720 



Abstract— 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 mea- 

 surement). The measurement error is 

 defined as the component of the var- 

 iability 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 offish density at 

 a given site at a given time. 



The measurement error in marine survey catches: 

 the bottom trawl case 



Vidar Hjellvik 

 Olav Rune Gode 



Institute of Marine Research 



Nordnesgt 33 



PO Box 1870, Nordnes 



N-5817 Bergen, Norway 



E-mail address (for V Hiellvik) vidarfi'a'imrno 



Dag Tjastheim 



Department of Mathematics 

 University of Bergen, 

 Johs Brunsgt 12 

 N-5008 Bergen, Norway 



Manuscript accepted 20 February 2002. 

 Fish. Bull. 100:720-726 (2002). 



Surveys are vital for estimating the 

 size and composition of marine pop- 

 ulations. In the data collection pro- 

 cess, trawl samples or acoustics (or 

 both) are usually employed. It is well 

 known that the resulting estimates are 

 subject to substantial variations, and it 

 is important to quantify and explain, 

 as much as is possible, the variability 

 in terms of relevant explanatory 

 variables or covariates. Typically these 

 variables will depend on the sampling 

 tool used, but for bottom trawl catches 

 important explanatory variables can 

 be depth and location of the haul (Pola- 

 check and Volstad, 199.3), time of the 

 day (Korsbrekke and Nakken, 1999), 

 season, strength of the year classes 

 involved, etc. 



Generally, as the number of covar- 

 iates increases, and the model becomes 

 more complex, the residual variation (or 

 remaining uncertainty) not explained 

 by the model decreases. But no matter 

 how refined the model is. there will 

 always be an unexplained random 

 component that cannot be attributed 

 to any observed variable. This residual 

 variation is caused by the interactions 

 between the fish, the measurement 

 device, and the environment (see e.g. 

 Engas, 1994). 



The purpose of this study was to 

 define and quantify this residual source 

 of random variation. This was done by 

 analyzing meastu-ements from parallel 

 tows of multiple vessels, which is the 



closest one can come to a controlled 

 statistical experiment in this context. 

 The importance of quantifying this type 

 of fluctuation lies in the fact that it 

 is a benchmark uncertainty, which is 

 inherent in the survey process itself, 

 and in this sense it may be termed a 

 measurement error If the measurement 

 error can be assessed from field data 

 and is consistent over time and space, 

 we improve our understanding and 

 quantification of other causal factors 

 behind the uncertainty associated with 

 survey estimates. 



In this study we looked at bottom 

 trawl catches of cod, but we would 

 like to stress that the concepts and 

 techniques developed can in principle 

 be applied to acoustic survey estimates 

 or indeed to any type of measurements 

 collected simultaneously by two or 

 more independent parallel sampling 

 devices. It should be noted that there 

 is a growing related literature on 

 comparative survey analysis. We refer 

 to the review paper by Pelletier (1998) 

 and references therein. 



Materials and methods 



Parallel trawling experiments 



During the annual combined bottom 

 trawl and acoustic survey of demersal 

 fish in the Barents Sea during winter 

 and autumn conducted by the Institute 



