Marinao and Yorio • SEABIRD MORTALITY AND FISHERY DISCARDS 
711 
Clara, and Escondida Island (Yorio et al. 1998b). 
Punta Tombo (44° 02' S. 65° 11' W) is one of the 
main breeding sites of Magellanic Penguins (Splw- 
niscus magellanicus; 175.000 pairs) on the Patago¬ 
nian coast, and has the main Kelp Gull colony in this 
coastal sector (5,400 pairs). Other species breeding 
at this location include Imperial Shag ( Leucocarbo 
iiirkeps) and Brown Skua ( Stercorarius atitare- 
ticus). Punta Clara (43 58' S, 65 15' W). 7 km 
north of Punta Tombo, includes 70,000 Magellanic 
Penguin and 40 Kelp Gull pairs. About 500 Imperial 
Shag pairs breed at Escondida Island (43 43' S, 65 
17' W). Kelp Gulls in the study area start laying in 
early November, most eggs hatch in early January, 
and chicks fledge during February (Bertcllotti and 
Yorio 1999). Magellanic Penguins arrive at breed¬ 
ing sites in early September, start laying in early 
October, eggs start hatching in early November, 
most chicks Hedge in February, and the last adults 
leave the colony for their winter migration during 
April (Bocrsmaetal. 1990). Imperial Shags arrive at 
the Punta Tombo colony during August, start laying 
in late October, eggs hatch in late November, and 
the last chicks fledge in late March (Malacalza 
1984). 
Species and Abundance of Seabirds Attending 
Vessels.— We gathered information on board 15 
coastal shrimp trawlers (38% of the fishing fleet) 
during regular fishing operations from December 
2006 to March 2007 and from December 2007 to 
February 2008. We gathered information during 
each trip for only one haul per day during midday, 
totaling 20 hauls (20 fishing days) in each of the 
fishing seasons. More than 80% of hauls in each 
season corresponded to the months of January and 
February, i.e.. during seabird incubation and chick 
rearing. We identified seabirds attending the 
vessel during all hauls to species level and 
recorded their numbers. We made counts during 
a l()-min observation period from the stern of the 
vessel, only once at the beginning of discarding 
(sorting and discarding fish while tow ing the net), 
covering up to a IO()-m radius. We also gathered 
information during haulback (lifting of the net to 
the vessel) during the 2007—2008 fishing season 
and identified Kelp Gulls into adult and young 
(juvenile and sub-adult) individuals based on 
plumage characteristics (Bo el al. 1995). We 
defined frequency of occurrence us the percentage 
of hauls in which each species was observed. In 
addition, we recorded the number of fishing 
vessels operating within sight (< 10 km) for each 
haul. 
Seabird Incidental Mortality. —We obtained 
information on incidental captures of seabirds in 
nets, including species' identity and number of 
birds caught in each haul, from the data base of 
the On-board Observer Program of Chubut 
Province, of which the first author was part, for 
the 2006-2008 period (n = 3,149 hauls). In 
addition, for the 40 hauls in W'hich we evaluated 
seabird attendance, wc quantified their interaction 
with w'arp cables to examine the occurrence of 
contacts and associated mortality. Observations at 
each haul were made from the stern of the vessel 
during the period when fish were discarded 
(— 10 min to 3 hrs. depending on size of the 
catch). 
Spatial Distribution of the Fishery and Discard 
Composition. —We mapped the distribution of 
hauls in both fishing seasons obtained from the 
data base of the On-board Observer Program of 
Chubut Province using Arc View 3.2 (ESRI 1998) 
to assess the spatial and temporal distribution of 
operating shrimp vessels. We present haul distri¬ 
bution as density maps on a 5 X 5 km grid. We 
calculated the distances of each haul to Punta 
Tombo, Punta Clara, and Escondida Island using 
the same software. 
We estimated the total amount discarded by the 
coastal trawl fishery in each study year extrapo¬ 
lating the mean value (in kg) corresponding to 
hauls evaluated by the On-board Observer Pro¬ 
gram during the entire fishing season (Nov to 
Mar) to the total number of hauls made by the 
fishery during that period. Wc estimated the total 
number of hauls made by die fishery by dividing 
the declared catch in the study period (Secretaria 
dc Agriculture, Ganaderia, Pesca y Alimentation 
of Argentina, unpub!, data) by the mean catch of 
observed hauls (On-board Observer Program of 
Chubut Province, unpubl. data). We estimated the 
amounts discarded per haul subtracting the 
retained catch from the total catch, and obtained 
the total catch for each haul averaging the 
independent estimates made by the vessel captain 
and the on-board observer. 
We obtained information on the catch compo¬ 
sition by shrimp vessels from the data base of the 
On-board Observer Program of Chubut Province, 
totaling 1,219 hauls corresponding to the months 
of December to March and December to February 
in the 2006-2007 and 2007-2008 fishing seasons. 
respectively. Personnel of the On-board Observer 
Program estimated the abundance of caught prev 
in each haul and assigned them to one of four 
