Working towards healthy selfsustaining populations for all Atlantic

Working towards healthy, self-sustaining populations for all Atlantic coast fish species or successful restoration well in progress by 2015 Ancillary Indices of Abundance for Black Sea Bass

Overview • The goal of the following presentation is to provide the SSC with information that they did not have during consideration of the black sea bass stock assessment • The point is not to second guess previous decisions but to supplement the existing information • This may allow for a case to be made to allow modest increases in harvest in this fishery

Data Treatment • Raw data from fishery independent trawl surveys conducted by the Atlantic Coastal states was gathered • The data were analyzed in a consistent fashion to provide some standardization between the different surveys • The surveys range throughout the biological range of black sea bass

Data Treatment • Black sea bass abundance (# fish) information was used from each dataset • A consistent set of metrics were examined E The metrics chosen were ones that were consistent amongst datasets • There are indications of zero inflation in the indices • To account for the zero inflation, the datasets were pivoted against the various metrics E Data rows were dropped in ranges or categories of the metric where there were less than 5 fish for the entire time series • In each case, the distribution of the data was analyzed, and is presented

Data Treatment • It was found that the distributions in each survey followed a negative binomial distribution • A negative binomial generalized linear model was performed on the datasets • Full models were run and then non significant variables were removed, the simplified model was rerun • In each case, if additional modifications to the dataset were performed, it is noted with a justification

VIMS Trawl Survey • Data distribution

VIMS Trawl Survey • Data distribution – zeroes removed

VIMS Trawl Survey • • Both MONTH and TEMP were signif, dropped MONTH as it’s a proxy for TEMP Model: ABUN ~ YEAR + TEMP + DEPTH + SAL + DO

New Jersey Trawl Survey • Data distribution

New Jersey Trawl Survey • Data distribution – zeroes removed and abundance truncated

New Jersey Trawl Survey • • Both MONTH and TEMP were signif, dropped MONTH as it’s a proxy for TEMP Model: ABUN ~ YEAR + TEMP + DEPTH

NEAMAP Trawl Survey • Data distribution

NEAMAP Trawl Survey • Data distribution – zeroes removed and abundance truncated

NEAMAP Trawl Survey • • • Both MONTH and TEMP were signif, dropped MONTH as it’s a proxy for TEMP Also STATION field was a unique identifier, not a station, so dropped Model: ABUN ~ YEAR + TEMP + DEPTH

NY Peconic Bay Trawl Survey • Data distribution

NY Peconic Bay Trawl Survey • Data distribution – zeroes removed

NY Peconic Bay Trawl Survey • Four years with 0 or 1 abundance removed from analysis (1987, 1989, 2005, 2010) • Model: ABUN ~ YEAR + TEMP + STATION

Long Island Sound Trawl Survey • Data distribution

Long Island Sound Trawl Survey • Data distribution – zeroes removed

Long Island Sound Trawl Survey • Several years removed from analysis due to missing covariates • Model: ABUN ~ YEAR + TEMP + DEPTH

Rhode Island Trawl Survey • Data distribution

Rhode Island Trawl Survey • Data distribution – zeroes removed

Rhode Island Trawl Survey • • Full model wouldn’t converge. Used output as a guide and removed non signif variables Model: ABUN ~ YEAR + TEMP

MA Inshore Trawl Survey • Data distribution

MA Inshore Trawl Survey • Data distribution – zeroes removed

MA Inshore Trawl Survey • Only one month in the survey so dropped • Model: ABUN ~ YEAR + TEMP + DEPTH

Scaled Survey Indices • All surveys

Scaled Survey Indices • Southern surveys

Scaled Survey Indices • Northern surveys

Standardized Index - All Surveys • Used same data as during previous analyses, all years shown • Model: ABUN ~ YEAR + TEMP + SURVEY

Standardized Index - All Surveys • 1990 – 2012 shown

Conclusions • A signal of increased abundance in latest years, though variable, was indicated throughout the analysis • The most notable and consistent signal is an increase in the last 2 years of data with the exception of VA • The data, especially data from NY north, corroborates the trends seen in the stock assessment • This information may add to the circumstantial case that abundance has continued to increase through time, and therefore cautious increases in allowed harvest may be warranted E This would help to test the system and gather more information on the stability of the stock assessment E An increase in harvest may help to decrease the high regulatory discard levels that are currently occurring in this fishery