Recreational harvest estimates for 1999-2000 (1 December 1999 to 30 November 2000) are presented for a wide range of fish and shellfish species. The estimates are based on a similar, but enhanced, survey methodology that has been adopted for estimating recreational harvests in previous surveys. The survey methodology involves combining results from at three separate but related surveys. An estimate of fisher prevalence derived from a nation-wide face to face survey is combined with detailed diary data of recreational harvests recorded by a nation-wide sample of recreational fishers recruited by telephone, to estimate recreational harvests in numbers of fish or shellfish. Estimated harvests in numbers of fish or shellfish were converted to total harvest weight using the results of a boat ramp survey to estimate the mean weight of recreationally harvested fish and shellfish.

Key enhancements over previous surveys included the use of a face to face survey for measuring fisher prevalence, improved methods for weighting up diarists' harvests using extensive demographic data and a more appropriate method for estimating coefficients of variation.

Estimates for the 1999-2000 national marine recreational fishing survey are much higher than the estimates from previous surveys. The harvest estimate for SNA1 which has a c.v. of 11% is in excess of 6.9 million fish and 6 200 tonnes. Very few of the harvest estimates have c.v.s of less than 20%. Most of the fishstock recreational harvest estimates presented in the report are higher than previous estimates by a factor of two to three times. Coefficients of variation (c.v.s) for the harvest estimates are much larger than estimated for previous surveys but are more reflective of the complex nature of the survey design and the highly skewed nature of diarists' harvests. These factors were not taken into account in the method used for estimates of c.v.s in previous surveys.

Some of the 1999-2000 harvest estimates, particularly the estimates for a number of key fishstocks in QMA2 appear to be implausibly high. While the reasons for this are not known, the small sample size for this area may have resulted in a biased sample of diarists.

Results from pilot surveys undertaken as part of the 1999-2000 survey together with a review of the available literature strongly suggest that previous harvest estimates from the 1996 national survey and earlier regional surveys are highly unreliable and probably much too low. Therefore, caution should be exercised in comparing the estimates presented here with the estimates from previous surveys as such comparisons are likely to be misleading.

The improved survey methodology and estimation procedures adopted for the 1999-2000 national marine recreational fishing survey mean that the reported harvest estimates should be more accurate than the estimates from prior surveys. The much higher recreational harvest estimates have significant fisheries sustainability and management consequences. Future surveys to estimate recreational harvests will need to focus on making further improvements to the survey methodology and improving the precision of estimates.

For the full document (26 pages) :

If you have Microsoft Word www.option4.co.nz/documents/2000est.doc (140Kb)

Or in a PDF document www.option4.co.nz/documents/2000est.pdf (79Kb)

If you do not have Acrobat reader (or need to upgrade) click here »»