Bycatch Estimates Research Articles

Modeling Low Rates of Seabird Bycatch in the U.S. Atlantic Longline Fishery

Total annual seabird bycatch in the U.S. Atlantic pelagic longline fishery was estimated from observer data by means of alternative methods to address the low frequency and uneven distribution of bird bycatch observations. Models used for estimation included generalized additive models (GAM), generalized linear models (GLM) and GLM with spatial or spatio-temporally auto correlated bycatch observations (s-GLM). Models were fitted separately to the probability of non-zero seabird capture (presence/absence) and to the numbers of seabirds per non-zero capture (positive bycatch); final estimates were obtained as the product of the two. Best-fitting models included latitude, longitude and quarter of year as covariates for presence/absence, and numbers of hooks per set for positive bycatch. Among the models compared, GLM gave the most consistent predictions of annual total seabird captures. For 2008, the most recent year modeled, GLM predicted 81 seabirds caught in 8,862 sets. The number is small on an Atlantic-wide or world-wide scale but is concentrated in an area of high species diversity that includes relatively small populations. GLM was recommended to be used in future Atlantic pelagic longline seabird bycatch estimation.

Cumulative estimates of sea turtle bycatch and mortality in USA fisheries between 1990 and 2007

Sea turtles interact with a variety of fishing gears across their broad geographic distributions and ontogenetic habitat shifts. Cumulative assessments of multi-gear bycatch impacts on sea turtle populations are critical for coherent fisheries bycatch management, but such estimates are difficult to achieve, due to low fisheries observer effort, and a single-species, single-fishery management focus. We compiled the first cumulative estimates of sea turtle bycatch across fisheries of the United States between 1990 and 2007, before and after implementation of fisheries-specific bycatch mitigation measures. An annual mean of 346,500 turtle interactions was estimated to result in 71,000 annual deaths prior to establishment of bycatch mitigation measures in US fisheries. Current bycatch estimates (since implementation of mitigation measures) are ∼60% lower (137,800 interactions) and mortality estimates are ∼94% lower (4600 deaths) than pre-regulation estimates. The Southeast/Gulf of Mexico Shrimp Trawl fishery accounts for the overwhelming majority of sea turtle bycatch (up to 98%) in US fisheries, but estimates of bycatch in this fishery are fraught with high uncertainty due to lack of observer coverage. Our estimates represent minimum annual interactions and mortality because our methods were conservative and we could not analyze unobserved fisheries potentially interacting with sea turtles. Although considerable progress has been made in reducing sea turtle bycatch in US fisheries, management still needs improvement. We suggest that sea turtle bycatch limits be set across US fisheries, using an approach similar to the Potential Biological Removal algorithm mandated by the Marine Mammal Protection Act.

Bycatch mitigation in context: Using qualitative interview data to improve assessment and mitigation in a data-rich fishery

Bycatch from pelagic longline fisheries has contributed to widespread population declines of turtles, sharks and other pelagic fishes. While large-scale estimates are needed to understand cumulative impacts on these highly migratory species, detailed information on targeting, setting, and discarding practices is needed to develop bycatch mitigation approaches. Data from qualitative fishers’ knowledge interviews with Canadian Atlantic pelagic longline captains was used to evaluate current bycatch estimation methods and to identify bycatch mitigation opportunities. Interviewed longline captains reported blue sharks ( Prionace glauca) were common bycatch during swordfish-targeted sets, but were sometimes absent from tuna-targeted sets. Discrepancies between longline captains’ observations and bycatch assessment methods identified needed improvements to data collection methods. Longline captains reported innovative uses of turtle dehooking gear, which two-thirds of interviewed captains had used to release other bycatch species in addition to turtles. Longline captains reported techniques for discarding pelagic stingray ( Pteroplatytrygon violacea), a common bycatch species in Pacific, Atlantic and Mediterranean pelagic longline fisheries. Therefore, such techniques could decrease fisheries impacts globally. While there can be major conservation benefits from fishers’ knowledge research, one-quarter of the active longline captains that we contacted declined interviews because they did not trust the larger research process. We urge conservation biologists to carefully design fishers’ knowledge research taking into account the often politicized context. Failure to do so may jeopardize future research and conservation efforts.

Protected Species Bycatch Estimating Approaches: Estimating Harbor Porpoise Bycatch in U.S. Northwestern Atlantic Gillnet Fisheries

Harbor porpoise bycatch estimates for federally managed gillnet fisheries in northwestern Atlantic US waters were calculated for a nine-year period (1999–2007) using two new methods, and the results were compared to the traditionally used stratified ratio estimation method. The aims of this research were to improve on the existing methods for estimating harbor porpoise bycatch for the Northeast and Mid-Atlantic US gillnet fisheries, to provide insight into the causes of harbor porpoise bycatch, and to compare bycatch estimation techniques that could be applied to other fisheries and species. The new methods included a model approach, and a ratio estimation approach that incorporated variables from both the regression model and the existing ratio estimation method. Initially, bycatch was modeled using a GAM forward stepwise process and included testing numerous variables describing the time, duration, and location of the fishing gear, the fishing gear configuration, and the environmental characteristics of the fished waters. The final model was simplified to a GLM and included variables describing port groupings, seasons, bottom depth, stretched mesh size, and year categories. The new mixed variable ratio estimation approach calculated both the Northeast and Mid-Atlantic bycatch estimates within the same framework by using the port groupings and mesh size variables from the model, and the season variable from the traditional ratio estimating approach. Bycatch estimates for the entire study area were similar between modeling and ratio estimator approaches, though estimates for the two new techniques were more stable from year to year in areas with less observed bycatch. The CVs for the model based estimates were much lower than ratio based estimates, and CVs for the two ratio estimation approaches were similar to each other. However, the model CVs may have been artificially low, as the model may have been over-parameterized in an attempt to accurately calculate annual estimates. Despite some differences, the estimates were not significantly different between approaches for the majority of comparisons.

Bycatch of the European purse seine tuna fishery in the Atlantic Ocean for the 2003–2007 period

Bycatch of several groups of species and their characteristics are presented for the period 2003 to 2007 for the European purse seine tuna fishery operating in the Atlantic Ocean. Data were collected through French and Spanish observer programmes and represented a total of 27 trips corresponding to 2.9% coverage. Bycatch is defined as non-targeted species and small or damaged target species. Bycatch species composition, main species length, sex ratio and the fate of the most common species are presented first. Stratified ratios relative to landings of major commercial tunas were then used to estimate the total bycatch; these ratios were considered the most appropriate variable for extrapolation. Stratification was based on the fishing mode (free school vs. floating object), season (quarters) and spatial areas. The annual average bycatch was estimated at about 6400 t, corresponding to a mean annual value of 80.8 t per 1000 t of tuna landed or 7.5% of the total catch. Tunas represent 83% (67.2 t/1000 t) of the total bycatch, followed by other bony fishes (10%, 7.8 t/1000 t), billfishes (5%, 4.0 t/1000 t), sharks (1%, 0.9 t/1000 t) and rays (1%, 0.9 t/1000 t). Based on estimates of the annual bycatch, 16% was kept on board and sold in local markets.

By-catch of cetaceans and other species of conservation concern during pair trawl fishing operations in the Adriatic Sea (Italy)

By-catch is one of the main sources of anthropogenic mortality in marine species of conservation concern worldwide. Between 2006 and 2008, the Consorzio Nazionale Interuniversitario per le Scienze del Mare (CoNISMa) coordinated a monitoring programme of cetacean by-catch in Italian pelagic trawlers, funded in compliance with European Regulation 812/2004. Sixteen independent observers monitored a total of 3141 hauls. The observation coverage ranged between 0.9 and 6.3% of the regional fishing effort. Almost all by-catch events were recorded in the northern Adriatic Sea. By-catch rates of bottlenose dolphins (Tursiops truncatus) and loggerhead turtles (Caretta caretta) were 0.0006 and 0.0255 individuals per haul, respectively. Given the low number of observed deaths, reliable estimates of total mortality for these two species were not obtained. The annual number of by-caught turtles was 863 (CV=0.15), with 99% released alive. A ‘hotspot’ for turtle captures was found off Goro (south Venice). The existence of lethal interactions makes it important to understand whether the scale of this mortality is sufficient to pose a threat at population level. Finally, annual by-catch estimates for rays and sharks were 5436 (CV=0.08) and 5414 (CV=0.15), respectively. Thintail threshers (Alopias vulpinus), piked dogfish (Squalus acanthias) and smooth-hounds (Mustelus mustelus), which are both commercial and vulnerable to overfishing, were taken in large numbers.

An interview-based approach to assess marine mammal and sea turtle captures in artisanal fisheries

Recent case studies have highlighted high bycatch mortality of sea turtles and marine mammals in artisanal fisheries, but in most countries there are few data on artisanal fishing effort, catch, or bycatch. With artisanal fisheries comprising >95% of the world’s fishermen, this knowledge gap presents a major challenge to threatened species conservation and sustainable fisheries initiatives. We report on results from an intensive pilot study to evaluate whether interview surveys can be effective in assessing fishing effort and threatened species bycatch. Fisheries and bycatch data from interviews with >6100 fishermen in seven developing countries were collected in <1 year for approximately USD $47,000, indicating that this approach may rapidly yield coarse-level information over large areas at low cost. This effort provided the first fisheries characterizations for many areas and revealed the widespread nature of high bycatch in artisanal fisheries. Challenges to study design and implementation prevented quantitative estimation or spatial comparisons of bycatch during this pilot research phase, but results suggested that annual sea turtle bycatch may number at least in the low thousands of individuals per country. Annual odontocete bycatch may number at least in the low hundreds per country. Sirenian bycatch occurred in all study areas but was frequent only in West Africa. We discuss lessons learned from this survey effort and present a revised protocol for future interview-based bycatch assessments.

Standardization of CPUE of loggerhead sea turtle (Caretta caretta) caught by pelagic longliners in the Southwestern Atlantic Ocean

The population abundance estimates used in stock assessments or required to establish management measures, depend on the sampling of the entire demographic spectrum of a population resident in a given area. However, for sea turtles, most population estimates are based mainly on nesting beach survey data and only consider a fraction of the population. The Southwest Atlantic Ocean (SWA) is an important foraging and development area for juveniles of the loggerhead sea turtle Caretta caretta where reproductive stocks from various nesting beaches mix. Declines in C. caretta populations have been observed in many parts of the world and bycatch rates of this species in the SWA are among the highest worldwide. This study standardizes the catch rates of loggerheads caught by pelagic longline fisheries in the region, using data collected by observer programs from Brazil and Uruguay. Generalized linear models (GLM) with a delta lognormal approximation were used. The variables used in the model take into account spatial and temporal variations as well as the characteristics of the fleet. In total, 6 272 344 hooks were observed between 1998 and 2007, with minimum effort registered in 2000 (12 010 hooks) and maximum effort in 2005 (1 989 431 hooks). During this period 3778 loggerheads were incidentally captured. The catch rates of loggerheads by the Uruguayan and Brazilian pelagic longline fisheries show oscillations through the years without a clear tendency; however, a low negative trend was observed from 1998 to 2005 with an increase in the last two years (2006 and 2007). The capture per unit of effort (CPUE) values varied between 0.38 to 1.78 ind/1000 hooks in 2005 and 2007, respectively. Distinct zones with differential catch rates were identified, with the higher CPUE values over the continental slope of Uruguay and adjacent waters. The incidental catch rates of this species are influenced, not only by fishing area, but also by year, season, sea surface temperature and gear type. In consequence, these variables and other potential ones should be considered in bycatch estimates by different fisheries because the loggerhead sea turtles are not uniformly distributed. This study intends to contribute not only to the general knowledge of loggerhead sea turtles in the SWA but to a future assessment of their populations at a global scale.

By-catch of franciscana Pontoporia blainvillei in Uruguayan artisanal gillnet fisheries: an evaluation after a twelve-year gap in data collection

The franciscana dolphin (Pontoporia blainvillei) is the most threatened small cetacean in the southwestern Atlantic Ocean. Its incidental capture in the Uruguayan fisheries of the Atlantic oceanic coast (AOC) has been recorded since 1940 and was systematically studied from 1974 to 1994, providing the most recent by-catch per unit of effort estimate (BPUE) for the AOC region (0.0064 franciscanas per 1000m net per fishing set). The lack of new by-catch data during the past 12 years has led to a data gap in by-catch estimates for the species in Uruguay. The current study was developed in two separate stages, a first stage (July 2004 - December 2005) designed to identify fisheries that interact with franciscana, and a second stage (2006), designed to determine new BPUE and franciscana mortality estimates for the selected fisheries. During the first stage, 13 artisanal fisheries of the Uruguayan coast were visited monthly, while in the second stage five fisheries (including the Rio de la Plata estuary and the AOC) were selected for monitoring. During 2006, 26 fishermen recorded all the information related to each fishing event, allowing the estimation of fishing effort calculated in linear units multiplied by hour (FEL). We also estimated a fishing set (FES) based by-catch rate which allowed the comparison with previously reported results. For 2006, the BPUEL (based on FEL) was estimated at 0.0020 franciscanas per 1000m net per hour and the BPUES at 0.0286 franciscanas per 1000m net per fishing set. The BPUEL was extrapolated to the surveyed fishing fleets during 2006, resulting in a mortality estimate of 289 (95% CI: 266-350) franciscanas. Based on the fishing dynamics reported herein, we consider the BPUEL the most accurate estimate of by-catch for the Uruguayan coast. This research updates the values of BPUE in the AOC after a 12-year information gap and includes the first by-catch evaluation in the Uruguayan estuarine coast. We highly recommend an abundance estimation of the species to complement the information reported herein in order to know the current status of franciscana dolphin population in Uruguay.

Assessing the effectiveness of conservation management decisions: likely effects of new protection measures for Hector's dolphin (Cephalorhynchus hectori)

Abstract Fisheries bycatch affects many species of marine mammals, seabirds, turtles and other marine animals. New Zealand's endemic Hector's dolphins overlap with gillnet and trawl fisheries throughout their geographic range. The species is listed as Endangered by the IUCN. In addition, the North Island subspecies has been listed as Critically Endangered. Estimates of catch rates in commercial gillnets from an observer programme (there are no quantitative estimates of bycatch by amateur gillnetters or in trawl fisheries) were used in a simple population viability analysis to predict the impact of this fishery under three scenarios: Option (A) status‐quo management, (B) new regulations announced by the Minister of Fisheries in 2008 and (C) total protection. Uncertainty in estimates of population size and growth rate, number of dolphins caught and other model inputs are explicitly included in the analysis. Sensitivity analyses are carried out to examine the effect of variation in catch rate and the extent to which fishing effort is removed from protected areas but displaced to unprotected areas. These methods are applicable to many other situations in which animals are removed from populations, whether deliberately (e.g. fishing) or not (e.g. bycatch). The current Hector's dolphin population is clearly depleted, at an estimated 27% of the 1970 population. Population projections to 2050 under Options A and B predict that the total population is likely to continue declining. In the case of Option B this is driven mainly by continuing bycatch due to the much weaker protection measures on the South Island west coast. Without fishing mortality (Option C) all populations are projected to increase, with the total population approximately doubling by 2050 and reaching half of its 1970 population size in just under 40 years. Copyright © 2009 John Wiley &amp; Sons, Ltd.

Characteristics and magnitude of sea turtle bycatch in US mid-Atlantic gillnet gear

From 1995 to 2006, US federal fisheries observers deployed aboard commercial fishing vessels in the US mid-Atlantic region documented captures of loggerhead Caretta caretta, green Chelonia mydas, Kemp's ridley Lepidochelys kempii, and leatherback Dermochelys coriacea turtles in commercial sink gillnet gear. Data collected by these observers were used to characterize sea tur- tle bycatch in sink gillnet gear, including the temporal and spatial distribution, fishing characteristics, species composition, and sizes of turtles captured. In addition, these data were used to develop a gen- eralized additive model to evaluate and predict bycatch rates (turtles per metric ton of fish landed) of sea turtles. These rates were then applied to commercial gillnet landings over the same time period to estimate total bycatch of loggerhead turtles. Bycatch rates of loggerheads were correlated with lat- itude, sea surface temperature, and mesh size. Highest predicted bycatch rates occurred in warm waters of the southern mid-Atlantic, in large-mesh gillnets. From 1995 to 2006, the average annual bycatch estimate of loggerheads was 350 turtles (coefficient of variation = 0.20, 95% CI over the 12 yr period: 234 to 504). Characteristics and magnitude of bycatch can help inform population assess- ments, while the distribution of bycatch rates can be used to help inform bycatch mitigation options.

Estimating seabird bycatch by fisheries and its impact on seabird populations

Estimating seabird bycatch by fisheries and its impact on seabird populations

Quantifying seabird bycatch: where do we go from here?

Bycatch of seabirds in offshore commercial fisheries has long been recognized as a serious conservation issue (e.g. Brothers, 1991; Brothers, Cooper & Lokkeborg, 1999), but until recently the problem was principally associated with industrial longline gear. In fact, pelagic seabird bycatch is very much a multi-gear problem. Watkins, Petersen & Ryan (2008) have presented the latest in a string of recent case studies documenting geographically widespread and sometimes severe levels of seabird bycatch in industrial trawl fisheries throughout the Southern Hemisphere (Weimerskirch, Capdeville & Duhamel, 2000; Sullivan, Reid & Bugoni, 2006a; Baker et al., 2007; Gonzales-Zevallos, Yorio & Caille, 2007). Substantial seabird bycatch occurs in trawl fleets of the Northern Hemisphere as well (e.g. in Alaskan trawl fisheries for groundfishes; NMFS, 2006). The issue is one of global importance that requires urgent international attention. National and international fisheries management bodies face several challenges as they move forward to address seabird bycatch within a multi-gear context. One fundamental question is what is the relative and cumulative threat to seabird populations from bycatch in trawl and longline gear? Answering this question depends on having reasonable estimates of mortality from these two gear types across the multitude of international fishing fleets with which particular seabird populations interact (e.g. Baker et al., 2007). This will require that new observer programs be established for national trawl fleets that interact with seabirds in their waters, and that international programs currently working to quantify seabird bycatch in high-seas longline fisheries (see Gilman & Moth-Poulsen, 2007) expand to quantify threats posed by trawl fisheries as well. Trawl gear poses a particularly difficult challenge for seabird bycatch estimation because a large but often unquantified proportion of birds killed by interactions with trawl cables (the dominant cause of mortality in trawl gears) is not hauled onto the vessel and therefore goes unobserved by traditional observer-data collection protocols (e.g. Gonzales-Zevallos & Yorio, 2006; Sullivan et al., 2006a). Watkins et al. (2008) observed that only two out of the 30 known seabird fatalities from cable interaction were eventually hauled aboard, corresponding at best to an overall 0.067 ( 0.046 binomial SE) detection rate for this type of mortality if only haul data had been used to estimated bycatch. This indicates the need to update observer protocols of seabird bycatch monitoring in trawl fisheries, to ensure proper accounting of bird mortalities resulting from cable collisions. Alternative methods for monitoring seabird bycatch in trawl fisheries should also be explored, as observer programs appropriate for monitoring seabird–trawl cable interactions are labor intensive and therefore expensive. Watkins et al. (2008) were only able to observe 0.5% of annual trawl fishing effort for this reason, and even some of the most well-funded fishery observer programs in the world (e.g. several in the United States) are too overstretched financially to meet observer coverage goals (Rossman, 2007). One promising approach is video monitoring (McElderry et al., 2004; Ames, Williams & Fitzgerald, 2005), which Watkins et al. (2008) used with apparent success during the first part of their study. Another strategy may be to augment traditional observer-data collection protocols, which are based on sampling hauled-in bycatch, over the long term with shorter intensive studies aimed at estimating species composition and non-detection rates for seabirds killed by cable interactions that are not retrieved during hauling of gear. These might then be applied as correction factors to statistically estimate actual seabird bycatch in trawls from data collected using conventional observer protocols. To our knowledge, Watkins et al. (2008) were the first to provide some empirical inference about the rate of bycatch non-detection in trawl fisheries and we encourage future studies to report similar data and estimates. As research and management programs respond to increase collection of trawl bycatch data, it would be helpful to adopt common metrics for reporting trawl bycatch. Watkins et al. (2008) reported bycatch per trawl hour, which

Loggerhead turtle ( Caretta caretta) bycatch in Canadian pelagic longline fisheries: Relative importance in the western North Atlantic and opportunities for mitigation

Little is known about the ecology of loggerhead turtles, Caretta caretta, in Canadian waters. However, Canada's eastern waters off the Scotian Shelf, Georges Bank and Grand Banks appear to be seasonal foraging habitats based on dates and locations where they have been captured as fishery bycatch. Estimates derived from data recorded by the international observer program (IOP) and used for this study suggest that thousands of mostly immature loggerheads have been captured in the Canadian pelagic longline fishery (PLF) in the western North Atlantic since 1999. These data suggest that the PLFs for tuna and swordfish are among the most important causes of incidental capture. Although U.S. and Canadian bycatch estimation methods differ and uncertainty associated with the estimates is high, it is clear from the IOP data that bycatch in the Canadian PLF is substantial and roughly within the same magnitude as that reported for the entire U.S. fleet in the North Atlantic for the same period (1999–2005). Analysis of bycatch observer data from the Canadian PLF and other empirical data suggests that fishing at temperatures below 20 °C using fish rather than squid bait and size-18 circle hooks could reduce loggerhead bycatch and mortality. However, research is needed to identify the best options and their commercial viability. Loggerheads that use Canadian Maritime waters are one of the least studied groups of loggerheads in the world, so systematic long-term studies on the ecology of loggerheads in Canada would help fill important information gaps and assist in developing effective recovery plans for this species.

A historical index of abundance for the blue shark ( Prionace glauca) in the western North Atlantic

The blue shark has been subject to bycatch fishing mortality in the western North Atlantic for almost a half-century. Nevertheless, stock status is ambiguous and improved input data are needed for stock assessments. It is particularly important to obtain reliable indices of abundance because of the uncertainty in estimates of bycatch. An index of relative abundance (catch-per-unit effort, CPUE) is developed for western North Atlantic blue sharks, starting from the mid-1950s, when industrial pelagic longline tuna fisheries began. Longline catch and effort records from recent observer programs (1980–1990s) were linked with longline survey records from both historical archives and recent cruises (1950–1990s). Generalized linear models (GLM) were used to remove the effects of diverse fishing target practices, geographical and seasonal variability that affect blue shark catch rates. The analysis revealed a decline in blue shark CPUE of approximately 30% in the western North Atlantic from 1957 to 2000. The magnitude of this CPUE decline was less than other recently published estimates and seems reasonable in light of the high productivity of the blue shark revealed by life-history studies and preliminary stock assessments.

Harbour porpoise Phocoena phocoena abundance in the southwestern Baltic Sea

Harbour porpoise Phocoena phocoena is the only cetacean species resident in the Baltic Sea. At least 2 different subpopulations occur in this area, with a presumed demarcation line in the western Baltic Sea. Aerial surveys were conducted during different seasons in the years 2002 to 2006, to obtain estimates of porpoise abundance for the southwestern Baltic. Within the survey area, 3 strata were defined. In total, 46 surveys were completed for these strata. Abundance estimates and corresponding confidence intervals (CI) were calculated for each stratum separately for each survey. Density between areas varied seasonally and spatially, with the western study area (Kiel Bight) generally showing the highest density. Whenever all 3 strata were covered within 25 consecutive days, abundance for the overall survey area was calculated. Ten surveys met these criteria. Overall abundance in the western Baltic varied between surveys, with the lowest value in March 2003 (457 ind.; 95% CI: 0 to 1632), and the highest estimate in May 2005 (4610 ind.; 95% CI: 2259 to 9098). The results of all other surveys were in the range from 1352 to 2905 ind., with largely overlapping confidence intervals. Applying the only available bycatch estimates (which include no measure of precision) currently available for the German Baltic region, the percent of porpoise bycatch in the western Baltic lies within a range of 1.78 to 17.94% of the local population. To increase the precision of bycatch rates, more detailed bycatch estimates for this region are needed. Nevertheless, the results indicate that bycatch is a major threat to harbour porpoises throughout the western Baltic Sea.

Estimation of seabird bycatch for North Pacific longline vessels using design- and model-based methods

Bycatch estimation for sensitive species is becoming increasingly important with the shift toward an ecosystem-based approach to fisheries management. Incidental mortalities for various seabird species occur on longline vessels throughout the world, including those in the North Pacific groundfish fleet. We present an approach to seabird bycatch estimation for North Pacific longline vessels using observer-collected data. Observers collect enormous amounts of data through a complex sampling design, but some information deficiencies preclude bycatch estimation using only probability sampling. Our approach combines probability sampling with model-dependent techniques to overcome these information deficiencies. The resulting bycatch estimator reflects the observer sampling design as closely as possible and minimizes reliance on untested model assumptions. We apply our estimator to black-footed albatross (Phoebastria nigripes) bycatch as an example and compare yearly estimates to those previously published. We also suggest changes in data collection that would further reduce dependence on model assumptions.

An assessment of the abundance and distribution of the Titicaca Flightless Grebe Rollandia microptera on Lake Titicaca and evaluation of its conservation status

Censuses were conducted from 15 June to 6 July and from 15 to 24 October 2003 throughout Lake Titicaca to determine the current distribution and abundance of the Titicaca Flightless Grebe Rollandia microptera. We surveyed 22 of an estimated 25 local population distribution ranges. We defined these as shallow offshore feeding areas in combination with coastal beds of tule-rushes (Schoenoplectus tatora). At this large spatial scale we found variously sized populations of grebes in all 22 of the areas surveyed and a negative correlation between the total area of tule-bed within each area and grebe density. Absolute counts of grebes pooled across all sites produced a total of 2,582 individuals on Lake Titicaca. Given that some sites were either partially surveyed or not surveyed at all we feel that this is an underestimate of the total lake population. Limitations in methodologies used precluded the classification to development stage of approximately 43% of the individuals counted. Among the remaining 57% of the individuals counted we identified 732 mature adults within the total population of Lake Titicaca. When combined with information from other sources and current knowledge of habitat fragmentation our results support the classification of the species in the IUCN Red List as Endangered under criterion C2a. Of 20 local fisherman interviewed on the Bolivian side of Lake Titicaca, 90% confirmed that individuals of Titicaca Flightless Grebe incidentally drown in their fishing nets although responses varied from weekly to monthly estimates of by-catch. Although this study provides the first estimate of the total population of Titicaca Flightless Grebes on Lake Titicaca, a standardized monitoring programme will be required in order (1) to provide a baseline for the interpretation of future population trends in the context of impacts generated by human activities and (2) to aid in future management decisions for the long-term conservation of the species.

Bycatch of Marine Mammals in U.S. and Global Fisheries

Fisheries bycatch poses a significant threat to many populations of marine mammals, but there are few published estimates of the magnitude of these catches. We estimated marine mammal bycatch in U.S. fisheries from 1990 to 1999 with data taken from the stock assessment reports required by the U.S. Marine Mammal Protection Act. The mean annual bycatch of marine mammals during this period was 6215 +/- 448 (SE). Bycatch of cetaceans and pinnipeds occurred in similar numbers. Most cetacean (84%) and pinniped (98%) bycatch occurred in gill-net fisheries. Marine mammal bycatch declined significantly over the decade, primarily because of a reduction in the bycatch of cetaceans. Total marine mammal bycatch was significantly lower after the implementation of take reduction measures in the latter half of the decade. We derived a crude first estimate of marine mammal bycatch in the world's fisheries by expanding U.S. bycatch with data on fleet composition from the Food and Agriculture Organization. The global bycatch of marine mammals is in the hundreds of thousands. Bycatch is likely to have significant demographic effects on many populations of marine mammals. Better data are urgently needed to fully understand the impact of these interactions.

A risk assessment for Pacific leatherback turtles (Dermochelys coriacea)

Leatherback turtles (Dermochelys coriacea) are critically endangered in the eastern and western Pacific Ocean. Here, I estimate the magnitude of two likely causes of their decline: (i) bycatch by longline fishing vessels and (ii) coastal sources of mortality. I calculate point estimates of longline bycatch based on turtle catch rates from the US Hawaii-based fleet and effort data for the international Pacific longline fleet. I estimate the intrinsic growth rate of the population and the magnitude of coastal mortality by fitting a simple logistic model. In the western and central Pacific, coastal sources lead to a 13% annual mortality rate, compared with a point estimate of 12% from longlining. In the eastern Pacific, coastal sources account for a 28% annual mortality rate, compared with a point estimate of only 5% from longlining. A Bayesian risk assessment reveals the importance of reducing coastal sources of mortality, as well as longline bycatch, if the populations are to avoid extinction. International efforts to protect the leatherback should expand beyond focusing solely on longline bycatch and should attempt to reduce coastal harvest of adult females and eggs, as well as reduce bycatch by inshore gears such as gillnets.