Using poststratification to improve abundance estimates from multispecies surveys: a study of juvenile flatfishes

Population assessments seldom incorporate habitat information or use previously observed distributions of fish density. Because habitat affects the spatial distribution of fish density and overall abundance, the use of habitat information and previous estimates of fish density can produce more precise and less biased population estimates. In this study, we describe how poststratification can be applied as an unbiased estimator to data sets that were collected under a probability sampling design, typical of many multispecies trawl surveys. With data from a multispecies survey of juvenile flatfish, we show how poststratification can be applied to a data set that was not collected under a probability sampling design, where both the precision and the bias are unknown. For each of four species, three estimates of total abundance were compared: 1) unstratified; 2) poststratified by habitat; and 3) poststratified by habitat and fish density (high fish density and low fish density) in nearby years. Poststratification by habitat gave more precise and (or) less design-biased estimates than an unstratified estimator for all species in all years. Poststratification by habitat and fish density produced the most precise and representative estimates when the sample size in the high fish-density and low fish-density strata were sufficient (in this study, n≥20 in the high fish-density stratum, n≥9 in the low fish-density stratum). Because of the complexities of statistically testing the annual stratified data, we compared three indices of abundance for determining statistically significant changes in annual abundance. Each of the indices closely approximated the annual differences of the poststratified estimates. Selection of the most appropriate index was dependent upon the species’ density distribution within habitat and the sample size in the different habitat areas. The methods used in this study are particularly useful for estimating individual species abundance from multispecies surveys and for retrospective st

Survey Assessment of Semi-pelagic Gadoids: The Example of Walleye Pollock, Theragra chalcogramma, in the Eastern Bering Sea

Assessment of walleye pollock, Theragra chalcogramma, in the eastern Bering Sea is complicated because the species is semi-pelagic in habit. Annual bottom trawl surveys provide estimates of demersal abundance on the eastern Bering Sea shelf. Every third year (starting in 1979), an extended area of the shelf and slope is surveyed and an echo integration-midwater trawl survey provides estimates of pollock abundance in midwater. Overall age-specific population and biomass estimates are obtained by summing the demersal and midwater results, assuming that the bottom trawl samples only pollock inhabiting the lower 3 m of the water column. Total population estimates have ranged from 134 x 109 fish in 1979 to 27 x 109 fish in 1988. The very high abundance observed in 1979 reflects the appearance of the unusually large 1978 year class. Changes in age-specific abundance estimates have documented the passage of strong (1978, 1982, and 1984) and weak year classes through the fishery. In general, older fish are more demersally oriented and younger fish are more abundant in midwater, but this trend was not always evident in the patterns of abundance of 1- and 2-year-old fish. As the average age of the population has increased, so has the relative proportion of pollock estimated by the demersal surveys. Consequently, it is unlikely that either technique can be used independently to monitor changes in abundance and age composition. Midwater assessment depends on pelagic trawl samples for size and age composition estimates, so both surveys are subject to biases resulting from gear performance and interactions between fish and gear. In this review, we discuss survey methodology and evaluate assumptions regarding catchability and availability as they relate to demersal, midwater, and overall assessment.

Crayfish survey of the Weaver, Dane, Goyt and Etherow catchments

This is the report on the Crayfish Survey of the Weaver, Dane, Goyt and Etherow catchments from 1998 by the Environment Agency. The aims of this report are: Firstly, to present the findings of the crayfish survey and details of the sites visited. Secondly to present the information on distribution maps with past records so that the current status can be seen and finally to use this information so that recommendations for the conservation of native crayfish can be made in accordance with the national action plan for this species and the Environment Agency’s Species Management Programme. The report contains sections on background, going through legislation, distribution and requirements of both native and non-native crayfish. Sections on methodology, results and discussion, conclusion and recommendations. The appendix I contains maps showing the sampling points locations. Details of sampling sites are summarized in appendix II. Appendix III contains previous crayfish records and Appendix IV shows the field data recording form. Finally, a collection of photographs are displayed in appendix V.

River Teign Fisheries Survey 1979

This is the River Teign Fisheries Survey from August 1979 by the South West Water Authority. The River Teign was sampled by electrofishing at fifteen sites and population estimates, average lengths, weights and biomass were calculated for each species present, and where possible for individual age classes of those species. Results indicated that a stable community structure existed, and little had changed from 1963. Salmonids accounted for the majority of the biomass, and within this group trout were the most abundant in the headwater, and salmon in the lower reaches. All tributaries sampled had higher densities and biomass than the main river. The results were compared with other data collected from similar surveys of other rivers in Devon by the South West Water Authority. Both the average length of each age class, and the biomass in the main River Teign appeared to be lower than in most other rivers, although in the Rivers Wray and Lemon these values were particularly high.

River Goyt and Etherow Crayfish Survey

This is the River Goyt & Etherow Crayfish Survey report from the Environment Agency held between August and November 2000. The report focuses on the need to verify the presence of the non-native American signal crayfish (Pacifastacus leniusculus) in the main River Goyt, and determine its distribution. Signal crayfish have been reported in the River Goyt after spreading from a known population centre in Hollywood End Brook, to the east of Marple Bridge. The report contains sections on results, discussion and a summary. The section on results shows maps of presence or absence of different crayfish species in River Goyt, and the Mersey/ Weaver river systems such American signal crayfish (Pacifastacus leniusculus) and European freshwater crayfish (Austopotamobius pallipes).

River Bovey (main Teign tributary) fisheries survey 1978

This is the River Bovey (main Teign tributary) fisheries survey 1978 report produced by South West Water Authority in 1978. The aim of this report is to confirm or disprove the presence of alien coarse fish species thought present in the Bovey system. A chosen site was electro fished using D.C. methods wading upstream. All fish taken were measured and identified. Since it was not possible to distinguish between the young parr stages of migratory and non migratory trout, all parr taken have been included under the heading of Brown trout whereas in fact some of these parr may become migratory later in their life cycle.

Modeling the Potential Spread of the Recently Identified Non-Native Panther Grouper (Chromileptes altivelis)in the Atlantic Using a Cellular Automaton Approach

The Indo-pacific panther grouper (Chromileptes altiveli) is a predatory fish species and popular imported aquarium fish in the United States which has been recently documented residing in western Atlantic waters. To date, the most successful marine invasive species in the Atlantic is the lionfish (Pterois volitans/miles), which, as for the panther grouper, is assumed to have been introduced to the wild through aquarium releases. However, unlike lionfish, the panther grouper is not yet thought to have an established breeding population in the Atlantic. Using a proven modeling technique developed to track the lionfish invasion, presented is the first known estimation of the potential spread of panther grouper in the Atlantic. The employed cellular automaton-based computer model examines the life history of the subject species including fecundity, mortality, and reproductive potential and combines this with habitat preferences and physical oceanic parameters to forecast the distribution and periodicity of spread of this potential new invasive species. Simulations were examined for origination points within one degree of capture locations of panther grouper from the United States Geological Survey Nonindigenous Aquatic Species Database to eliminate introduction location bias, and two detailed case studies were scrutinized. The model indicates three primary locations where settlement is likely given the inputs and limits of the model; Jupiter Florida/Vero Beach, the Cape Hatteras Tropical Limit/Myrtle Beach South Carolina, and Florida Keys/Ten Thousand Islands locations. Of these locations, Jupiter Florida/Vero Beach has the highest settlement rate in the model and is indicated as the area in which the panther grouper is most likely to become established. This insight is valuable if attempts are to be made to halt this potential marine invasive species

A survey of the marine nematodes of Chesapeake Bay, Maryland

The study here reported is a survey of the most common non-parasitic nematode families of Chesapeake Bay, Maryland, with descriptions and figures, so that ecological workers and students of invertebrate zoology may be encouraged not to pass over this highly interesting and abundant invertebrate phylum. This survey is not a complete account of the free-living nematode population of the Bay, however, since only the middle section of the Bay was sampled and since the collections were not made systematically throughout the year. The physical and chemical factors of Chesapeake Bay may be found in several publications of the Chesapeake Biological Laboratory, Solomons Island, Maryland, and in the records of the Chesapeake Bay Institute, Annapolis, Maryland.

Capture probability of a survey trawl for red king crab (Paralithodes camtschaticus)

The relative abundance of Bristol Bay red king crab (Paralithodes camtschaticus) is estimated each year for stock assessment by using catch-per-swept-area data collected on the Alaska Fisheries Science Center’s annual eastern Bering Sea bottom trawl survey. To estimate survey trawl capture efficiency for red king crab, an experiment was conducted with an auxiliary net (fitted with its own heavy chain-link footrope) that was attached beneath the trawl to capture crabs escaping under the survey trawl footrope. Capture probability was then estimated by fitting a model to the proportion of crabs captured and crab size data. For males, mean capture probability was 72% at 95 mm (carapace length), the size at which full vulnerability to the survey trawl is assigned in the current management model; 84.1% at 135 mm, the legal size for the fishery; and 93% at 184 mm, the maximum size observed in this study. For females, mean capture probability was 70% at 90 mm, the size at which full vulnerability to the survey trawl is assigned in the current management model, and 77% at 162 mm, the maximum size observed in this study. The precision of our estimates for each sex decreased for juveniles under 60 mm and for the largest crab because of small sample sizes. In situ data collected from trawl-mounted video cameras were used to determine the importance of various factors associated with the capture of individual crabs. Capture probability was significantly higher when a crab was standing when struck by the footrope, rather than crouching, and higher when a crab was hit along its body axis, rather than from the side. Capture probability also increased as a function of increasing crab size but decreased with increasing footrope distance from the bottom and when artificial light was provided for the video camera.

Assessing the precision of frequency distributions estimated from trawl-survey samples

In trawl surveys a cluster of fish are caught at each station, and fish caught together tend to have more similar characteristics, such as length, age, stomach contents etc., than those in the entire population. When this is the case, the effective sample size for estimates of the frequency distribution of a population characteristic can, therefore, be much smaller than the number of fish sampled during a survey. As examples, it is shown that the effective sample size for estimates of length-frequency distributions generated by trawl surveys conducted in the Barents Sea, off Namibia, and off South Africa is on average approximately one fish per tow. Thus many more fish than necessary are measured at each station (location). One way to increase the effective sample size for these surveys and, hence, increase the precision of the length-frequency estimates, is to reduce tow duration and use the time saved to collect samples at more stations.

Longitudinal logbook survey designs for estimating recreational fishery catch, with application to ayu (Plecoglossus altivelis)

Longitudinal surveys of anglers or boat owners are widely used in recreational fishery management to estimate total catch over a fishing season. Survey designs with repeated measures of the same random sample over time are effective if the goal is to show statistically significant differences among point estimates for successive time intervals. However, estimators for total catch over the season that are based on longitudinal sampling will be less precise than stratified estimators based on successive independent samples. Conventional stratified variance estimators would be negatively biased if applied to such data because the samples for different time strata are not independent. We formulated new general estimators for catch rate, total catch, and respective variances that sum across time strata but also account for correlation stratum samples. A case study of the Japanese recreational fishery for ayu (Plecoglossus altivelis) showed that the conventional stratified variance estimate of total catch was about 10% of the variance estimated by our new method. Combining the catch data for each angler or boat owners throughout the season reduced the variance of the total catch estimate by about 75%. For successive independent surveys based on random independent samples, catch, and variance estimators derived from combined data would be the same as conventional stratified estimators when sample allocation is proportional to strata size. We are the first to report annual catch estimates for ayu in a Japanese river by formulating modified estimators for day-permit anglers.

The measurement error in marine survey catches: the bottom trawl case

We have formulated a model for analyzing the measurement error in marine survey abundance estimates by using data from parallel surveys (trawl haul or acoustic measurement). The measurement error is defined as the component of the variability that cannot be explained by covariates such as temperature, depth, bottom type, etc. The method presented is general, but we concentrate on bottom trawl catches of cod (Gadus morhua). Catches of cod from 10 parallel trawling experiments in the Barents Sea with a total of 130 paired hauls were used to estimate the measurement error in trawl hauls. Based on the experimental data, the measurement error is fairly constant in size on the logarithmic scale and is independent of location, time, and fish density. Compared with the total variability of the winter and autumn surveys in the Barents Sea, the measurement error is small (approximately 2–5%, on the log scale, in terms of variance of catch per towed distance). Thus, the cod catch rate is a fairly precise measure of fish density at a given site at a given time.

A report of the Fisheries Catch Assessment Survey in the Ugandan waters of Lake Victoria for the February 2008 survey

Between July 2005 and February 2008, ten Catch Assessment Surveys (CASs) were conducted at 54 pre-selected fish landing sites in the Ugandan part of Lake Victoria comprising approximately 10% of all landing sites in each of the 11 districts sharing the lake. The CASs were conducted following regionally harmonised Standard Operating Procedures (SOPs). This report covers the CAS conducted in February 2008 and puts into context the trends generated by results of the previous surveys. The catch rates of Nile perch in gillnetting boats with motor/sail, reduced from 26.9 kg boat-1 day-1 in August 2007 to 22.8 kg boat-1 day-1 in February 2008. Whereas the catch rates of paddle Sesse boats remained more or less the same as in August 2007. The Nile perch catch rates of the long line fishery of the boats using motor/sail was similar, 35 and 36 kg boat-1 day-1 in August 2007 and February 2008 respectively but the catch rates of paddle Sesse boats using long lines showed some more increase from 19 to 22 kg boat (-1) day (-1). In the tilapia fishery, the catch rates of the parachute boats using gillnets showed further decrease in a row from 12.6 kg boat-1 day-1 in December 2006, 11.6 kg boat-1 day-1 in March 2007, 11.2 kg boat-1 day-1 in August 2007 and 10.0 kg boat-1 day-1 in February 2008. The overall impact of reduced catch rates in the predominant effort groups, e.g. gillnetting boats using motor/sail in the Nile perch fishery and Parachute boats using gillnets in the tilapia fishery overshadowed the increases in less dominant effort groups and resulted in the lowest monthly catch estimates recorded in the surveys conducted since 2005. Whereas there was a clear downward trend in the Nile perch catch rates of boats using gillnets, which corroborates with the information of declining stocks from the recent Acoustics surveys, the catch rates in the long line fishery remained stable and even somewhat increasing in the last four surveys. The factors that maintain high catches against reduction of fish biomass in the long line fishery and their effects on sustainability of the Nile perch fishery should be investigated further. The Mukene fishery, characterised by large fluctuations in the catch rates did not show much change in the last two surveys in August 2007 and February 2008 and the annual catch estimates showed an overall increase of 7% from 2005 to 2007. The Mukene fishery in the Ugandan waters of Lake Victoria remained a near shore fishery in which paddle Sesse boats using small seines or scoop nets were the dominant craft.

Report of Catch Assessment Survey of Lake Albert – Albert Nile conducted in July 2012

A Catch Assessment Surveys (CAS) with the overall objective of generating information on the fish production and commercial value of the fisheries of Lake Albert and Albert Nile system was implemented by the National Fisheries Resources Research Institute (NAFIRRI) in collaboration with the Department of Fisheries Resources (DFR), Local Government staff (FOs) and BMU members at selected landing sites on Lake Albert (12 landing sites) and Albert Nile (26 landing sites) in July 2012. A total 622 and 313 boat days on Lake Albert and Albert Nile respectively were sampled for a period of 9 days. Results indicate an annual landing of 151,600 and 5,900 tonnes (t) of fish with an estimated beach value of 122.5 and 14 Billion (UShs) from Lake Albert and Albert Nile respectively. Over 80% of the catch from Lake Albert comprised the small pelagic species; Neobola bredoi (Muziri) and Brycinus nurse (Ragoogi) followed by Nile perch (6%). However, due to low market value of the small fishes and the high prices attached to Nile perch for industrial processing and export market, the contribution of the latter to beach value rose to 34% of the total. The contribution of the light fishery based on small pelagic species (B. nurse and N. bredoi) are insignificant on Albert Nile. Even if the small pelagic species may be present in the river system, a light fishery based on these two is yet to be developed. Proportionally, Albert Nile still remain a multispecies fishery with over 20 fish species harvested commercially. Interestingly, the Albert Nile fishery still remains primitive with simple crafts and gears (mainly dugout canoes, traps, and gillnets). This could suggest that the more developed the system becomes the higher the level of transformation in its fisheries leading to simplification, characterized by reduction in multispecies nature and dominance of few species. Illegal gears especially undersized gillnet of mesh size less than 4 inches were the most dominant in the Lake Albert and Albert Nile fisheries. They captured large quantities of immature fish particularly when used to target Nile perch, Bagrus, Nile tilapia, and large Barbus spp. Their impact when used to target the smaller species (Ragoogi, Angara & Ngasia) is yet to be evaluated. A specific study to analyze selectivity and impacts of these nets is a recommended. However, the dominance of 1.5” mesh sizes especially on Albert Nile to target Angara, Ngassia and Barbus, is definitely destructive to their fisheries and should be checked forthwith. In addition, there is an emerging fishing method locally referred to as “Salsio or Luzira” whereby fishers stay on the lake from 3 days up to 2 weeks without returning to the landing site. They carry with them food and salt for processing the catches on the lake, and in the case of Albert Nile on make shift shelters on islands and in the game park. They normally use gillnets of 3-3½ inch mesh size and caught mainly Nile perch & Bagrus (Pethi & Munama). On return they land several tons of fish. Most of these Catches are not captured in the estimates presented in our analyses since we target daily fishing boats. The possible impacts of this fishing method should be studied and appropriate action recommended.

Final report of the Fisheries Catch Assessment Survey in the Ugandan waters of Lake Victoria for the May 2011 survey

This report presents findings of the CAS conducted in the Ugandan waters of Lake Victoria in May 2011. The results of the previous eleven CASs conducted under the IFMP of the LVFO programme in July, August, September and November 2005; in March, August and December 2006; in March and August 2007; in February and December 2008; and March 2010 are included to show the emerging trends. The report also presents annual catch estimates for the Ugandan part of the lake from 2005 to 2011. Through these CASs, information is building up to show the emerging picture of fish production in the Ugandan waters of the lake. Similar surveys are conducted in the Kenyan and Tanzanian parts of the lake, which provide the lake wide perspective of fisheries production but this time not simultaneously as under the LVFO effort due to different sources and timing of funding. These data can now be utilised together with other Resource and Socio-economic Monitoring survey data for a stock assessment of the lake to provide a firm basis for planning and management of the fisheries resources.