An optimal allocation hauls sampling in order to reduce bias within fishing trip in trawlers operating in the Grand Sole and Porcupine areas (NE Atlantic)
Spanish demersal trawlers which operate in the Grand Sole and Porcupine Areas usually make between 50 and 80 hauls in each fishing trip, lasting around 15 days where fishing is almost continuous. Since the observer is not able to sample every haul, it is crucial to define a robust number of hauls to...
| Main Authors: | , |
|---|---|
| Format: | Report |
| Language: | English |
| Published: |
2004
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10508/9099 http://hdl.handle.net/10261/327908 |
| Summary: | Spanish demersal trawlers which operate in the Grand Sole and Porcupine Areas usually make between 50 and 80 hauls in each fishing trip, lasting around 15 days where fishing is almost continuous. Since the observer is not able to sample every haul, it is crucial to define a robust number of hauls to be sampled by the observer. A bootstrap analysis was carried out to determine the minimum number of hauls to be sampled to reduce significantly intra-variance within a fishing trip. Additionally, number of vessels and trips per vessel to be sampled was also analysed. Taking account the multistage sampling design and partitioning the over-all variability over the various stages, optimum sample sizes was estimated. On the other hand, fishermen may vary its retained catch and discard patterns whilst fishing trip occurs. Particular targets, discarded fish length or proportion of discarded/retained may change according to weather, occasional presence/absence of main target, storage space, quota exceeded, etc. To look into these plausible differences, every fishing trip was divided in three periods, beginning of the trip, half period and ending period, containing each one the same number of hauls. Several statistical tests were applied to these three periods to look for differences as well as to fit a consistent division of hauls to every period. Finally, an optimal allocation sampling scheme is suggested, both in terms of minimum number of hauls to be sampled and temporal division of those hauls along the fishing trip. |
|---|