft, drifting slowly seaward on the current, the prelude to the most spectacular mass activity of all, takeoff for the days fishing foray.

Socotra cormorants bathe at the earliest opportunity each time they leave the land, a routine that never varies. By snorkelling amongst the bathing birds it was possible to view both individual and the massed bathing routines at close quarters viewing underwater, surface and aerial activity. The bathing routine for individual birds varies only slightly and in general is started almost immediately a bird settles on the water. Initiated as if looking for fish by the ducking of the head on an extended neck under the water, the head and neck are jerked downwards in stabbing motion and the feet used to scratch at the feathers of the breast or the neck to assist with the removal of dirt or parasites. Whilst in the head down position the tail is raised and the feet used to vigorously flick water up into the ventral area, occasionally the feet were also used to propel or torpedo the bird along the surface at considerable speed to force water up over the back. Birds were only very occasionally seen to actually swim underwater as a prelude to the bathing routine. After attending to the head, the bird would rise and with a quick flap of its wings, reassume a sitting position on the water, then with wings extended in a cupped or half cock position to the side, beat them repeatedly against the surface of the water, holding the wings at an angle that would ensure that wing tips are driven vertically down into the water. The wings are then drawn into the body and used to move or scoop more water up and over the back with semi folded wings used to work the water into the back feathers. The processes of wing beating and water scooping over the back or occasionally the entire routine was repeated several times. The bird would then rise up as with wings outstretched and flap them vigorously, holding itself erect with just the tail and feet in the water before finally holding the wings out to dry from a slightly raised sitting position on the water. Preening of individual feathers continues after bathing from the swimming position for some considerable time. When undertaken en mass, bathing routines are spectacular and dramatic, birds are constantly dropping down into or taking off while others bathe creating a mass of apparently confused movements. Such bathing activity could easily be confused with fishing activity when undertaken by an entire roost of tens of thousands of birds as they roll over each other in a continuous leap frog motion.

Food finding movements

The mass departure of Socotra cormorants from a raft or a roost is totally unlike the choreographed ballet of the more commonly observed flight formations. These are the continuous chains or long strings and "V" shaped formations that are often observed along distant horizons or from shorelines. These strings, spectacular in their own right (as such a passage can be continuous for many hours), generally represent the processional return of tens of thousands of birds to roosting or breeding sites after fishing, or a localised movement from one area to another. Mass fishing departures, in contrast, involves the almost simultaneous takeoff and flight in a single direction of an entire flock; passage past an observer will be completed in minutes rather than hours even for the largest of congregations. The flock retains an identity; it takes the form of a dense black sheet of birds flying at wave height barely a metre or so above the water. Flying within an almost definable rectangular area, no single bird leads, the flock moves en masse; on a broad front several hundred metres across, whist the remainder follow as a unit in close pursuit behind.

Fishing

Flying along their selected route with thousands of eyes searching along the front line of the moving flock, when fish are encountered forward movement would come to an abrupt halt as the birds alighted onto the sea all together to opportunistically dive for fish. Birds always settle onto the sea before diving below to fish. If the shoal of fish encountered is sizeable a fishing frenzy would develop as those coming up from behind would also alight en masse on the sea in the same place. Birds were often seen to return to the surface with fish in their beaks and instantly fly off to settle away from the fishing area before attempting to swallow their catch. Those that attempted this in the fishing area risked being robbed of their catch by others. It is not known if these cormorants can swallow fish on the move either above or below the water. In a very short time a fishing frenzy takes shape with birds dropping down from what soon becomes a circling mass above the hot spot, large numbers of birds, some barely visible with heads down, cover the surface of the sea as others that have completed a dive continuously rise from the fishing area to fly off, either to circle again or find a haven to swallow their catch. Many appear to be just sitting on the sea as if taking a rest as the back markers catch up and overtake the leading edge. From a distance, only the birds dropping down or those taking off are visible in what appears to be a continuous motion, which could be mistaken, given the surface commotion during such frenzies, for birds plunge diving straight into the sea.

The longest feeding frenzy observed (recorded on video) lasted over thirty minutes and involved around 5000 birds however most events were of much shorter duration. Fishing frenzies were occasionally observed to occur along the fringes of the moving mass but seldom behind the leading edge of birds. Once the leading birds were otherwise engaged in feeding, the moving sheet of birds coming up from behind would over fly the area, generally instinctively knowing whether it was worth stopping to fish or continue as a new front. Such a leap frog motion characterises most fishing forays and provides in due course. a consequence of the long flight path, an opportunity for most birds to be at the front. Occasionally more than one feeding area would develop resulting in the flock being longitudinally stretched out along the flight path, but such was the manner of feeding forays that in time the entire flock would eventually reform into a single sheet of birds again and again, to move across the more open and deeper areas of sea grass.

Return flights to roost

Eventually appetites are satisfied and the mass fishing effort breaks down and skeins of birds break off from fishing to return to their point of departure. It is on the return flight that the cormorants create the long processional low flying lines of the familiar continuous strings or "V" shaped flight formations, with seemingly choreographed movements and 'Mexican waves'. Mass fishing forays were followed on several occasions throughout the summer of 2003, to the point when flocks begin to return. (Provided the fishing areas remained inside Bahrain waters). Flocks would often cover distances in excess of 40 or 50 kilometres on the outward leg daily. Such long distance fishing forays were observed to be the norm rather than the exception. Since fishing flocks move at a slower speed than birds returning to roost areas it was possible to follow them on the outward leg using a boat with a top speed of 28 knots. However it was not possible to keep up with the returning skeins of birds.

The outward leg was seldom a straight line and on numerous occasions contained right angle turns and even back tracks. For the return, cormorants always took the shortest possible route home. But on such flights transits overland are avoided or kept to an absolute minimum, thus flights often were via key turning points around islands or land obstructions.

On or about 14 August 2003 the large roosts on western islands of Bahrain start to dissolve with the flocks making a leisurely flight southwards to reform as a large roost on Rubud al Gharbiyah or Rubud Ash Sharqiyah (the Rubuds) the most northerly islands of the Hawar archipelago. This large seasonal movement follows the western coast of Bahrain south around Ras Al Bar, the southern tip of the main island of Bahrain, before crossing the 22 kms of open sea to Hawar. The roost on the Rubuds remained in place until around 23 September 2003 when the birds occupied the breeding grounds on Suwad. The first eggs at the breeding colony can be expected seven to ten days from that time.

Pre-breeding fishing Forays

Fishing forays from the Rubuds are massive affairs involving numbers thought to be in excess of 50,000 birds. Only small rafts amongst the islands were encountered during this time, the birds remain largely inactive during the day until around 13.30 hrs when, in a similar fashion to that observed at Umm Nassan, the entire roost would leave together to first bathe then move as a compact flock out to sea to fish. The birds from this pre-breeding location seem to have a circular route for their daily fishing activities, which results in regular fishing forays inside the territorial waters of Qatar. The chosen fishing direction for the day however is announced within the roost by an initial wholesale movement of birds to either the eastern or western corner of the islands, points of departure coincident with access to deeper water. The roost site lacks a good adjacent bathing area as a large sand bank less than half a metre in depth at high tide, extends nearly 2 km offshore directly to the north, however the bank ensures little or no human (fishermen) disturbance at this time.

Despite the preference of the gathered flock to fish communally small fishing parties of generally less than fifty birds were regularly encountered particularly late afternoon along the flyway home to roost. Unlike that employed by the massed flock during fishing forays the behaviour and technique of these small fishing groups was found to he quite different. These parties congregate as in a miniature raft sitting tightly packed on the water, one or more birds would be seen looking under the water Oust their backs visible) so that once fish were spotted, with a raise of a head and an arch of the back one would disappear from view in a dive. Alert to the slightest movement of any in the party almost simultaneously the balance of the fishing party would dive, disappearing as if this were a rehearsed synchronous motion. Surprisingly the party would surface in a similar manner retaining its tight formation. Such fishing parties however would not allow a close approach by snorkelling thus the actual fishing technique employed underwater or the nature of the schooling fish remains unknown.

Fish species observed in the diet

Divers record that Socotra cormorants can dive straight down to depths in excess of 18 metres, they are also regularly found drowned in fishing traps at varying depths up to 20 metres. Observations of identifiable fish taken indicate that the following species are included, (some only as fingerlings) in the diet.

 

 -  Atherinomorus lacunosus Silverside

 -  Atule mate Yellowtail Scad

 -  Selar crumenophthalmus Bigeye Scad

 -  Sardinella albella White Sardine

 -  Sardinella gibbossa Goldstripe Sardine

 -  Hemiramphus far Spotted Halfbeak

 -  Siganus javus Streaked Rabbit Fish

Howard King, Po Box 15344, Manama, Bahrain, Tel 39642729,

Email: howardk@batelco.com.bh

*Note  numbers indicated *(QB28) relate to squares used in the Arabian Breeding Bird Atlas  (ABBA)