UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of prey abundance on the diet, maternal attendance and pup mortality of the South American… Soto, Karim H. 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2005-0115.pdf [ 4.64MB ]
Metadata
JSON: 831-1.0074872.json
JSON-LD: 831-1.0074872-ld.json
RDF/XML (Pretty): 831-1.0074872-rdf.xml
RDF/JSON: 831-1.0074872-rdf.json
Turtle: 831-1.0074872-turtle.txt
N-Triples: 831-1.0074872-rdf-ntriples.txt
Original Record: 831-1.0074872-source.json
Full Text
831-1.0074872-fulltext.txt
Citation
831-1.0074872.ris

Full Text

THE EFFECTS OF PREY ABUNDANCE ON THE DIET, MATERNAL ATTENDANCE AND PUP MORTALITY OF THE SOUTH AMERICAN SEA LION (OTARIA FLAVESCENS> IN PERU b y " Karim H. Soto B.Sc. Zoology, Universidad Nacional Agraria La Molina, Peru, 1999 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES ' ZOOLOGY NOVEMBER 2004 We accept this thesis as conforming to the reauired standard THE UNIVERSITY OF BRITISH COLUMBIA © Karim H. Soto, 2004 Abstract The Peruvian upwelling system is one of the most unpredictable but productive marine ecosystems in the world that is regularly affected by El Nino and La Nina events. As a consequence, South American sea lions (Otaria flavescens) inhabiting the Peruvian coast face stochastic fluctuations in the abundance and distribution of their prey. The goal of my study was to determine the effects of marine environmental changes on the reproductive success of South American sea lions in Peru on the Ballestas Islands during the 1997-2002 breeding seasons. I also wanted to assess whether changes in diet and maternal behaviour during this time were proxies of major changes in the abundance of marine resources. South American sea lions consumed primarily anchovy and squat lobster during the 1999-2001 La Nina, but consumed a greater diversity of prey when abundance was low during the 1997-1998 El Nino. Lactating females also modified the length of time they spent at sea and onshore in response to changes in the abundance of prey. Low abundances and quality of prey during El Nino caused females to continue foraging for longer periods. This resulted in greater times between nursing episodes, and ultimately led to the starvation of pups. In contrast, during the 1999-2001 La Nina, females spent shorter times at sea and longer times onshore as prey became more abundant. Pup production and mortality were also directly related to the abundance of prey. Pup mortality reached 100% during El Nino and remained high during the favourable conditions of La Nina due to the short-term effects that El Nino had on population dynamics and subsequent maternal behaviour. Abortions were also higher during El Nino compared to other years, while numbers of pups bom were significantly reduced during and after El Nino due to the death of a high proportion of the adult population. My study suggests that South American sea lions in Peru are highly vulnerable to extreme changes in prey abundance. It also demonstrates that sea lion reproductive parameters, maternal attendance, and diet are strongly related to marine environmental conditions and annual variations in prey abundance. As a result, South American sea lions appear to be good ii indicators of stochastic changes in the distribution and abundance of marine resources in the Peruvian upwelling ecosystem. Table of contents Abstract ii Table of contents iv List of Tables :'. vi List of Figures vii Acknowledgements viii Chapter 1: General Introduction 1 Chapter 2: The Effects of Prey Abundance on Pup Mortality and the Timing of Birth ...4 Introduction 4 Materials and methods .". 6 Study site 6 Births and mortality 7 Prey abundance 8 Results • 8 Numbers ofpups born and timing of births 8 El Nino 11 Pup mortality : .'. .;•:......- 15 Aberrant behaviour 17 Discussion 18 Pup mortality 18 Pup production 19 Birth synchrony 21 Prey abundance 22 Timing of births and pup mortality 23 Summary 23 Chapter 3: Diet and Maternal Attendance as Indicators of Prey Abundance and Changes in the Marine Environment 25 Introduction 25 Materials and methods 27 Study site 27 iv Diet 27 Maternal attendance 28 Prey abundance and oceanographic conditions 29 Results .' 30 Diet 30 Maternal attendance 31 Arrivals 37 Departures 37 Numbers of females onshore 39 Prey abundance, oceanographic conditions and maternal attendance patterns 40 Discussion 42 Diet 42 Maternal attendance patterns and prey abundance 44 Perinatal Period. 44 Time at sea 45 Time onshore 48 Arrivals and departures 49 Numbers of females onshore 50 Conclusions 50 Summary 51 Literature cited . 52 v List of Tables Table 2.1 Characteristics of the breeding season from 1997-2002 at the Ballestas Islands 9 Table 3.1 Frequencies of occurrences of the principal prey during the breeding season from 1998 to 2002 at San Gallan Island 31 Table 3.2 Mean duration of foraging trips and visits onshore, and mean number of females onshore during the 1998 to 2002 breeding seasons 34 Table 3.3 Mean duration of the perinatal period, first foraging trip and first visit onshore during the 1998-2002 breeding seasons 35 Table 3.4 Relationship between an index of prey abundance and the duration of the maternal attendance cycle .V •. 41 vi List of Figures Fig. 2.1 The study site "La Maternidad" located at the North Ballestas Island, Peru 6 Fig. 2.2 Numbers of pups born and the estimated biomass of prey during the 1997- 2002 breeding seasons 9 Fig. 2.3 Relationship between anchovy biomass and annual death rates of pups, and the numbers of pups born per females counted onshore from 1997-2002 11 Fig. 2.4 Distributions of daily numbers of pups that were born or had died during the breeding season from 1997 to 2002 12 Fig. 2.5 The relationship between median dates of birth (1997-2002) and the abundance of prey in the previous year (1996-2001). , 14 Fig. 2.6 Distribution of daily abortions, premature births and full term births during the 1997-1998 El Nino 14 Fig. 2.7 Running rate of pup mortality through the breeding season from December to March (1997 to 2002) 16 Fig. 3.1 Mean time that females spent at sea and onshore during the breeding season 1998-2002 contrasted with an index of prey abundance 32 Fig. 3.2 Mean duration of female trips at sea and visits onshore over the breeding seasons from 1998 to 2002 36 Fig. 3.3 Timing of arrivals and departures of lactating females observed during day light hours from 1998 to 2002 at the Ballestas Islands 38 Fig. 3.4 Daily number of females onshore during the 1997-98 El Nino contrasted with the 2000 La Nina 40 Fig. 3.5 Relationship between the time that females spent at sea and onshore during the breeding seasons (1998-2002) and sea surface temperature anomaly and an index of prey abundance 41 Fig. 3.6 Distribution and abundance of anchovy around the Ballestas Islands 43 vn Acknowledgements I am grateful to the Peruvian Marine Research Institute (EVIARPE) who provided the financial and logistical support for my entire project. Also thanks to Idea Wild for providing logistical support. I am very grateful to my friend Milena Arias-Schreiber from UVIARPE, who trusted me and gave me the opportunity to lead this project. Special thanks to Miguel Niquen, Erick Diaz, Martin Trujillo and Mariano Gutierrez for providing me with unpublished data on the abundance and distribution of anchovy as well as data from anchovy catches from commercial fisheries from EVIARPE. Also thanks to Milena and Jose Carlos Marquez for the data on diet that I included in my thesis and did not always collect. Thanks also to Renato Guevara for helping me through financial support from LMARPE for the last breeding season at the islands. Thanks a lot to my field assistants during this long journey. Thanks Roberto, Ximena, Cynthia, Fernando, Angel, Francesca, Judith, Marcelo, Akemi, Gaby, Martin, Karina, Giss, Christian, Susana, Milagros, Jean, Angie and Doris. Without you guys, it would not have been possible to collect all the data. Thanks for your patience and friendship! We had a great time! I would like to specially thank my friends Roberto Manrique "Peto" and Julian Uribe "Pele", for being there when I needed you the most. Thanks Roberto for give me the strength to continue at the islands, and thanks Pele for your friendship, help and interest for ensuring our well being at the islands. Thanks to my supervisor Andrew Trites for introducing me to the world of SPLUS. Now I see the light you were talking about! Thanks Andrew for your guidance and for helping me to improve my writing. Thanks also to the members of my Committee, David Rosen and Anthony Sinclair for the helpful comments on my thesis. Thanks to my friends at the Marine Mammal Research Unit. Thanks to those nice people who made me feel not far away from home. Thanks especially to my friend Ruth Joy, for helping me so much in the fight with SPLUS to analyze data and get the graphics done. viii Thanks also to the amazing people of Room 10, Laura, Michelle and Saeko. Girls you made my time fun at UBC! Thanks a lot Laurita for your time correcting my English, especially my "grammar", and also for correcting my second chapter over and over again. Thanks to Arliss, Dom and Mary-Anne for the genial suggestions you gave me that improved my manuscript. Thanks Pamela for the times you made my life easy at the office. I would like to specially thank my family, my parents Maria, Antonio, and my brother Bell, for giving me the encouragement to leave home for something different and better. Thanks a lot for the incredible effort in helping me to cover my studies here. Thanks also for your love and for being always by my side. Thanks for caring for my Josefina!!! Finally, I would like to thank my Martin, for all his help and support. Just for being the way you are. Thank you!! ; , • . ix Chapter 1: General Introduction The South American sea lion (Otaria flavescens) is distributed along the coasts of South America, from northern Peru along the Pacific coast to southern Brazil in the Atlantic coast (Vaz-Ferreira 1982). They breed during the austral summer and have a breeding season that appears to change in length with latitude (Campagna 1985, Soto 1999). Female sea lions give birth to a single pup onshore, and will nurse for approximately one week before initiating regular attendance cycles (foraging trips and subsequent visits onshore to nurse their pups; Campagna & Le Boeuf 1988). Little is known about the timing of weaning in South American sea lions or when dependent young may supplement their milk diet with fish. However, known individuals of up to three years of age have been observed nursing on a regular basis in Peru (Soto 1999). The Peruvian upwelling ecosystem is one of the richest ecosystems of the world. However, it is subject to drastic and stochastic fluctuations of productivity, which in turn affect the abundance of principal prey resources sought by pelagic predators. These fluctuations in productivity are caused by one of the strongest natural disturbances of marine and terrestrial ecosystems, the El Nino Southern Oscillation. An El Nino event results in warm sea surface temperatures, decreased primary production, and a deepening of the thermocline. This reduces the upwelling and concentration of nutrients, and thereby alters the abundance, distribution and availability of prey to pelagic predators (Barber & Chavez 1986, Fiedler 2002, Trillmich 1993). The other major oceanographic event in Peru is La Nina (known as the El Nino cold phase), and is characterized by an increase in primary marine productivity, a shallow thermocline and cool sea surface temperatures (Fiedler 2002) resulting in favourable conditions for predators feeding on pelagic resources. Such diametrically opposed oceanic events as El Nino and La Nina presumably mean' that apex predators that inhabit the Peruvian coast have incurred strong selective pressures to thrive in the face of contrasting episodes of prey abundance. Provisioning of young is energetically the most expensive component of reproduction and parental care in mammals (Clutton-Brock 1991). Female sea lions provide all of the nutritional resources required by their young. Thus, parental care is dependant on the 1 foraging success of the female, which ultimately determines offspring survival and future reproductive success (Bonner 1984). Lactating females need to acquire enough energy to recover from the perinatal fast and also require energy to travel, forage and produce milk to nourish their pups (Costa et al: 1989), Since lactating females have limited body reserves, they rely on the abundance of prey close to the breeding rookery to produce enough milk to care for their pups and ensure their survival. Therefore, reduced prey abundance may result in females being unable to acquire enough energy to meet their metabolic needs, and may prevent them from adequately provisioning their young, thereby affecting pup survival. Reproductive and behavioural parameters of sea lions are probably a proxy for the state of the physical environment and a measure of the overall health of the ecosystem. In several otariid species, changes in maternal attendance behaviours have been associated with fluctuations in the abundance of prey resources (e.g., Costa et al. 1989, DeLong & Antonelis 1991, Heath et al. 1991, Majluf 1991, Ono et al. 1987, Trillmich & Dellinger 1991). The numbers of pups born and rates of pup survival have also been linked to fluctuations in prey abundance (e.g., DeLong & Antonelis 1991, Francis & Heath 1991, Guinet et al. 1994, Lunn & Boyd 1993a, b, Trillmich 1985, Trillmich & Limberger 1985, Trillmich et al. 1991). However, the strength of the relationship between the abundance of prey and changes in the reproductive ecology of otariids due to drastic and stochastic fluctuations in the marine environment has not yet been investigated. The goal of my thesis was to determine the effects of stochastic environmental changes on the reproductive success of the South American sea lions in Peru. I also wanted to assess whether their diet and maternal attendance patterns may indicate major changes in the abundance and distribution of prey. By examining the relationship between environmental change and reproductive parameters of South American sea lions, I sought to gain insights into the adaptative responses of females to cope with prey uncertainty and ensure pup survival. 2 Overview My thesis has two principal chapters. The first (Chapter 2) explores how changes in the abundance of prey due to major environmental disturbances can affect the timing of births as well as current and future reproductive success of South American sea lions in Peru. The second (Chapter 3) considers whether some parameters of the reproductive ecology of South American sea lions (such as diet and maternal attendance) may reflect changes in the abundance and distribution of prey in the unpredictable Peruvian marine ecosystem. Chapters 2 and 3 were written as independent manuscripts, and therefore contain some redundancies since both relate different behavioural measurements of South American sea lions to the abundance and distribution of prey. 3 Chapter 2: The Effects of Prey Abundance on Pup Mortality and the Timing of Birth Introduction Changes in the abundance or availability of prey may shape breeding, maternal patterns and survival, and compromise the future reproductive success and population dynamics of apex predators. Such changes have been noted among such species as Humboldt penguins (Spheniscus humboldti; Paredes and Zavalaga 1998), southern fulmars (Fulmarus glacialoides; Jenouvier et al. 2003), Galapagos fur seals (Arctocephalus galapagoensis; Trillmich & Ono 1991), water rats (Nectomys squamipes; Bergallo & Magnusson 1999), wild boars (Sus scrofa; Massei et al. 1996), and fat dormouse (Myoxus glis; Bieber 1998). Among pinnipeds, the effects of prey shortages have been noted to increase the length of foraging trips taken by lactating females (Boyd et al. 1994, Costa et al. 1989, Georges et al. 2000a, Georges et al. 2000b, Heath' et al. 1991, Lunn & Boyd 1993b, McCafferty et al. 1998, Trillmich & Ono 1991). They have also been seen to affect the timing of implantation (Boyd 1984, Boyd 1996), decrease the growth rate of pups (Lunn et al. 1994, Trillmich et al. 1991) and increase pup mortality (DeLong & Antonelis 1991, Francis & Heath 1991, Le Boeuf & Reiter 1991, Lunn & Boyd 1993b, Majluf 1991, Roux 1997, Trillmich & Dellinger 1991). Other effects of prey shortage include reducing pup production, delaying the timing of births, and extending the length of breeding cycles (Boyd 1996, Guinet et al. 1998, Lunn & Boyd 1993a, Ono et al. 1987, Pomeroy & Duck 2000, Trillmich 1985, Trites & Donnelly 2003). The Peruvian upwelling system is one of the most unpredictable but productive marine ecosystems in the world. As a consequence, pinnipeds and sea birds inhabiting the south eastern Pacific coast have to endure interannual fluctuations of their food supply (Trillmich & Ono 1991). One of the largest predators in this ecosystem is the South American sea lion (Otaria flavescens). How stochastic changes in the marine ecosystem affect the reproductive ecology of this species is unknown. From 1997-2002 I investigated how changes in prey abundance affected the reproductive success of South American sea lions. I was particularly interested in determining how this apex predator copes with the highly unpredictable Peruvian marine ecosystem. My study encompassed one major El Nino and one La Nina event. An El Nino event results in a deep 4 thermocline, with warm sea surface temperatures, less upwelling and a decrease in primary marine productivity (Barber & Chavez 1986, Fiedler 2002, Trillmich 1993). This causes a severe shortage of anchovy (Engraulis ringes) and other prey for apex predators, such as sea lions. During an El Nino, anchovy, the most important prey of South American sea lions (Arias-Schreiber 2003) tend to be of poor quality, in low numbers and at deeper depths than normal (Alamo & Espinoza 1998, Arntz et al. 1991, Barber & Chavez 1986, Santander & Zuzunaga 1984, Trillmich 1993). In contrast, a shallow thermocline, cool sea surface temperature and high marine productivity characterize a La Nina event as occurred from 1998-2001 (Fiedler 2002). South American sea lions breed during the austral summer starting in mid-December. Duration of the breeding season is affected by latitude (Campagna 1985, Soto 1999), with longer seasons occurring at low latitudes and shorter seasons occurring further south at high latitudes (Soto 1999). Pregnant females give birth to a single pup and remain onshore to nurse for approximately seven days (the perinatal period) before copulating and undertaking regular feeding trips (Campagna & Le Boeuf 1988). Lactation can last up to three years (Soto 1999), but little is known about the timing of weaning or when dependent young may supplement their milk diet with fish. Lactating females have limited body reserves and are therefore dependent on the abundance of prey near the rookery to produce the milk needed to care for their pups. Reduced prey abundance may result in a mother only acquiring enough energy to meet her metabolic needs, and may leave her unable to adequately provision her pup. Thus, a shortage of prey can directly affect pup survival. The following examines whether fluctuations of prey influence pup survival and the timing of births of South American sea lions. Data were obtained from 1997-2002 at the Ballestas Islands, Peru (Fig. 2.1) and span one El Nino and one La Nina event. They provide insights into the female adaptative responses when faced with prey shortages and document the behavioural changes associated with prey abundance that may affect pup survival. These behavioural observations are also the first to systematically describe the effects of the El Nino on future reproductive success. 5 Fig. 2.1 The study site "La Maternidad" located at the North Ballestas Island, Peru. Materials and methods Study site The Ballestas Islands (13°44'S 76°24'W) are located 250 km south of Lima (Fig. 2.1). Data were collected for 27 consecutive months from January 1997 to March 1999, and from 04 January to 10 March for three additional years (2000 - 2002). In all, the study encompassed six consecutive breeding seasons. The Ballestas Islands constitute a representative group of reproductive colonies of South American sea lions and is one of the largest breeding concentrations in Peru (Arias-Schreiber & Rivas 1998). These Islands are protected by the Peruvian Government as a guano bird reserve and are an important tourist 6 attraction, where sea lions are the primary draw. The main study area, La Maternidad or "The Maternity", is a narrow stretch of open pebble beach without pools or shaded areas. It is bounded to the north by open ocean, to the south by a rocky cliff, and by small caves to the east and west. I divided La Maternidad into 5 zones to facilitate the daily mapping of births and deaths (denotated as Cave, CI, C2, C3 and C4). Conspicuous rocks marked the boundaries between zones. Births and mortality The number of pups born each year and the mean number of females onshore (estimated from daily counts at 06:00 hrs.) were determined by two observers during day light hours (06:00 to 18:00 hrs). Parturitions occurring at night were estimated by counting fresh placentas each morning. The estimated number of pups born each day equalled the pups born during daylight hours plus the total number of placentas seen early in the morning. The number of pups that died each day was estimated from counts of fresh carcasses. Each dead pup was noted on a map to avoid double counts. Four categories of pup mortality were distinguished: male harassment, female harassment, stillborn and unknown reasons. Unknown reasons were designated when a pup was found dead and could not be placed in any of the other categories. The low density of sea lions and the open flat topography of the study beach, as well as the low number of scavengers and small tides meant that few if any carcasses (or placentas) were missed. The median dates of pupping and pup mortality were computed as the point during the breeding season when 50% of the total estimated births or deaths had occurred respectively. This was assumed to be a better measure of central tendency than the mean because of the large differences in the start date of the breeding season among years. 7 Prey abundance Data on abundance of prey (anchovy) was obtained from the Peruvian Marine Research Institute (LMARPE) and included anchovy biomass and monthly catches from commercial fisheries along the Peruvian coast. Anchovy biomass was estimated from hydroacoustic surveys conducted once each year (between January and April) to determine the abundance of pelagic resources along the Peruvian coast. Anchovy catches represent where the fishery was concentrated and may not have reflected all of the places where anchovy were available to sea lions. Catches were also affected by partial or total closed seasons enacted by the Peruvian Government to protect spawning anchovy. This typically occurred twice per year (normally between February - March and August - September). Duration of the closed seasons differed between years and may therefore have affected the annual amount of anchovy caught for interannual comparisons. " Anchovy catch data were standardized using the catch per unit effort index (CPUE) because total catch figures were influenced by many variable factors (e.g., number of holds, trips with catch, searching time, storage capacity in the ship, etc.). CPUE was estimated by month by IMARPE and was used to compare with pup production and mortality to test for possible relationships with prey abundance. Results Numbers ofpups born and timing of births The number of pups born at La Maternidad fell from 1249 in 1997 to only 50 during the 1997-98 El Nino (Fig. 2.2, Table 2.1). All 50 of these pups died. Pup production in subsequent years increased from 406 in 1999 to 799 in 2002. Biomass of anchovy, like pup production, dropped during the 1997-98 El Nino, and increased over the next four years (Fig. 2). Birth rates (expressed as the ratio of pups born to the mean number of females counted on shore) were positively correlated with prey abundance (anchovy biomass) during the breeding season (Pearson r = 0.868, p - 0.025, n - 6; Fig. 2.3). In non El Nino years, distributions of 8 Fig. 2.2 Numbers of pups born (dashed line) and the estimated biomass of prey (anchovy; solid line) during the 1997- 2002 breeding seasons. Shaded regions denote the periods of El Nino and La Nina. Anchovy biomass is measured in millions of tons (MT). Table 2.1 Characteristics of the breeding season from 1997-2002 at the Ballestas Islands showing the percent of pups that died, numbers of pups born, numbers of abortions, the mean number of females counted onshore, and median dates of birth and death. Years Pup mortality Pups born Abortions Females onshore Median birth date Median death rate 1997 13.3 1229 no data 885 31-Jan 09-Feb 1998 El Nino 100 50 -1000 72 22-Jan 10-Feb 1999 La Nina 26.4 406 16 548 06-Feb 02-Feb 2000 La Nina 32.6 561 no data 470 22-Jan 22-Jan 2001 La Nina 11.4 729 no data 458 23-Jan 20-Jan 2002 15.1 799 >100 390 29-Jan 30-Jan 9 births varied between years (KruskalrWallis^ =65.24,/? < 0.001; Fig. 2.4)., Although pupping started at different dates each year, there was general synchrony in the timing and frequency of the total number of pups born (Fig. 2.4). Numbers of pups born were correlated with the mean numbers of females counted on the beach each year (Pearson r = 0.860,/? = 0.028,« = 6). As with pups, mean number of females onshore varied significantly during the study period (Kruskal-Wallis Xl = 172.5, p < 0.001) and decreased by 90% during the 1997-98 El Nino, and by 30% from 1999 to 2002 (Table 2.1). The median birth date differed among years by as much as 16 days between the earliest peak of pupping (22 January 2000) and the latest (06 February 1999) (Median test, xl - 262.9, p < 0.001; Fig. 2.4). Individually recognizable females from previous breeding seasons (n = 8 females) varied their pupping dates - giving birth later in the season in 1999 and earlier in 2000 and 2001. Their timing of birth was consistent with the overall pattern observed from other females. In all years the peak of pup mortality occurred around the peak of pupping (measured as median dates, Fig. 2.4). A strong negative relationship was found between the median birth date and the assumed abundance of prey during the previous year (as measured by the amount of anchovy caught per unit of commercial fishing effort CPUE; Fig. 2.5). Median birth dates tended to be later following a year of low catches, and were earlier when catches had been high (Fig.2.5). Similar significant negative relationships were found between timing of birth and other indexes of CPUE such as anchovy catch per total hours of trip (Pearson r = -0.959,/? = 0.003,n = 6), anchovy caught per trip (r = -0.934,/? = 0.006), catch per register of brute tonnage estimated from total hours of trips (r = -0.918,/? = 0.01), catch per total hold (r = -0.972,p = 0.001) and catch per hold with fish (r = -0.979,/? = 0.001). Although the numbers of pups born per unit of area within the five zones of the La Maternidad beach were equally distributed each year, individual females appeared to have a 10 Fig. 2.3 Relationship between anchovy biomass (MT) and annual death rates of pups (top panel), and the numbers of pups born per females counted onshore (an index of birth rate) from 1997-2002 (bottom panel). preference for where they gave birth. The Cave was the first zone where females gave birth each year. It differed from the other zones by having shade and lacking intraspecific attacks by subadult males. El Nino The 1997-1998 El Nino started in May 1997, after the breeding season ended. Abortions were first noted in late July 1997, and continued progressively until late October. Foetuses seen after October were classified as premature births based on their size and their different 11 D. CD Fig. 2.4 Distributions of daily numbers of pups that were born or had died during the breeding season from 1997 to 2002. Data were fit with loess curves. The vertical lines indicate the median dates of births (solid) and deaths (dashed). 12 pelage colour. The frequency of premature births increased dramatically in early November at a rate of approximately 25 per day (Julian Uribe, pers. comm.) and peaked in mid November (Fig. 2.6). In all, 175 abortions and 440 premature births were observed in 1998 during the non-breeding season. Assuming 25 premature births occurred on each of the 15 days we were not present, the total number of reproductive failures numbered about 1000. High numbers of premature pups occurred at other rookeries in Peru during the 1997-1998 El Nino, but no systematic data of the daily rate are available. During the 1997-1998 El Nino breeding season (January - March), only 50 pups were born (-5% of the expected total pup production) (Figs. 2.4 & 2.6). Most of these pups were abandoned on the beach by their mothers, who were visibly emaciated. In contrast to the El Nino, only 16 premature births were noted during La Nina (April-November 1998). A second El Nino of moderate intensity occurred in 2002-2003 after we had completed our intensive observations, and again resulted in an increase in abortions (at a rate of 3 per day between September and November 2002). No data were recorded before or after these months. Sexually mature females in 2002 appeared to be older, in better condition, and were probably more experienced than those seen at the peak of the first El Nino (inferred by their large size). There was no evidence from other rookeries that massive abortions occurred during the 2002-2003 El Nino as seen during the 1997-1998 El Nino. Unusual maternal behaviours were observed during the 1997-1998 El Nino when a few females (n=5) abandoned the beach as they were giving birth (the pup was emerging from the mother). Parturition was probably completed at sea where the pup would have drowned. A number of other females arrived on the beach (n=28) with newborns in their mouths. Of these 28 pups, 39% were alive (n=ll), 46% were dead (n=13), and 7% were undetermined (n=2). The umbilical cord and remains of placenta were seen in all cases. Two other females arrived at the beach without a pup, but expelled their placenta minutes later. A few other females took their pups to sea after giving birth (n=7 live pups, and n=l dead pup). 13 100 350 600 850 1100 Anchovy catch per unit effort Fig. 2.5 The relationship between median dates of birth (1997-2002) and the abundance of prey in the previous year (1996-2001). Prey abundance was assumed to be proportional to the amount of anchovy caught per unit of commercial fishing effort. Data were fit with a linear regression (r= - 0.97, p < 0.01). 30 20 10 prematures abortions \ « - •• y x v . * • . t. pups born 1 15 1 15 1 15 1 15 1 15 1 15 1 15 1 15 1 15 Jul Aug Sept Oct Nov Dec Jan Feb Mar Fig. 2.6 Distribution of daily abortions, premature births and full term births during the 1997-1998 El Nino. 14 During the 1998 El Nino breeding season, yearlings were observed trying to take milk from unrelated females (n=17). Only six were successful (35%) and three of the 17 (18%) were injured by the females. All individuals (yearlings and adult females) showed signs of severe undernourishment. Pups were also observed trying to take milk (n=39), but only 39% of them were successful. Pups were rejected frequently and 5% were injured through their persistence. Pups in groups of up to four individuals were observed following and trying to take milk insistently from mothers arriving on the beach. Although not quantified, pups were also seen trying to take milk during the 2002-2003 El Nino breeding season. However, the frequencies appeared to be much lower than during the 1997-98 El Nino. Such behaviour was not observed in other years. Pups appeared to be undernourished during the 2002-2003 El Nino breeding season as a result of infrequent nursing associated with longer maternal feeding trips (Chapter 3.). Pups often pursued their mothers to the shoreline as they prepared to leave on a foraging trip, and tried unsuccessfully to continue suckling. During the 1997-1998 El Nino, 100% mortality of yearlings and juveniles (individuals of approximately two years old) occurred. No individuals of these cohorts were observed at the end of this El Nino at the study site or other important rookeries. During the 1999 breeding season, only adults were on the beach. Overall, approximately 60% adult mortality was estimated from monthly counts before, during and after the El Nino. Although carcasses were uncommon at the study beach during the non-El Nino years, emaciated carcasses of 73 yearlings, 41 females, and 35 juveniles were counted during El Nino. Yearlings were the first cohort affected by the El Nino food shortage beginning two months after its onset. Pup mortality The number of dead pups differed significantly among years and ranged from as few as 13% before El Nino to 100% during El Nino (Chi-square test xl = 159.8,/? < 0.001; Fig. 2.7, Table 1). Pup mortality after El Nino also differed significantly among years (Chi-square test 15 Fig. 2.7 Running rate of pup mortality through the breeding season from December to March (1997 to 2002). The rate of mortality for any given date equaled the total number of dead pups (since the first day of observation) divided by the total number of pups born over this time period. 16 xl = 17.6, p < 0.01). There was a strong negative correlation between the death rate of pups and prey abundance as measured by catch per unit of effort. These indices of abundance included tons of anchovy caught per trip (Pearson r = -0.82, p = 0.04, n = 6), catch per total hours of trip (r = -0.87, p = 0.02), catch per total hold (r = -0.86, p = 0.02), catch per hold with fish (r = -0.84, p = 0.03) and catch per brute tonnage register estimated from total hours of trip (r = -0.89,p- 0.01). There was also negative correlation between death rate and anchovy biomass during the breeding season (Pearson r = -0.80, p = 0.058, n = 6; Fig. 2.3). Higher rates of mortality were associated with lower estimates of anchovy biomass. Pup mortality was consistently high throughout the breeding season in all years (Fig. 2.7), but the distribution of deaths differed among years (Kruskal-Wallis xl = 22.57, p < 0.001; Fig. 2.4). Pup mortality was independent of the number of pups born and of the number of females on the beach. The only significant correlation between births and deaths occurred during the El Nino (Pearson r = 0.93, p < 0.001, n = 6; Fig 2.7). Pup starvation was evident during the El Nino breeding season when only 1% of the total pups born were still alive at the end of the season. One month later, only one pup was observed alive on the entire beach. The median dates of death differed by as much as 22 days between the earliest peak of pup mortality (20 January 2000) and the latest (10 February 1998) (Median test xl - 50.5, p < 0.001; Fig. 2.4). Numbers of pups were evenly distributed (per unit of area) over the five study zones in all years. Although the 2000 La Nina breeding season could be considered as optimal and favourable for the reproductive success of the South American sea lions (based on the high abundance of anchovy reflected in the landings and stocks assessments), pup mortality was at its highest compared with other non-El Nino years. Aberrant behaviour Pup mortality was caused by several factors that varied in importance among years. Most noteworthy was the aberrant behaviour observed particularly during the 1999-breeding season (a year after the peak of El Nino). Females in groups ranging from 4 to 6 animals were seen 17 taking and killing unrelated newborn pups. This behaviour accounted for 2.5 to 8.4% of total pup mortality. In most cases, females killed a pup by physical trauma while trying to keep it aggressively away from other females. Sub-adult males sometimes took pups to the sea from a group of disputing females. Aberrant maternal behaviour was first observed after the 1998 El Nino breeding season and continued during the rest of the study years in a lower frequency. Discussion An important finding of my study is that the reproductive success of South American sea lions in Peru is directly linked to fluctuations of prey availability. The 1997-1998 El Nino was the strongest on record and resulted in severe short-term effects on their reproductive ecology. My study suggests that South American sea lions face highly contrasting episodes of prey abundance — yet can evoke strategies that enable them to adapt to stochastic changes in their ecosystem and recover under favourable environmental conditions. Pup mortality Interannual changes in the Peruvian upwelling system can significantly affect the distribution and abundance of pelagic resources and the reproductive performance of species such as the South American sea lions that depend upon them. The strongest 1997-98 El Nino on record caused 100% of the few pups that were born that year to starve. Females returned infrequently to the beach to nurse their pups during the period of acute prey reduction (Chapter 3), and probably did not have sufficient milk to adequately nourish their pups. The acute prey shortage resulted in high adult mortality (approximately 60%) and the loss of entire young cohorts (ages zero, one and two years old). High number of carcasses recorded during annual censuses along the Peruvian coast confirms high mortality rates (Arias-Schreiber, unpublish.). Stillbirths were observed in low numbers and accounted for an average of 2.3% of the total mortality in all years. In contrast to El Nino, high abundance of prey should have been available to sea lions during the 1999-2001 La Nina. However, the sea lions did not appear to have been able to 18 fully capitalize on this change in events due to the lagged effects of the acute prey shortages experienced the previous year (1997-1998 El Nino). High pup mortality recorded during the 1999-2001 La Nina breeding seasons seem to have been an indirect consequence of three different but related factors associated with short-term effects of the 1997-98 El Nino: 1) high mortality of territorial males, 2) high incidence of intraspecific aggressions and 3) aberrant female behaviour. Only 60% of the territorial males that attempted to breed before the El Nino was alive to breed after this event (Soto 1999). The high mortality of bulls allowed for increased intraspecific aggressions by subadult males on pups. The low numbers of territorial males present after the El Nino was insufficient to defend the relatively large breeding beach (and indirectly protect the pups). As a consequence, injuries resulting from intraspecific aggressions by the higher proportion of subadult males that appeared on the beach during the 2000 La Nina breeding season caused up to 50% of the total pup mortality. Aberrant female behaviour, whereby females took and killed pups that were not their own, also contributed to the increased pup mortality. This aberrant behaviour seemed to be related to a low pregnancy rate and low numbers of females pupping successfully after the El Nino; and appeared to abate in subsequent years as more females successfully gave birth. Pup production My data showed that the ratio of pups born to the number of females present each year declined as the anchovy abundance dropped during the breeding season. This ratio is effectively an index of birth rates. Most mature females return to the rookery even if they do not give birth. Thus, it appears that birth rates dropped as the abundance of prey declined. Increases in pup production have been noted in some otariids following a year of high food availability (see Lunn & Boyd 1993a). Indeed, during La Nina, we observed a doubling in the number of pups born (from 1999 to 2002). Although environmental conditions were not equal during all La Nina years, there was increased pupping during 2000 and 2001 when food abundance was higher than other years. 19 It is important for animals to detect environmental signals that permit them to coordinate or possibly adapt the physiological events of reproduction with changes in their food supply (Boyd 1991). The 1997-1998 El Nino started soon after the end of the breeding season when females would have had high energetic demands placed on them by their suckling young. Poor body condition probably increases the frequency with which South American sea lion abort their foetuses during years when nutritional stress becomes acute. Although the cost of producing a full term foetus is likely to be small (see Winship et al. 2002), the cost of caring for a pup is not (Clutton-Brock 1991). Thus a female would have to continue to nurse a pup for one or more years to ensure its survival and future reproductive success. High incidence of abortions when food is scarce has also been reported in species such as the South African fur seal, Arctocephalus pussillus (Roux 1997) and the California sea lion, Zalophus californianus (Trillmich et al. 1991). Abortions tended to be associated with females in significantly poorer body condition compared to females that maintained pregnancies to parturition (Guinet et al. 1998). High mortality of females during the 1997-1998 El Nino -combined with the low pregnancy rates probably explain the low number of births that followed the El Nino year. Although pinnipeds are capable of ovulating and conceiving when prey is restricted (Lunn & Boyd 1993a, b, Trites & Donnelly 2003), it seems improbable that high numbers of females whose blastocysts successfully implanted, failed to carry foetuses to parturition under the favourable environmental conditions that followed the 1998 El Nino breeding season. Scarce numbers of abortions observed in my study area and other important rookeries preceding this event support this hypothesis. The reduced number of copulations and the abandonment of breeding areas I observed during the peak of the El Nino may be additional factors accounting for the low recruitment we noted after the El Nino breeding season. Duration of the breeding cycle (from oestrus to parturition) varies according to the food availability (Boyd 1996) and may reflect the timing of implantation (Lunn & Boyd, 1993a) or changes in the fetal growth rate (Guinet et al. 1998). Although birth synchrony is maintained and high precision in the peak of pupping from year to year is observed in several species (Boulva 1975, Campagna 1985, Coulson 1981, Duck 1990, Jemison & Kelly 2001, Majluf 20 1991, Pitcher et al. 2001, Trites 1992), the South American sea lions appear to have considerable flexibility in the start and duration of their breeding season according to the abundance of prey. Thus, females tended to give birth later in the season after a year of low prey abundance, and earlier when prey abundance was high. Birth synchrony The peak of pup mortality occurred around the peak of pupping in all years. This differs from other South American sea lion rookeries on the Atlantic coast (see Campagna et al. 1992), and with other pinnipeds such as the Weddell seal (Thomas & DeMaster 1983). Food availability, suitable weather, and environmental cues acting together or independently may be the primary factors that determine the timing of births and enhance the reproductive success of females (Boness et al. 1995, Boyd 1991, Ims 1990, Lunn & Boyd 1993a, Majluf 1992, Pitcher et al. 2001). Additionally, females should give birth when the probability of offspring mortality is the lowest in order to maximize their fitness (Clutton-Brock 1988, Ritcher 1999, Trites & Antonelis 1994). However, the distribution of pupping dates for all breeding female South American sea lions within a season tended to be consistent across all years, despite differences in risk of losing their pups. In general, older females tend to be more experienced and in better condition than younger females. A number of pinniped studies have documented that such females also tend to give birth earlier in the season compared with young mothers - thereby increasing the survival of their pups (Boyd & McCann 1989, Duck 1990, Reiter et al. 1981, York & Scheffer 1997). Although there is.no evidence to indicate-whether most dead pups belonged to young or inexperienced sea lion females before the 1997-98 El Nino, there were few young females left alive after this event and pup mortality remained at its highest around the peak of pupping. Although a bias towards older females giving birth earlier in the season was expected after the El Nino, the distribution of observed births did not change. Distribution of births during the breeding season may be independent of the age of females in South American sea lions. 21 The closeness of the median dates of births and deaths (Fig. 4) may be related to a concomitant rise in copulations that lead to higher incidences of male-male interactions (Soto, unpublish.). Although most pups died, from unknown causes, intraspecific attacks were the main documented cause of direct pup mortality at the peak of births. These intraspecific attacks resulted from peripheral males that succeeded in evading the territorial males and entering the beach. Male interactions may also have contributed to mother-pup separations as have been observed elsewhere (Campagna et al. 1992, Pistorius et al. 2001). The degree of polygyny appeared higher in Peru as indicated by the mean ratio of males to females (1:30) compared to the ratio (1:4) at rookeries on the Atlantic coast (see Campagna, 1985). My study site also appears to have had a shortage of suitable territories that may have affected male behaviour and lead to intense male competition and a high incidence of intraspecific attacks on pups. An absence of pools or shady areas (necessary to reduce thermal stress due to intense solar radiation) favours territories along the water edge and restricts the number of territories. This in turn increases intraspecific competition, and favours a reproductive system where few males defend shoreline territories and have access to most copulations. This suggests that having more territories with thermoregulatory advantages besides the shoreline, would have lessened male competition and would have lowered pup mortality. Although further studies are necessary to test this hypothesis, topographies that are suitable for reproduction and enhance thermoregulation are scarce in Peru. Thus, sea lions tend to be concentrated in large numbers on relatively few rookeries that have limited access by humans. Prey abundance I found a strong relationship between reproductive success (current and future) of the South American sea lion and the abundance of prey in the Peruvian upwelling ecosystem. The impact that fluctuations in the abundance of prey may have on the sea lion reproductive ecology will depend on the intensity and timing of such changes relative to the timing of the breeding cycle. Prey shortage during El Nino events may result in higher numbers of 22 abortions and reproductive failures in subsequent years. In contrast, high abundance of prey during La Nina can favour reproductive success and an increase in pup production. Timing of births and pup mortality The timing of births and pup mortality were strongly related to the interannual fluctuations in the abundance of prey. Births occurred later in the season after years of low prey abundance and earlier in the season after a year of high prey abundance. Pup mortality reached its highest when prey scarcity was acute because mothers were unable to adequately supply energy resources to their pups. Moreover, high numbers of abortions were observed during years when prey abundance appeared to be low. This may have been a physiological response of pregnant females that negated future energetic costs under adverse conditions. In contrast, when resources were abundant, the numbers of pups born rapidly increased. However, rates of pup mortality remained high during the years of high prey abundance as an indirect consequence of altered population dynamics and maternal behaviour caused by the previous severe food shortages that sea lions experienced during El Nino. Global warming models predict stronger and more frequent future El Ninos events along the Peruvian coast. This does not bode well for South American sea lions if the lagged aberrant behaviours associated with high adult mortality further impact their reproductive success. As such, the direct and indirect effects of the drastic and stochastic changes in the upwelling ecosystem on births and pup mortality may endanger the South American sea lions in Peru. Summary Pup mortality and the timing of birth of South American sea lions were investigated to determine the possible relationship between fluctuations in prey abundance in the Peruvian upwelling ecosystem and current and future reproductive success during six consecutive breeding seasons. My study from 1997 to 2002 encompassed the strongest El Nino on record and one La Nina event. Pup mortality ranged from 13% before El Nino to 100% during El 23 Nino, and was negatively correlated with prey abundance. Abortions were also more frequent when prey abundance was low. However, pup mortality remained high following El Nino due to its short-term effects on population dynamics and subsequent maternal behaviour. Births occurred later in the season after years of low prey abundance and earlier following years of high prey abundance. The peak of pupping occurred around the peak of mortality in all years, and may have been the product of intensive competition between bulls at the peak of the breeding. Stronger and more frequent El Ninos appears to be occurring along the Peruvian coast and could produce significant stochastic changes in births and pup mortality that place the vulnerable South American sea lion population in Peru at great risk. 24 Chapter 3: Diet and Maternal Attendance as Indicators of Prey Abundance and Changes in the Marine Environment Introduction In theory, changes in ocean productivity directly affect the abundance and distribution of prey species, which in turn affect the reproductive and foraging ecology of their predators (Bester & Van Jaarstveld 1997, Boyd & Murray 2001, Georges et al. 2000a, Guinet et al. 2001, McCann 1987, Trillmich & Ono 1991). A number of studies have related changes in the foraging behaviour of such apex marine predators as pinnipeds, whales and sea birds to the abundance of prey they feed upon (e.g., Boyd 1999, Boyd et al. 1994, Boyd et al. 2002, Costa et al. 1989, Georges et al. 2000a, Lunn et al. 1994, McCafferty et al. 1998). However, only a few studies have attempted to validate the relationship between prey and foraging behaviour by assessing oceanographic conditions or the distribution and abundance of prey due to the difficulty and costs of conducting such research (e.g., Boehlert et al. 2001, Boyd & Arnbom 1991, Boyd & Murray 2001, Campagna et al. 2000, Hakoyama et al. 1994, Hooker & Boyd 2003, Lea & Dubroca 2003, Lydersen et al. 2003, McCafferty et al. 1999, Wilson et al. 1994). The degree to which behavioural responses of apex predators are proxies for changes in the marine environment is not clear. A better understanding of this relationship would help guide the use of apex predators as indicators of marine environmental change. It might also help to assess the consequences of fishing on the foraging ecology of apex predators and the potential for competition to occur with humans for marine resources. Female South American sea lions offer a unique opportunity to evaluate the strength of the relationship between fluctuations in the marine ecosystem and the reproductive ecology of apex predators. These sea lions inhabit the productive but unpredictable Peruvian upwelling ecosystem (Ryther 1969), and are exposed to interannual and stochastic fluctuations in the distribution and abundance of their principal prey — the anchovy (Arias-Schreiber 2003). The distribution and abundance of anchovy, as well as the effects of reduced ocean productivity on its life history have been well documented due to the economic importance of 25 anchovy to the Peruvian fishing industry (Barber & Chavez 1986, Muck et al. 1987, Niquen & Bouchon 2002, Pauly & Palomares 1987, Pauly & Tsukayama 1987, Senocak et al. 1987). Similarly, considerable knowledge has been gathered on the foraging, reproduction and diet for the sea lions throughout their distributional range along the Atlantic and Pacific coasts of South America (Campagna et al. 1992, Campagna & Le Boeuf 1988, Campagna et al. 2001, Koen Alonso et al. 2000, Soto 1999,. Thompson et al. 1998, Werner & Campagna 1995). Females are easy to monitor since they give birth on land and alternate intermittent trips at sea to feed (foraging trips) with visits onshore to nurse their pups during the austral summer (Campagna & Le Boeuf 1988). Female sea lions also provide all the nutritional resources needed to rear their pups, and do not rely on fat reserves stored before parturition (Trillmich et al. 1991). Thus, stochastic fluctuations of the marine environment should directly affect the maternal performance and future reproductive success in this species. The goal of my study was to determine whether behavioural measurements from apex predators reflect physical changes in the marine environment and alterations in the abundance and distribution of prey. Sea lion diets and behavioural observations of maternal attendance patterns were assessed at the Ballestas Islands, Peru, over 5 consecutive breeding seasons of contrasting prey abundance that included the strongest El Nino on record (1997-1998, McPhaden 1999) and a moderate La Nina (1999-2001). I investigated the reproductive ecology (maternal attendance and dietary habits of South American sea lions) and its relationship with oceanographic variables (sea temperature anomalies) and marine prey resources (distribution and abundance of anchovy) at matching spatial and temporal scales. The strength of these relationships were expected to reveal the extent to which South American sea lions could be used as indicators of abundance and distribution of prey in the unpredictable Peruvian marine ecosystem. My results also provided valuable information about the effects of prey fluctuations on the reproductive ecology and behaviour of sea lions for better understanding the ecology of other marine mammals and their adaptative responses to cope with marine environmental change. 26 Materials and methods Study site Data were collected on the Ballestas Islands (13°44'S 76°24'W), Pisco, Peru, which are home to one of the largest breeding concentrations of South American sea lions in Peru (Arias-Schreiber & Rivas 1998). The islands are protected by the Peruvian Government as a guano bird reserve. The main study area, La Maternidad ("The Maternity"), is an open pebble beach without pools or shaded areas that allows for all of the sea lions using the rookery to be observed. It is bounded to the north by open ocean, to the south by a rocky cliff, and by small caves to the east and west. Diet Diet was determined from scats collected by EvIARPE at San Gallan Island (13°51' S, 76°26' W), a nearby breeding rookery that was approximately 7 km from my study site. The topography of my behavioural study site at La Maternidad prevented access to the beach to collect scats. Numbers of scats collected at San Gallan Island varied between years (n = 139 scats in January and March 1998; 81 scats in January and March 1999; 50 in April 2000; 21 in January 2001 and 84 samples in January and February 2002). Hard parts that could be identified to species were recovered in all scats, except 60 in 1998 (El Nino), 8 in 1999 (La Nina) and 10 in 2002 that contained no identifiable hard parts. Scats were frozen after collection, then later thawed and washed using a series of sieves with mesh sizes of 2.0, 0.425 and 0.25 mm (U.S.A. Standard Test, Tyler). Fish otoliths, cephalopod beaks and crustacean remains were separated. Otoliths were identified to the lowest taxonomic level using the reference collections of the Peruvian Marine Research Institute (EVIARPE) and the identification guide of Garcia-Godos (2001). Cephalopod beaks were not identified to species and crustacean remains did not accurately reflect the number of individuals in each sample due to their advanced stage of digestion. Thus, only the frequency of occurrence (defined as the number of scats in which a given prey occurred, Dellinger & Trillmich 1988) for each taxonomic group was included in the analysis. 27 Maternal attendance Observations were made during 5 consecutive breeding seasons (Jan 4-Mar 10) from 1998 to 2002. Maternal attendance was estimated by measuring the duration of individual female attendance cycles (time spent at sea foraging and time spent onshore nursing their pups). Females were marked with paint pellets fired from a C O 2 gun (Nel-Spot Company), and with paint-filled-eggs thrown up to 35 m from the animals. Females were marked on their day of parturition or during the interval between parturition and a mother's first departure to sea (the perinatal period) (1998: n = 45; 1999: n = 80; 2000: n = 100; 2001: n = 144; 2002: n = 103). The small sample size during the 1998 El Nino breeding season resulted from the low numbers of females using the study beach, given most females with pups were marked that season. Paint markings and natural scarring were drawn on animal identification cards. Marks lasted from two to three weeks. Some females were remarked depending on the quality of the initial mark. Female attendance was recorded using focal sampling methodology (at 10 minute intervals) during day light hours (06:00 to 18:00 hrs) from a cliff that was about 70 meters above the colony and 100 meters distant. The duration of the perinatal period, foraging trips and visits onshore were recorded for each marked female. Females that lost their pups were removed from the sample. If a female was recorded at the beach at the end of the day and was not found the next morning, her time of departure was assumed to have occurred at midnight (the midpoint between when observations stopped and began again). Females whose date of pupping were unknown and had remained onshore with their pups for more than 3 days after being marked were assumed to have not made their first post-partum trip to sea. First feeding trips and first visits onshore were analyzed independently from the rest of the data because the maternal investment is known to be high during the perinatal period. Sample sizes represented the number of females rather than the numbers of trips and visits onshore. Thus, only mean durations of trips and visits onshore for each female were included in the analysis. Daily counts of females were conducted at 06:00, 12:00 and 18:00 hrs each year. 28 Prey abundance and oceanographic conditions Interannual oceanographic and biological fluctuations occurred during my study from 1998 to 2002. The strong 1997-1998 El Nino, resulted in increased sea surface temperatures and the deepening of the thermocline, which drastically reduced the primary marine productivity (Fiedler 2002, McPhaden 1999). This in turn affected the distribution and abundance of anchovy and other potential prey for the sea lions. In contrast, the moderate 1999-2001 La Nina was characterized by cold surface temperatures, shallow thermoclines and high marine productivity (Fiedler 2002), which in turn resulted in favourable prey conditions for the sea lions. The 2002 breeding season took place during the early stages of the moderate 2002 El Nino (Moron 2002, Niquen & Bouchon 2002). Data on prey abundance and the distribution of pelagic resources were obtained from IMARPE. These included CPUE (catch per unit of effort) indexes from catches of anchovy from commercial fisheries and, abundance and distribution of Peruvian anchovy derived from hydroacoustic surveys along the Peruvian coast. Hydroacoustic surveys encompassed the area between shore and 100-120 nm and were carried out using the echo sounding SLMRAD EK-500 and EY-500. Abundance of anchovy from hydroacoustic surveys was estimated using echo integration techniques (see MacLennan & Simmonds 1992). CPUE is often used as a measurement of relative abundance for anchovy (Bouchon et al. 2002, Niquen & Bouchon 2002, Santojanni et al. 2003), and was compared with the duration of the foraging trips and visits onshore to test for possible relationships. CPUE indexes included the catch of anchovy per hours of trip, per hold with fish, per registered brute tonnage, and per registered brute tonnage estimated from hours of trip. According to IMARPE (M. Niquen unpublish.) the anchovy catch per hours of trip has showed the strongest correlation with the assessments of anchovy biomass from hydroacoustic surveys. . , Oceanographic conditions were assessed from sea surface temperatures recorded daily by the IMARPE, at the Coastal Laboratory of Pisco. Temperature anomalies were calculated from data collected since 1976 by IMARPE. Since sea surface temperatures can be used as an index of environmental uncertainty (Gentry & Kooyman 1986) and have been associated with changes in the abundance of the Peruvian anchovy (Barber & Chavez 1986) and maternal 29 performance in otariids (Francis et al. 1998, Lea 2002), I analyzed these oceanographic features to test for possible relationships with the maternal attendance pattern. The occurrence of principal prey species in the diet and the potential link to their distribution and abundance close to the breeding area was also explored. Results Diet A total of 1719 otoliths were found in 307 scats. All but 1.8% of the otoliths (n=31) could be identified. The main prey items consumed were fish (76.7% frequency of occurrence), crustaceans (88.3%) and cephalopods (17.9%; Table 3.1), with the highest numbers of prey species occurring in 2002 (17 spp.) and during the 1997-98 El Nino (12 spp.). An average of 8 species were found in the scats during La Nina (1999 to 2001). Significant differences were found in the occurrences of principal prey among and within years. Only anchovy (Engraulis ringes) and squid (Loligo sp.) occurred in all years. However, consumption of anchovy differed among years (logistic regression deviance test — with year as a categorical variable: xl = 32.86, p < 0.001) as did consumption of squid (Xl= 23.20, p< 0.001). No differences were found in the relative occurrences of anchovy within a season in any year. Occurrences of anchovy represented an average of 75% of the total fish identified in the scats (excluding the El Nino year when anchovy was scarce and other fish species were consumed). Including the El Nino year resulted in anchovy making up 68% of all fish occurrences. The only year that squid varied within a season was 1999 (logistic regression deviance test: xl = 4-73, p = 0.03). The pelagic squat lobster (Pleuroncodes monodon) was found in all years except during the 1997-1998 El Nino and in February 2002. Its occurrence differed between years (logistic regression deviance test — with year as a categorical variable: Xl =36.61, p < 0.001), but only differed within a season in 2002 when no remains were found in the scats collected in February of that year (early stages of the 2002 El Nino, see Table 3.1). 30 Table 3.1 Frequencies of occurrences (%) of the principal prey of South American sea lions during the breeding season from 1998 to 2002 at San Gallan Island. Mean values excluded the 1998 El Nino. (*) Occurrences of crustaceans in 2002 correspond to percentages for January 2002 only since no crustacean remains were found in the scats during February 2002. Years Fish (%) Crustaceans (%) Cephalopods (%) 1998 El Nino 97.5 0.0 5.1 1999 La Nina 41.1 83.6 27.4 2000 La Nina 86.0 96.0 6.0 2001 La Nina 100.0 76.2 28.6 2002 79.7 97.5* 9.5 1999-2002 Mean 76.7 88.3 17.9 Important demersal species for commercial fisheries occurred in most years. These included the lumptail searobin (Prionotus stephanophrys, mean = 21.6%), Peruvian hake (Merluccius gayi, 8.2%) and camotillo (Normanichthys crockeri, 29.2%). However, differences in their occurrences were found among years (lumptail searobin, Logistic regression deviance test xl = 57.43, p = 0.027; Peruvian hake, xl = 8-82, p - 0.032; and camotillo xl =9.09,p = 0.028). Only during the 1997-1998 El Nino did the demersal cachema (Cynoscion analis) occur in high percentages (22.2%). Maternal attendance The duration of foraging trips (Kruskal-Wallis, xl = 133.1, p < 0.001) and visits onshore (Kruskal-Wallis, ^ = 18.7, p < 0.001) differed significantly among years, even after removing the El Nino year (trips: xl =121.97, p < 0.001; visits: xl = 8-87, p = 0.031; Fig. 3.1; Table 3.2.). Pairwise comparisons showed that the foraging trips recorded during the 31 El Nino La Nina 1998 1999 2000 2001 2002 Breeding season Fig. 3.1 Mean time that females spent at sea and onshore during the breeding season 1998-2002 contrasted with an index of prey abundance (anchovy catches, CPUE). 1998 El Nino did not differ significantly from those in 2001 (Dunnett, p = 0.93) and 2002 (Dunnett, p = 0.60). The lack of difference may reflect the small sample size or the high variability in foraging times during the 1998 El Nino. Duration of visits onshore was similar between all years except between 2000 and 2002 (Dunnett, p = 0.031). Mean duration of foraging trips and visits onshore were inversely related, with the longest foraging trips and shortest visits onshore occurring during El Nino. Similarly, duration of foraging trips were the shortest during the 2000 La Nina, while the visits onshore were the longest (Fig. 3.1). During El Nino, foraging trips were twice the mean duration of trips in subsequent years (Table 3.2). 32 Duration of the first trip and visit onshore after parturition were longer than all other trips and visits in all years (Table 3.3) and also varied significantly among years (trips: Kruskal-Wallis xl = 91-75, p < 0.001, visits: Kruskal-Wallis, xl = 17.35, p = 0.002). The perinatal period averaged 7.46 days over the 5 years of study (Table 3.3), and varied significantly among years (Kruskal-Wallis xl =38.61, p < 0.001), with the shortest period occurring during the 1998 El Nino and the longest perinatal period during the 2000 La Nina. Between year comparisons revealed significant differences in the duration of the perinatal period between 2000 and 2001 (Dunnett,/? = 0.025), and 2000 and 2002 (Dunnett,/? = 0.049). During La Nina, the perinatal period was about 2 days longer than during El Nino. Of the 42 females marked during the 1997-98 El Nino breeding season, 22% (n = 10) did not return to the beach after the first feeding trip, and another 22% abandoned their pups after the second feeding trip. Removing females that lost their pups from the samples in all years (see Methods) resulted in numbers of recorded foraging trips being lower during El Nino compared to other years (Fig. 3.2) since most pups died before the end of the breeding season (Chapter 2). During El Nino, no trend was observed in the mean time that females spent at sea, until the 5 th trip, when mean time at sea began decreasing with each subsequent trip (Fig. 3.2). In 1999, the first season after El Nino, no trend was detected for the mean time spent at sea. However, mean times at sea increased over the breeding season in 2000 and 2002, and decreased in 2001 (Fig. 3.2). No trends were found during El Nino in the time spent onshore. However, for all other years (1999 to 2002) the mean time onshore decreased, as the pups grew older (Fig. 3.2). The proportion of time that females spent at sea per attendance cycle differed among years (Kruskal-Wallis, xl =132.05, p < 0.001), and was the greatest during the 1998 El Nino (71%) and the lowest during La Nina (53%). Differences were also noted among years for the proportion of time females spent at sea during,the first trip after parturition (Kruskal-Wallis, Xl =102.25, p< 0.001). 33 Table 3.2 Mean duration ± SE (in days) of foraging trips and visits onshore, and mean number of females onshore during the 1998 to 2002 breeding seasons. Sample sizes in parentheses indicate the number of counts and the number of females per trip and per visit each breeding season included in the analysis. Breeding Season Mean trip at sea Mean visit onshore Mean number females 1998 El Nino 2.37 ±0.23 (22) - 0.68 ± 0.04 (22) 72.11 ±6.15 (54) range 0.45-4.41 0.43 - 1.21 13 - 154 1999 La Nina 1.40 ±0.04 (92) 0.88 ± 0.03 (88) 548.04 ± 18.64 (63) range 0.49-2.88 0.28- 1.62 287 - 1113 2000 La Nina 1.12 ±0.03 (122) 0.96 ±0.02 (119) 470.15 ±7.22 (60) range 0.45 - 2.08 0.54- 1.66 311-596 2001 La Nina 1.66 ± 0.07 (117) 0.89 ±0.08 (112) 457.90 ± 7.43 (43) range 0.36-3.36 0.34- 1.62 321-536 2002 1.92 ±0.06 (107) 0.87 ±0.02 (103) 390.19 ± 12.01 (47) range 0.57 - 3.60 0.36-1.64 205 - 519 1998-2002 Mean 1.56 ±0.07 0.90 ± 0.04 470.06 ± 14.56 There was considerable variability in the duration of foraging trips between females within a year. Variability in time spent at sea was highest during the 1998 El Nino (CV = 45.15%) and lowest during the 2000 La Nina (25.89%). Such variability was statistically consistent among years (CV = 32.3% for all years combined). However, the lack of statistical significance may be related to a small sample size during the 1998 El Nino and the lower number of female foraging trips recorded that year due to pup mortality (see Fig. 3.2). Variability observed in the duration of visits onshore was similar among years (26.94%). 34 Table 3.3 Mean duration ± SE (in days) of the perinatal period, first foraging trip and first visit onshore during the 1998-2002 breeding seasons at the Ballestas Islands. Sample sizes are in parentheses and indicate the number of females included in the analysis each breeding season. Breeding season Perinatal period First trip First visit 1998 El Nino 5.72 ± 0.36 (26) 2.98 ± 0.29 (32) 1.01 ± 0.11 (32) range 1.26-7.24 0.83-8.29 0.27 + 2.43 1999 La Nina 7.54 ±0.10 (114) 1.56 ± 0.07 (113) 1.25 ±0.05 (101) range 3.65 - 12.00 0.30-4.00 0.32-3.37 2000 La Nina 7.69 ±0.10 (104) 1.32 ±0.05 (143) 1.23 ±0.04 (128) range 4.43 - 10.34 0.33-3.72 0.30-2.66 2001 La Nina 7.23 ±0.11 (80) 1.69 ±0.06 (117) 1.11 ±0.04(113) range 5.00- 10.00 0.28 - 3.74 0.30-3.00 2002 7.21 ±0.14(65) 2.21 ±0.98 (104) 1.03 ±0.04 (96) range 4.75 - 10.71 0.49 - 5.74 0.28-2.45 1998-2002 Mean 7.46 ±0.39 1.66 ±0.08 1.16±0.06 The proportion of time that females spent at sea was significantly correlated with the duration of attendance cycles each year (1998 El Nino: Pearson r = 0.742, p < 0.001, n = 22; 1999: r = 0.372, p < 0.001, n = 91; 2001: r = 0.548, p < 0.001, n = 111; 2002: r = 0.492, p < 0.001, n = 103) except in 2000 when the foraging trips were the shortest (r = 0.097, p = 0.292, n = 119). Females tended to spend a greater proportion of their time at sea in years when they made longer foraging cycles. No relationship was detected between the time spent at sea and the preceding or subsequent time spent onshore each year. 35 T R I P S A T S E A V I S I T S O N S H O R E 1998 El Nino j . 1 78 75 69 62 64 51 45 38 28 21 15 1 101 OJ 68 78 77 73 ( 1999 La Nifla v *• T MO 125 119 109 105 98 90 BO 75 67 61 57 53 44 38 33 24 IB 17 ' 125 123 115 107 102 94 H 79 71 66 60 55 M 42 34 28 20 1 2000 La Nina -o c ro a) 2001 La Nifla I 60 54 44 34 28 25 22 1 1 if ,1 1 4 7 10 13 16 19 22 1 4 7 10 13 16 19 22 Trip number Visit number Fig. 3.2 Mean duration ± SE (days) of female trips at sea and visits onshore over the breeding seasons from 1998 to 2002. Sample sizes (number of individual females) are shown at the top of each panel. Dashed lines represent the grand mean of trips and visits respectively excluding the El Nino year. Trip numbers show the trips that females made after parturition from their first to their 23 rd trip, while visits indicate the time spent onshore. 36 Arrivals Time of arrival onshore varied significantly among years (%u = 112.9,/? < 0.001), but did not differ significantly among the La Nina years (1999-2001). The time of arrival during La Nina was not random during the hours of observation (06:00 to 18:00 hrs) (pooled data: xl =310.7,/? < 0.001; Fig. 3.3). In general, most females arrived after midday in all years except during the 1998 El Nino, when they arrived more frequently before midday. Arrivals during La Nina occurred mostly at night and dawn rather than during the day (2000: Xl =29.43, p< 0.001, 57.14%; 2001: xl = 4.15,/? = 0.042, 53.14%) except in 1999 and 2002 when no differences were found. In contrast, most females arrived during the daytime during the 1998 El Nino rather than at night and dawn (xl = 13.26,/? < 0.001). Departures Timing of departures from land varied significantly among years (x]2 = 148.2,/? < 0.001). Significant differences were also found during La Nina years (xl = 19.55,/? = 0.003). In general, most females departed in the morning in all years except during El Nino when most females departed at the end of the day (Fig 3.3). Most departures occurred at night and dawn than during the day light from 1998 to 2002 (mean ± SE: 72.2 ± 3.7 % and 27.7 ± 3.7 % respectively). Departures outnumbered arrivals at night and dawn in all years (72 ± 3.7 % and 48 ± 3.5 % respectively). 37 D E P A R T U R E S ARRIVALS Fig. 3.3 Timing of arrivals and departures of lactating females observed during day light hours from 1998 to 2002 at the Ballestas Islands. Percentages were calculated as number of arrival (or departures) each hour as a percent of total (day and night) arrivals (or departures). Data were smoothed with Loess curves (solid lines). Total numbers of documented arrivals and departures are indicated in each panel. 38 Numbers of females onshore Mean numbers of females that were onshore during the breeding season (based on counts at 06:00 hrs) varied significantly among years (Kxuskal-Wallis xl = 159.71,p< 0.001). Numbers fell from a mean of 885 females before the 1997-98 El Nino (Chapter 2) to 72 during the El Nino, and decreased again from 548 females after El Nino in 1999, to 390 in 2002 (Table 3.2). One major difference during El Nino (as opposed to other years such as the 2000 La Nina shown in Fig. 3.4) was the low numbers of females that were onshore during the breeding season due to drastic abandonment of the breeding beaches. Numbers of females onshore during non-El Nino years did not vary significantly before or after the breeding season (K. Soto pers. obs.). Mean numbers of females onshore during the day (based on counts at 06:00, 12:00 and 18:00 hrs) varied significantly each year (Kruskal-Wallis: 1999: xl = 5 6 - 4 0 > P < 0.001; 2000: Xl =114.16, p < 0.001; 2001: xl =48.31, p < 0.001; 2002: "xl =26.59, p < 0.001). In general, numbers tended to be high early in the morning, dropping to a minimum around midday, and increasing to a maximum towards the end of the day when most of the arrivals were recorded (Fig. 3.3). Each morning, most of the females and their pups moved towards the water's edge to occupy the shoreline, presumably to avoid overheating as soon as the sunlight fell on the beach (at approximately 07:30 hrs.). Other females abandoned the beach at this time, as noted by the high number of recorded departures. Females did not make such movements during cloudy days. Instead, the animals tended to increase the spacing between them. Although no significant differences were found in the number of females onshore throughout the day, during El Nino, the highest numbers of females were counted at midday with lower numbers registered at the end of the day, and minimum numbers in the morning. Differences between the mean numbers of females at the end of the day (18:00 hrs) and following morning (06:00 hrs) indicated that most females departed at night, which is consistent with my observations that the highest numbers of marked females departed over night. -39 2000 La Nina —i 1 1 1 1 1 1 1 15 1 , 15 1 15 1 Dec' Jan Feb Mar Fig. 3.4 Daily number of females onshore (based on counts at 06:00 hrs) during the 1997-98 El Nino contrasted with the 2000 La Nina (which was similar to other non El Nino years). Data during Dec 1999 were not collected, but would have been in the range of values collected from Jan - Mar 2000 based on observations in other years. Data were smoothed with Loess curves. Prey abundance, oceanographic conditions and maternal attendance patterns Females appeared to adjust their attendance cycles in response to changes in prey abundance (Fig. 3.1), as shown by the inverse relationship between the time spent at sea and the abundance of prey (measured as the amount of anchovy caught per unit of commercial fishing effort, CPUE, as is shown in Table 3.4). Conversely, females spent more time onshore as abundance of prey increased (measured using CPUE indexes shown in Table 3.4). Durations of foraging trips were positively correlated with the sea surface temperature anomalies (Pearson: r = 0.896, p = 0.039, n = 5; Fig. 5). hi contrast, durations of visits onshore were inversely correlated with the temperature anomalies (r = -0.877, p = 0.051, n = 5; Fig 3.5). Therefore, foraging trips were longer and visits onshore were shorter when anomalies were positive (such as during El Nino) and foraging trips were shorter and visits onshore were longer when anomalies were negative (as during La Nina). 40 Table 3.4 Relationship between an index of prey abundance (anchovy catches from commercial fisheries, CPUE) and the duration of the maternal attendance cycle (time at sea and time onshore). (* estimated from total hours of trip). Anchovy C P U E Time at sea Time onshore r P n r P n Catch per total hours o f trip -0.946 0.015 5 0.820 0.080 5 Catch per hold with fish -0.832 0.081 5 0.874 0.052 5 Catch per registered brute tonnage capacity -0.990 0.001 5 0.897 0.039 5 Catch per registered brute tonnage capacity* -0.955 0.011 5 0.815 0.093 5 • Fig. 3.5 Relationship between the time that females spent at sea and onshore during the breeding seasons (1998-2002) and sea surface temperature anomaly (left panels) and an index of prey abundance (anchovy catches, CPUE: Catch per total hours of trip; right panels). 41 Discussion There appears to be a strong relationship between diet and the maternal attendance pattern, and the prey that was available to South American sea lions in the Peruvian marine ecosystem. The types and proportions of prey consumed as well as the maternal attendance cycles changed according to year to year shifts in the abundance and distribution of fish and crustaceans. However, maternal flexibility was insufficient during the 1997-1998 El Nino conditions to compensate for acute prey shortages, resulting in females spending prolonged times at sea that exceeded the fasting abilities of their pups and caused high mortalities due to starvation. Diet South American sea lions fed mainly on fish, crustaceans and cephalopods, but the frequency of occurrence of each, differed between and within seasons. Anchovy and the pelagic squat lobster were the main prey consumed by sea lions during La Nina. However, their spatial distribution and concentration with respect to the breeding rookery varied in those years (despite being close to the surface, Fig. 3.6). Interestingly, occurrence of anchovy and squat lobster in the diets increased as prey became more abundant close to the rookery (IMARPE 1999, 2000, 2001, 2002; Fig. 3.6). Sea lions fed on a variety of prey when their principal prey were scarce as occurred during the 1997-98 El Nino and in 2002. High occurrences of demersal species in the diet during these years suggest that sea lions foraged at deeper depths than normal. The only time the pelagic squat lobster was not found in the diet was during the 1997-98 El Nino and in February 2002 (during the early stages of the moderate El Nino 2002). This was probably due to the preference of squat lobsters for colder waters (Paredes & Elliot 1997), and is consistent with the observation that squat lobsters were scarce near the breeding rookery based on hydroacoustic surveys conducted during warm episodes (see IMARPE 2002). 42 Fig. 3.6 Distribution and abundance of anchovy around the Ballestas Islands. Data from hydroacoustic surveys during the 1999 La Nina (left panel) and 2000 La Nina (right panel). Dashed lines indicate the limits of the surveys. Adapted from EVIARPE (1999, 2000). Sea lions appear to have foraged on their preferred prey according to its relative abundance in the marine environment. They showed strong prey preferences that were reflected in the dominant occurrences of these resources in their diets. Scarcity of resources in the marine ecosystem was also reflected by low occurrences in their diet as well. The apparent relationship between diet and the distribution and abundance of their principal prey (obtained from hydroacoustic surveys and commercial catches) suggests that the South American sea lion diet is a good indicator of the distribution and abundance of marine resources in the Peruvian marine ecosystem. 43 Maternal attendance patterns and prey abundance Perinatal Period. Duration of the perinatal period has been proposed to be directly linked to the availability of prey and the energy reserves acquired before parturition (Bowen 1991). In my study, the mean duration of the perinatal period during years when prey was not restricted did not differ from the results of other studies along the Atlantic coast of South America (Campagna & Le Boeuf 1988, Cappozzo et al. 1991, Werner & Campagna 1995), nor did it differ from the findings for other species of sea lions (Heath et al. 1991, Higgins et al. 1988, Higgins & Gass 1993, Milette & Trites 2003, Ono et al. 1987, Trillmich 1986). The short perinatal period observed during the El Nino implies that initial female body condition was significantly reduced compared to other years, and probably led to a faster depletion of scarce maternal fat reserves to critical -values. A reduced perinatal period might have significantly decreased the chances of pups surviving if they did not initially gain enough energy reserves to survive the initial fast while their mothers were foraging at sea. Similar reductions in perinatal attendance have been reported for other otariids due to changes in the abundance of prey (see Lunn et al. 1993, Ono et al. 1987). Gentry et al. (1986) proposed that depletion of fat stores regulates the time spent onshore after a foraging trip. Similarly, the length of the perinatal period is probably determined by the rate that female fat reserves are depleted to produce milk and adequately nourish their pups. The length of the perinatal period must also be limited by the maximum energy a female can store before parturition, which might be related to the size of the female and the abundance of key prey species. This limit is probably easily attained under favourable oceanographic conditions (such as during La Nina). Consequently, the length of the perinatal period might not increase significantly during years when prey is abundant due to inherent physical and physiological constraints placed on the female. 44 Time at sea. Females appear to have modified the duration of their foraging trips to compensate for changes in prey quality and abundance (Fig. 3.1). The deepening of the thermocline during the 1998 El Nino (by more than 100 m; Pizarro 1997, Sanchez et al. 2000) would have meant that females had to dive to greater depths than normal. Also, they may have had to travel longer distances from the rookery to follow the migrating anchovy to cooler areas of southern Peru (Niquen & Bouchon 2002). Females in my study responded to the acute prey shortage during the 1997-98 El Nino by increasing the length of their foraging trips. In contrast, females spent shorter times at sea during the favourable conditions of La Nina. Increased times at sea due to reductions of prey have also been reported for several otariid species (Boyd 1999, Costa et al. 1991, DeLong & Antonelis 1991, Heath et al. 1991, Limberger et al. 1983, Majluf 1991, McCafferty et al. 1998, Ono et al. 1987, Trillmich & Dellinger 1991, Trillmich et al. 1986). Although favourable oceanographic conditions (in terms of sea lion prey) were observed along the Peruvian coast during La Nina years, the spatial distribution and abundance of pelagic resources differed between years (see IMARPE 1999, 2000, 2001, 2002). These apparent differences in the accessibility of prey close to the rookery might explain the differences in foraging trip duration during La Nina, since shorter foraging trips corresponded to higher abundances of prey close to the breeding area (IMARPE 2000; Fig. 3.6), and longer foraging trips corresponded to a mid- to high-abundance of prey located further away from the breeding beaches (IMARPE 1999; 2001; Fig. 3.6). During the 1997-98 El Nino breeding season, no assessments of pelagic resources by hydroacoustic surveys were conducted by IMARPE. However, data from landings of commercial fisheries indicated that anchovy catches were minimal at that time because of the deepening of the thermocline (Niquen & Bouchon 2002). Interestingly, the longest foraging trips were also observed at that time. The poor body condition I observed during El Nino indicated that females were unable to gain enough resources to meet their own energy requirements. As a consequence, females were apparently unable to sustain normal rates of energy delivery to their pups or compensate 45 adequately by increasing foraging effort because the prey shortage was extreme. Females appear to have had no alternative but to abandon their pups or maintain lactation with milk of low quality and quantity after spending long periods at sea. Insufficiently replenished body reserves may have forced females to continue foraging for longer periods. Long trips did not appear to result in an increase in energy gained or milk transferred to pups. Emaciated females and severely undernourished pups confirmed the failure of foraging and the severity of this event. Although most females abandoned their pups, it did not ensure their own survival. High numbers of visibly emaciated females died during the peak of El Nino in February 1998 at several rookeries along the Peruvian coast (K. Soto per.obs.). Similarly, many pups were unable to successfully fast between the infrequent visits onshore, and subsequently starved to death (Chapter 2). One emaciated pup attempted to continuously take milk from other females, but died after waiting 9 days for his mother. She returned 2 days later and was also visibly emaciated. While prey shortages would have resulted in females increasing their foraging effort for a reduced rate of foraging success, the energy content of anchovy was also lower during El Nino (Alamo & Espinoza 1998). This would have resulted in reduced energy intake per unit of prey caught by the sea lions. The concentration of low quality and reduced prey abundance at deeper depths than normal would have also caused the sea lions to expend more energy per unit of prey consumed. Moreover, poor body condition may have reduced female diving abilities and further decreased foraging success. A longer first foraging trip after parturition seems to reflect the time needed for females to replenish the fat reserves they depleted during the perinatal period. However, when faced with prey restrictions, longer trips might also have reflected lower foraging success. Although other otariids are known to increase diving effort near their rookery when prey is limited (Feldkamp et al. 1991, Lea et al. 2002), it is unknown whether the longer foraging trips of South American sea lions noted during El Nino were due to increased foraging effort associated with swimming greater distances from the rookery and/or increasing the rate of diving closer to the rookery. Nevertheless, it is apparent that none of these foraging behaviours were sufficient to compensate for acute prey reduction. Long distances were 46 apparently traveled since a number of known females who abandoned their pups were sighted as far as 250 km south of the breeding rookery (K. Soto per.obs.). Boyd et al. (1991) considered two factors that might set the duration of female foraging trips. One was that females need to attain a minimum amount of energy before returning onshore to nurse their pups. The second was that the time spent away must not exceed the ability of the pups to fast, regardless of the energy gained. If the former factor occurs, females may return before or after a determined time limit depending on the availability of prey (Lunn et al. 1993). However, if a minimum of energy is gained before the time limit, females may continue foraging until they reach the set time limit or their maximum storage capacity (Lunn et al. 1993). The length of foraging trips for female South American sea lions seems to be mainly determined by their storage capacity than by a set time limit. During the 2000 La Nina, when prey abundance was the highest, females presumably reached their storage capacities faster than in other years based on the shortness of their trips. If trip durations were time limited, trips would have been expected to be of similar durations during good years such as the La Nina years. Instead, average trip durations correlated with the abundance and distribution of prey, such that time at sea was reduced as prey became more abundant (Figs. 3.5 and 3.6). The flexibility in the duration of maternal foraging trips observed in the South American sea lions seems to depend on a combination of environmental, physical and physiological constraints. Females presumably require a minimum time at sea to replenish their resources and/or satisfy a minimum gain of energy. Such a restriction is probably determined by physiological and physical constraints (Boyd et al. 1991, Thompson et al. 1998) as well as by prey abundance. Similarly, there must also be a maximum time spent at sea before pups starve, which must be a function of the fasting abilities of pups, the female storage capacity and the abundance of prey. However, extreme restrictions in the availability of prey (such as during the 1997-98 El Nino), presumably results in females extending the time they spend at sea to ensure their own survival, which must ultimately depend on their physiological foraging capabilities and their foraging success: "• 47 The duration of foraging trips recorded during years when prey was not restricted were similar to values obtained for the South American sea lions at the Falkland Islands (Thompson et al. 1998). However, South American sea lions from Argentina (Campagna & Le Boeuf 1988, Campagna et al. 2001, Cappozzo et al. 1991, Werner & Campagna 1995) spent as much as twice the time foraging as did the sea lions I observed. Since distance to the foraging grounds is an important component of trip duration (Boyd 1999, Merrick & Loughlin 1997), and time at sea increases with travel distance (Bonadonna et al. 2000, Boyd 1999), these long trip durations may indicate that longer distances are traveled or longer times are spent foraging in Argentina to satisfy prey energy requirements. Time onshore. The time that females spent onshore after foraging also varied according to the abundance of prey (Figs. 3.1 & 3.5). Shorter visits onshore occurred when prey abundance was lower as observed during the 1997-98 El Nino - and longer visits occurred when prey was not restricted (as during La Nina). In several otariid species, the time that females spent onshore has been less sensitive to prey reductions than the time spent at sea. Although shorter times onshore have been observed in several species during shortages of prey, they did not differ significantly from values obtained in normal years (see Ono et al. 1987, Heath et al. 1991, Majluf 1991). Low caloric prey might have prevented the females from producing milk of high quantity and quality. Consequently, milk reserves might have been depleted faster during El Nino, thereby forcing females to depart sooner than in other years. Pup body condition deteriorated continuously during the long fasting periods and reflected the reduced foraging success of their mothers. 48 Arrivals and departures. Although arrivals and departures occurred throughout the day, most females arrived during daylight and departed at night and dawn (Fig. 3.3). These patterns suggest that female South .American sea lions foraged mostly at night and dawn, when anchovy were closer to the surface and more accessible to the sea lions (Arntz et al. 1991). Such nocturnal foraging patterns have been reported for the sympatric South American fur seal (Trillmich et al. 1986), which also prey mainly on anchovy (Arias-Schreiber 2003). Higher proportions of departures in the morning and arrivals after midday (Fig. 3.3) have also been observed for other otariid species (see Trillmich et al. 1986), and probably relate to female behavioural adaptations to avoid thermal stress, since both patterns avoid overheating during the hot midday hours. Diurnal foragers avoid the costs of overheating, although diurnal foraging might demand higher energy investment than nocturnal foraging since anchovy are located at deeper depths during the day (Arntz et al. 1991). In all likelihood, diurnal foragers probably compensate for higher costs of foraging if prey are abundant and easily accessible. During El Nino, females arrived onshore before midday when higher air temperatures would have increased the risks of overheating on land. Similar pattern of arrivals and departures have been observed in the South American fur seal during El Nino 1982-83 (Trillmich et al. 1986). Apparently, reduced prey would have meant that the costs of foraging during the day might not have compensated for the cost of thermoregulation on land. Species that inhabit the tropics and subtropics experience high air temperatures during the breeding seasons and incur high costs since the reduction of thermal stress is associated with behaviours that compromise reproductive success (e.g., aggressiveness, Limberger et al. 1986, Majluf 1987, Trillmich et al. 1986, Chapter 2). In such circumstances, sea lions probably avoid the costs of thermoregulation through behavioural adaptations by arriving soon after midday and departing before midday in all years, except during El Nino when maternal behaviour was altered because of reduced prey (Fig. 3.3). 49 Numbers of females onshore. Numbers of females onshore decreased dramatically during the 1997-98 El Nino (Fig. 3.4). Females progressively abandoned the rookery throughout the El Nino starting in November 1997, and no animals were observed at the end of the breeding season in March 1998. High adult mortality and the loss of the entire young cohorts during the El Nino (Chapter 2) accounts for the low number of survivors that remained in April 1998. Although favourable oceanographic conditions were observed after El Nino, the recovery of the South American sea lion numbers might take many years to be noticed due to the time needed for the young females to reach the age of reproduction and successfully pup. However, since favourable oceanographic conditions and high survival of young cannot be guaranteed in subsequent years, a slow increase in the sea lion population might be expected due to the unpredictability of the Peruvian upwelling ecosystem and the loss of an important part of the reproductive population. If another El Nino of strong magnitude occurs, the recovery of the remaining population of sea lions could be seriously compromised since young cohorts would be the most affected. Conclusions My study demonstrates that South American sea lion maternal responses and diets reflected the interannual fluctuations of the Peruvian upwelling ecosystem and suggests that South American sea lions may be good indicators of relative changes in the distribution and abundance of marine resources. Although the dramatic changes in the marine ecosystem due to the occurrence of strong El Nino and moderate La Nina events are extreme compared to fluctuations in other marine ecosystems, the responses of apex predators such as the South American sea lions reveals important insights about the behavioural adaptations and the limitations of those adaptations that might have evolved to cope with uncertainty in prey distribution and abundance, and rapid changes in the ecosystem. The fact that behavioural measurements from apex predators may be considered good proxies of the abundance and distribution of prey as well as oceanographic features is promising and opens new windows for the implementation of research using marine mammals as data collectors. Thus, the integration of biological and oceanographic information has the 50 potential to provide a better understanding of the dynamics of the marine ecosystem and its influence in shaping the life history of apex predators. Summary The diet and maternal attendance pattern of South American sea lions in Peru were investigated during the 1998-2002 breeding seasons when drastic changes in marine productivity occurred as a consequence of the strong 1997-1998 El Nino and a moderate 1999-2001 La Nina. The goal of my study was to determine whether changes in the diet and maternal attendance pattern reflected physical changes in the marine environment and alterations in the abundance and distribution of prey. Behavioural observations conducted at the Ballestas Islands found maternal attendance patterns were strongly linked to the abundance of prey and oceanographic features close to the rookeries. Females responded to the acute prey shortage during the El Nino by increasing the length of their foraging trips and decreasing the time they spent onshore with their pups. In contrast, females spent shorter times at sea and longer times onshore during the favourable conditions of La Nina. Acute prey shortages appeared to cause high pup mortalities because females had to spend more time at sea searching for prey and did not return frequently enough to nurse their pups. The high abundance of anchovy and squat lobster near the breeding areas during La Nina (as indicated by independent fisheries assessments) suggest that sea lions did not have to travel very far to acquire their preferred prey. However, the lower occurrence of these preferred prey species during El Nino, resulted in sea lions increasing the number of species consumed, particularly demersal fishes. Females appear to have adjusted their maternal attendance patterns and diets according to how the abundance and distribution of their prey changed from year to year. Based on these observations, diet and maternal responses appear to reflect interannual fluctuations of the unpredictable Peruvian upwelling ecosystem, and suggest that the South American sea lions may be good indicators of relative changes in the distribution and abundance of marine resources. 51 Literature cited Alamo, A. and Espinoza, J. 1998. Variaciones alimentarias en Engraulis ringes y otros recursos pelagicos durante Invierno-Primavera de 1997. Inf. Inst. Mar Peru 130: 45-52. Arias-Schreiber, M . 2003. Prey spectrum and feeding behaviour of two sympatric pinnipeds (Arctocephalus australis and Otaria flavescens) in relation to the 1997-98 ENSO in southern Peru. Pages 59. M.Sc. Thesis. University of Bremen, Bremen. Arias-Schreiber, M . and Rivas, C. 1998. Distribution, tamano y estructura de las poblaciones de lobos marinos Arctocephalus australis y Otaria. byronia en el litoral peruano en Noviembre 1996 yMarzo 1997. Inf. Prog. Ins. Mar Peru 73: 17-32. Arntz, W., Pearcy, W. G. and Trillmich, F. 1991. Biological consequences of the 1982-83 El Nino in the Eastern Pacific. In: Pinnipeds and El Nino: responses to environmental stress. 22-42. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Barber, R. T. and Chavez, F. P. 1986. Ocean variability in relation to living resources during the 1982-83 El Nino. Nature: 279-285. Bergallo, H. G. and Magnusson, W. E. 1999. Effects of climate and food availability on four rodent species in southeastern Brazil. J. Mammal. 80: 472-486. Bester, M . N. and Van Jaarstveld, A. S. 1997. Growth in Subantarctic fur seal Arctocephalus tropicalis pups as a possible indicator of offshore food availability. In: Marine mammal research in the southern hemisphere. 88-91. Hindell, M . A. & Kemper, C. Pretoria, S. Africa: Surrey Beatty & Sons, Chipping Norton. Bieber, C. 1998. Population dynamics, sexual activity, and reproduction failure in the fat dormouse (Myoxus glis). J. Zool., London 244: 223-229. Boehlert, G. W., Costa, D. P., Crocker, D. E. , Green, P., O'Brian, T., Levitus, S. and Le Boeuf, B. J. 2001. Autonomous pinniped environmental samplers: using instrumented animals as oceanographic data collectors. J. Atmos. Oceanic Technol. 18: 1882-1893. 52 Bonadonna, F., Lea, M . A. and Guinet, C. 2000. Foraging routes of Antarctic fur seals (Arctocephalus gazella) investigated by the concurrent use of satellite tracking and time-depth recorders. Polar Biol. 23: 149-159... Boness, D. J., Bowen, D. and Iverson, S. I. 1995. Does male harassment of females contribute to reproductive synchronic in the grey seal by affecting maternal performance? Behav. Ecol. Sociobiol. 36: 1-10. Bonner, W. N. 1984. Lactation strategies for pinnipeds: problems for a marine mammal group. Symp. Zool. Soc. Lond. 51: 253-272. Bouchon, M . , Diaz, E. , Cahuin, S. and Ochoa, M . 2002. Effects of "El Nino" 1997-98 on the schoolong behaviour of the Peruvian Anchovy (Engraulis ringes). Investig. Mar. 30: 182-183. ISSN 0717-7178. Boulva, J. 1975. Temporal variations in birth period and characteristics of newborn harbour seals. Rapp. P.-v. Reun. Cons. int. Explor. Mer 169: 405-408. Bowen, W. D. 1991. Behavioural ecology of pinniped neonates. In: The behaviour of pinnipeds. 66-127. Renouf, D. London: Chapman and Hall. Boyd, I. L. 1984. The relationship between body condition and the timing of implantation in pregnant grey seals. J. Zool., Lond. 203: 113-123. Boyd, I. L. 1991. Environmental and physiological factors controlling the reproductive cycles of pinnipeds. Can. J. Zool. 69: 1135-1148. Boyd, I. L. 1996. Individual variation in the duration of pregnancy and birth date in Antarctic fur seals: the role of environment, age, and sex of fetus. J. Mammal. 77: 124-133. Boyd, I. L. 1999. Foraging and provisioning in Antarctic fur seals: interannual variability in time-energy budgets. Behav. Ecol. 10: 198-208. Boyd, I. L. and Arnbom, T. 1991. Diving behaviour in relation to water temperature in the southern elephant seal: foraging implications. Polar Biol. 11: 259-266. 53 Boyd, I. L. , Arnould, J. P. Y., Barton, T. and Croxall, J. P. 1994. Foraging behaviour of Antarctic fur seals during periods of contrasting prey abundance. J. Anim. Ecol. 63: 703-713. Boyd, I. L. , Lunn, N. J. and Barton, T. 1991. Time budgets and foraging characteristics of lactating Antarctic fur seals. J. Anim. Ecol. 60: 577-592. Boyd, I. L. and McCann, T. S. 1989. Pre-natal investment in reproduction by female Antarctic fur seals. Beh. Ecol. Socio. 24: 377-385. Boyd, I. L. and Murray, A. W. A. 2001. Monitoring a marine ecosystem using responses of upper trophic level predators. J. Anim. Ecol. 70: 747-760. Boyd, I. L. , Staniland, I. J. and Martin, A. R. 2002. Distribution of foraging by female Antarctic fur seals. Mar. Ecol. Prog. Ser. 242: 285-294. Campagna, C. 1985. The breeding cycle of the southern sea lion, Otaria byronia. Mar. Mamm. Sci. 1: 210-218. Campagna, C , Bisioli, C , Quintana, F., Perez, F. and Vila, A. 1992. Group breeding in sea lions: pups survive better in colonies. Anim. Behav. 43: 541-548. Campagna, C. and Le Boeuf, B. J. 1988. Reproductive behaviour of southern sea lions. Mar. Mamm. Sci. 104:233-261: " ' Campagna, C , Rivas, A. L. and Marin, M . R. 2000. Temperature and depth profiles recorded during dives of elephant seals reflect distinct ocean environments. J. Mar. Syst. 24: 299-312. Campagna, C , Werner, R., Karesh, W., Marin, M . R., Koontz, F., Cook, R. and Koontz, C. 2001. Movements and location at sea of South American sea lions (Otaria flavescens). J. Zool., London 257: 205-220. Cappozzo, H. L. , Campagna, C. and Monserrat, J. 1991. Sexual dimorphism in newborn Southern sea lions. Mar. Mamm. Sci. 7: 385-394. Clutton-Brock, T. H. 1988. Reproductive success: studies of individual variation in contrasting breeding systems. Chicago: University of Chicago Press. 54 Clutton-Brock, T. H. 1991. The evolution of parental care. Princeton: Princeton University Press. Costa, D. P., Antonelis, G. A. and DeLong, R. L. 1991. Effects of El Nino on the foraging energetics of the California sea lion. In: Pinnipeds and El Nino: responses to environmental stress. 156-165. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Costa, D. P., Croxall, J. P. and Duck, C. D. 1989. Foraging energetics of Antarctic fur seals in relation to changes in prey availability. Ecology 70: 596-606. Coulson, J. C. 1981. A study of the factors influencing the timing of breeding in the grey seal Halichoerus grypus. J. Zool., London 194: 553-571. Dellinger, T. and Trillmich, F. 1988. Estimating diet composition from scat analysis in otariid seals (Otariidae): is it reliable? Can. J. Zool. 66: 1865-1870. DeLong, R. L. and Antonelis, G. A. 1991. Impact of the 1982-1983 El Nino on the northern fur seal population at San Miguel Island, California. In: Pinnipeds and El Nino: responses to environmental stress. 75-83. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Duck, C. D. 1990. Annual variation in the timing of reproduction in Antarctic fur seals, Arctocephalus gazella, at Bird Island, South Georgia. J. Zool., London 222: 103-116. Feldkamp, S. D., DeLong, R. L. and Antonelis, G. A. 1991. Effects of El Nino 1983 on the foraging patterns of California sea lions (Zalophus californianus) near San Miguel Island, California. In: Pinnipeds and El Nino: responses to environmental stress. 146-155. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Fiedler, P. C. 2002. Environmental change in the eastern tropical Pacific Ocean: review of ENSO and decadal variability. Mar. Ecol. Prog. Ser. 244: 265-283. Francis, J., Boness, D. J. and Ochoa-Acuna, H. 1998. A protacted foraging and attendance cycle in female Juan Fernandez fur seals. Mar. Mamm. Sci. 14: 552-574. Francis, J. and Heath, C. 1991. Population abundance, pup mortality,' and copulation frequency in the California sea lion in relation to the 1983 El Nino on San Nicolas 55 Island. In: Pinnipeds and El Nino: responses to environmental stress. 119-128. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Garcia-Godos, I. 2001. Patrones morfologicos del otolito sagitta de algunos peces oseos del mar peruano. Boletin Inst. Mar Peru 20: 1-70. Gentry, R. L. and Kooyman, G. L. 1986. Fur seals: maternal strategies on land and at sea. Princeton, NJ: Princeton University Press. Georges, J.-Y., Bonaddona, F. and Guinet, C. 2000a. Foraging habitat and diving activity of lactating subantarctic fur seals in relation to sea surface temperatures at Amsterdam Island. Mar. Ecol. Prog. Ser. 196: 291-304. Georges, J.-Y., Tremblay, Y. and Guinet, C. 2000b. Seasonal diving behaviour in lactating subantarctic fur seals on Amsterdam Island. Polar Biol. 23: 59-69. Guinet, C., Dubroca, L. , Lea, M . A., Goldsworthy, S., Gherel, Y., Duhamel, G., Bonadonna, F. and Donnay, J. P. 2001. Spatial distribution of foraging in female Antarctic fur seals Arctocephalus gazella in relation to oceanographic variables: a scale-dependent approach using geographic information systems. Mar. Ecol. Prog. Ser. 219: 251-264. Guinet, C., Jouventin, P. and Georges, J.-Y. 1994. Long-term population changes of fur seals Arctocephalus gazella and Arctocephalus tropicalis on subantarctic (Crozet) and subtropical (St. Paul and Amsterdam) islands and their possible relationship to El Nino Southern Oscillation. Antarct. Sci. 6: 473-478. Guinet, C., Roux, J. P., Bonnet, M . and Mison, V. 1998. Effect of body size, body mass, and body condition on reproduction of female South African fur seals (Arctocephalus pusillus) in Namibia. Can. J. Zool. 76: 1418-1424. Hakoyama, FL, Le Boeuf, B. J., Naito, Y. and Sakamoto, W. 1994. Diving behaviour in relation to ambient water temperature in northern elephant seals. Can. J. Zool. 72: 643-651. Heath, C. B., Ono, K. A., Boness, D. J. and Francis, J. M . 1991. The influence of El Nino on female attendance patterns in the California sea lion. In: Pinnipeds and El Nino: 56 responses to environmental stress. 138-145. Trillmich, F. & Ono, K. New York: Springer-Verlag. Higgins, L. V., Costa, D. P., Huntley, A. C. and Le Boeuf, B. J. 1988. Behavioural and physiological measurements of maternal investment in the Steller sea lion, Eumetopias jubatus. Mar. Mamra. Sci. 4: 44-58. ' Higgins, L. V. and Gass, L. 1993. Birth to weaning: parturition, duration of lactation, and attendance cycles of Australian sea lions Neophoca cinerea. Can. J. Zool. 71: 2047-2055. Hooker, S. K. and Boyd, I. L. 2003. Salinity sensors on seals: use of marine predators to carry CTD data loggers. Deep-Sea Res. Part I 50: 927-939. IMARPE. 1999. Crucero 9902-03 de Evaluation Hidroacustica de Recursos Pelagicos. Informe Ejecutivo-Direccion Cientifica. Inst. Mar Peru, Lima. IMARPE. 2000. Crucero 0001-03 de Evaluation Hidroacustica de Recursos Pelagicos. Segundo Informe de Campo. Inst. Mar Peru, Lima. IMARPE. 2001. Crucero 0102-04 de Evaluation Hidroacustica de Recursos Pelagicos. Informe Ejecutivo-Direccion Cientifica. Inst. Mar Peru, Lima. IMARPE. 2002. Crucero 0202-03 de Evaluation Hidroacustica de Recursos Pelagicos. Resumen Ejecutivo. Inst. Mar Peru, Lima. Ims, R. A. 1990. The ecology and evolution of reproductive synchrony. Trends Ecol. Evol. 5: 135-140. Jemison, L. A. and Kelly, B. P. 2001. Pupping phenology and demography of harbour seals (Phoca vitulina richardsi) on Tugidak Island, Alaska. Mar. Mamm. Sci. 17: 585-600. Jenouvrier, S., Barbraud, C. and Weimerskirch, H. 2003. Effects of climate variability on the temporal population dynamics of southern fulmars. J. Anim. Ecol. 74: 576-587. Koen Alonso, M . , Crespo, E. A., Garcia, N. A. and Coscarella, M . A. 2000. Food habits of the South American sea lion, Otaria flavescens, off Patagonia, Argentina. Fish. Bull. 98: 250-263. 57 Le Boeuf, B. J. and Reiter, J. 1991. Biological effects associated with El Nino, Southern Oscillation 1982-83, on northern elephant seals breeding at Ano Nuevo, California. In: Pinnipeds and El Nino: responses to environmental stress. 206-218. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Lea, M . A. 2002. The foraging activity and provisioning strategies of Antarctic fur seals in relation to changes in oceanographic conditions at the Kerguelen Archipelago. Pages 177. PhD Thesis. University of Tasmania, Hobart. Lea, M . A. and Dubroca, L. 2003. Fine-scale linkages between the diving behaviour of Antarctic fur seals and oceanographic features in the southern Indian Ocean. ICES J. Mar. Sci. 60: 990-1002. Lea, M . A., Hindell, M . A., Guinet, C. and Goldsworthy, S. D. 2002. Variability in the diving activity of Antarctic fur seals, Arctocephalus gazella, at lies Kerguelen. Polar Biol. 25: 269-279. Limberger, D., Trillmich, F., Biebach, H. and Stevenson, R. D. 1986. Temperature regulation and microhabitat choice by free-ranging Galapagos fur seal pups (Arctocephalus galapoensis). Oecologia 69: 53-59. Limberger, D., Trillmich, F., Kooyman, G. L. arid Majluf, P. 1983. Reproductive failure of fur seals in Galapagos and Peru in 1982-83. Trop. Ocean-Atmos. Newsl. October 1983: 16-17. Lunn, N. J. and Boyd, I. L. 1993a. Effects of maternal age and condition on parturition and the perinatal period of Antarctic fur seals. J. Zool., Lond. 229: 55-67. Lunn, N. J. and Boyd, I. L. 1993b. Influence of maternal characteristics and environmental variation on reproduction in Antarctic fur seals. Symp. Zool. Soc. Lond. 66: 115-129. Lunn, N. J., Boyd, I. L., Barton, T. and Croxall, J. P. 1993. Factors affecting the growth rate and mass at weaning of Antarctic fur seals at Bird Island, South Georgia. J. Mamm. 74: 908-919. 58 Lunn, N. J., Boyd, I. L. and Croxall, J. P. 1994. Reproductive performance of female Antarctic fur seals: the influence of age, breeding experience, environmental variation and individual quality. J. Anim. Ecol. 63: 827-840. Lydersen, C , Nost, O. A., Lovell, P., McConell, B., Gammelsrod, T., Hunter, C , Fedak, M . A. and Kovacs, K. M . 2003. Salinity and temperature structure of a freezing Arctic fjord-monitored by white whales (Delphinapterus leucas). Geophys. Res. Lett. 29: 34. MacLennan, D. N. and Simmonds, J. E. 1992. Fisheries acoustics. London: Chapman and Hall. Majluf, P. 1987. Reproductive ecology of female South American fur seals at Punta San Juan, Peru. PhD thesis. University of Cambridge. Majluf, P. 1991. El Nino effects on pinnipeds in Peru. In: Pinnipeds and El Nino: responses to environmental stress. 55-65. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Majluf, P. 1992. Timing of births and juvenile mortality in the South American fur seal in Peru. J. Zool., Lond. 227: 367-383. Massei, G., Genov, P. V. and Staines, B. W. 1996. Diet, food availability and reproduction of wild boar in a mediterranean coastal Area. Acta Theriologica 41: 307-320. McCafferty, D. J., Boyd, I. L. , Walker, T. R. and Taylor, R. I. 1998. Foraging responses of Antarctic fur seals to changes in the marine environment. Mar. Ecol. Prog. Ser. 166: 285-299. McCafferty, D. J., Boyd, I. L., Walker, T. R. and Taylor, R. I. 1999. Can marine mammals be used to monitor oceanographic conditions? Mar. Biol. 134: 387-395. McCann, T. S. 1987. Female fur seal attendance behaviour. N O A A Techn. Rep. NMFS. 51: 199-200. McPhaden, M . J. 1999. Genesis and evolution of the 1997-98 El Nino. Science 283: 950-954. Merrick, R. L. and Loughlin, T. R. 1997. Foraging behaviour of adult female and young-of-year Steller sea lions in Alaskan waters. Can. J. Zool. 75: 776-786. 59 Milette, L. L. and Trites, A. W. 2003. Maternal attendance patterns of lactating Steller sea lions (Eumetopias jubatus) from a stable and a declining population in Alaska. Can. J. Zool. 81: 340-348. Moron, O. 2002. Comparative analysis of warm events in the series 1970-2002 off the coast of Peru. Investig. Mar. 30: 98-99. ISSN 0717-7178. Muck, P., Zafra de Moreno, A. and Aranda, C. 1987. The seasonal dynamics of sea surface temperature and its impact on anchoveta distribution off Peru. In: The Peruvian upwelling ecosystem: dynamics and interactions. 33-44. Pauly, D., Muck, P., Mendo, J. and Tsukayama, I. ICLARM Conference Proceedings 18, 438 p. Instituto del Mar del Peru (IMARPE), Callao, Peru; Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), GmbH, Eschborn, Federal Republic of Germany; and International Center for Living Aquatic Resources Management (ICLARM) Manila, Philippines. Niquen, M . and Bouchon, M . 2002. Interactions between distribution and concentration of main pelagic resources in Peruvian waters during 1983-2001. Investig. Mar. 30: 196-197. ISSN 0717-7178. Ono, K. A., Boness, D. J. and Oftedal, O. T. 1987. The effect of a natural environmental disturbance on maternal investment and pup behaviour in the California sea lion. Behav. Ecol Sociobiol. 21: 109-118. Paredes, F. and Elliot, W. 1997. Estudio biologico pesquero del "camaroncito rojo" o "munida" Pleuroncodes monodon frente a Pucusana, Lima. Inf. Prog.Inst. Mar Peru 57: 21-34. Paredes, R. and Zavalaga, C. 1998. Overview of the effects of El Nino 1997-98 on Humboldt penguins and other seabirds at Punta San Juan, Peru. Penguin Conservation 11: 5-7. Pauly, D. and Palomares, M . L. 1987. New estimates of monthly biomass, recruitment and related statistics of anchoveta (Engraulis ringes) off Peru (4-14° S), 1953-1985. In: The Peruvian upwelling ecosystem: dynamics and interactions. 438. Pauly, D., Muck, P., Mendo, J. and Tsukayama, I. ICLARM Conference Proceedings 18. Instituto del 60 Mar del Peru (EVIARPE), Callao, Peru; Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), GmbH, Eschborn, Federal Republic of Germany; and International Center for Living Aquatic Resources Management (ICLARM) Manila, Philippines. Pauly, D. and Tsukayama, I. 1987. The Peruvian anchoveta and its upwelling ecosystem: three decades of change. ICLARM Studies and Review 15. Instituto del Mar del Peru (EVIARPE), Callao, Peru; Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), GmbH, Eschborn, Federal Republic of Germany; and International Center for Living Aquatic Resources Management, Manila, Philippines. Pistorius, P. A., Bester, M . N., Kirkman, S. P. and Taylor, F. E. 2001. Pup mortality in southern elephant seals at Marion Island. Polar Biol. 24: 828-831. Pitcher, K. W., Burkanov, V. N., Calkins, D. G., Le Boeuf, B. J., Mamaev, E. G., Merrick, R. L. and Pendleton, G. W. 2001. Spatial and temporal variation in the timing of births of Steller Sea Lions. J. Mammal. 82: 1047-1053. Pomeroy, P. and Duck, C. 2000. Breeding performance of grey seals at the Isle of May. In: The effect of large-scale industrial fisheries on non-target species. Harwood, J. Final Report to DGXTV, Directorate General Fisheries of the European Commission, under Contract 95/78. Reiter, J., Panken, K. J. and Le Boeuf, B. J. 1981. Female competition and reproductive success in northern elephant seals. Anim. Behav. 29: 670-687. Ritcher, T. A. 1999. The effects of density-dependent offspring mortality on the synchrony of reproduction. Evol. Ecol. 13: 167-172. Roux, J.-P. 1997. Proceedings of the International Workshop on Research and Management of Cape Fur Seals in Namibia. Swakopmund, Namibia, 24-26 June 1997: Ministry of Fisheries and Marine Resources of Namibia Ryther, J. H. 1969. Photosynthesis and fish production in the sea. Science 166: 72-76. 61 Santander, H. and Zuzunaga, J. S. 1984. Cambios en algunos componentes del ecosistema marino frente al Peru durante el fenomeno El Nino 1982-83. Rev. Com. Perm. Pacifico Sur 15: 311-331. Santojanni, A., Arneri, E. , Barry, C , Belardinelli, R., Cingolani, N., Gianetti, G. and Kirkwood, G. 2003. Trends in anchovy (Engraulis encrasicolus, L.) biomass in the northern and central Adriatic Sea. Sci. Mar. 67: 327-340. Senocak, T., Schnack, D. and Pauly, D. 1987. Mean monthly distribution, abundance and production of anchoveta eggs off Peru (4-14° S), 1964-1971 and 1972-1986. In: The Peruvian upwelling ecosystem: dynamics and interactions. Pauly, D., Muck, P., Mendo, J. and Tsukayama, I. ICLARM Conference Proceedings 18, 438 p. Instituto del Mar del Peru (IMARPE), Callao, Peru; Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), GmbH, Eschborn, Federal Republic of Germany; and International Center for Living Aquatic Resources Management (ICLARM) Manila, Philippines. Soto, K. H. 1999. Efectos de El Nino 1997-98 sobre el ciclo reproductivo del lobo marino chusco Otaria hyronia en las Islas Ballestas, Pisco, Peru. Pages 92. Tests Universidad Nacional Agraria La Molina, Lima. Thomas, J. A. and DeMaster, D. P. 1983. Parameters affecting survival of Weddell seal pups (Leptonychotes weddelli) to weaning. Can. J. Zool. 61: 2078-2083. Thompson, D., Duck, C. D., McConnell, B. J. and Garrett, J. 1998. Foraging behaviour of lactating female southern sea lions (Otaria flavescens) in the Falkland Islands. J. Zool., London 246: 135-146. Trillmich, F. 1985. Effects of the 1982-83 El Nino on Galapagos Fur Seals and Sea Lions. Noticias de Galapagos 42. Trillmich, F. 1986. Attendance behaviour of Galapagos fur seals. In: Fur seals: maternal strategies on land and at sea. 168-185. Gentry, R. L. & Kooyman, G. L. Princeton, NJ: Princeton University Press. 62 Trillmich, F. 1993. Influence of rare ecological events on pinniped social structure and population dynamics. Symp. Zool. Soc. Lond. 66: 95-114. Trillmich, F. and Dellinger, T. 1991. The Effects of El Nino on Galapagos Pinnipeds. In: Pinnipeds and El Nino: responses to environmental stress. 66-1 A. Trillmich, F. & Ono, K. New York: Springer-Verlag. Trillmich, F., Kooyman, G. L., Majluf, P. and Sanchez-Grinan, M . 1986. Attendance and diving behaviour of South American fur seals during El Nino in 1983. In: Fur Seals: maternal strategies on land and at sea. 153-167. Gentry, R. L. & Kooyman, G. L. Princeton, NJ: Princeton University Press. Trillmich, F. and Limberger, D. 1985. Drastic effects of El Nino on Galapagos pinnipeds. Oecol. (Berl.) 67: 19-22. Trillmich, F. and Ono, K. A. 1991. Pinnipeds and El Nino: responses to environmental stress. New York: Springer-Verlag Trillmich, F., Ono, K. A., Costa, D. P., DeLong, R. L. , Feldkamp, S. D., Francis, J. M . , Gentry, R. L. , Heath, C. B., Le Boeuf, B. J., Majluf, P. and York, A. E. 1991. The effects of El Nino on pinniped populations in the eastern Pacific. In: Pinnipeds and El Nino: responses to environmental stress. 247-260. Trillmich, F. & Ono, K. A. New York: Springer-Verlag. Trites, A. W. 1992. Reproductive synchrony and the estimation of mean date of birth from daily counts of northern fur seal pups. Mar. Mamm. Sci. 8: 44-56. Trites, A. W. and Antonelis, G. A. 1994. The influence of climatic seasonality on the life cycle of the Pribilof northern fur seal. Mar. Mamm. Sci. 10: 311-324. Trites, A. W. and Donnelly, C. P. 2003. The decline of Steller sea lions in Alaska: a review of the nutritional stress hypothesis. Mamm. Rev. 33: 3-28. Vaz-Ferreira, R. 1982. Otaria flavescens (Shaw), South American sea lion, hi: Mammals of the world. 447-495FAO, Rome: FAO Fish. Series. Werner, R. and Campagna, C. 1995. Diving behaviour of lactating southern sea lions (Otaria flavescens) in Patagonia". Can."J. Zool. 73: 1975-1982. 63 Wilson, R. P., Culik, B. M . , Bannasch, R. and Lage, J. 1994. Monitoring Antarctic environmental variables using penguins. Mar. Ecol. Prog. Ser. 106: 199-202. Winship, A. J., Trites, A. W. and Rosen, D. A. S. 2002. A bioenergetic model for estimating the food requirements of Steller sea lions (Eumetopias jubatus) in Alaska, USA. Mar. Ecol. Prog. Ser. 229:291-312. York, A. E. and Scheffer, V. B. 1997. The timing of implantation in the northern fur seal, Callorhinus ursinus. J. Mamm. 78: 675-683. 64 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0074872/manifest

Comment

Related Items