Pāua catch per unit effort from logger data

Pāua (primarily Haliotis iris) fisheries in New Zealand are currently managed in Quota Management Areas (QMAs). An further assumption used in pāua stock assessment is that catch per unit effort (CPUE) provides an index of abundance There are several key problems that can confound the relationship between haliotid CPUE and abundance. In particular, abalone are vulnerable to serial depletion where accessible pāua are fished initially, but there is a progression of fishing to other reefs with a similar CPUE once the accessible pāua have been harvested.

In addition, management of pāua is hampered by the limitations of current data collection that only gives aggregate data at a statistical area and daily scale. To address these issues, pāua loggers were introduced into the New Zealand commercial fishery in October 2010. The loggers are compact units that fit in a pocket on the back of a diver’s wetsuit. When the diver is at the surface they record their position, using a GPS, and when they are underwater the loggers record the diver’s depth. A unit is also kept on the boat, and is used to record the location and time of catch bags being brought onboard.

In order to help develop methods for interpreting the pāua logger data, a fishdown experiment was carried out in Fighting Bay (in the outer Marlborough Sounds) in the summer of 2011—12. Fighting Bay was closed to all fishing in 1996, and so has a high pāua abundance. During the fishdown, a designated area within Fighting Bay was repeatedly fished, using commercial pāua divers, until the catch per unit effort was reduced to levels similar to a fished control site. Pāua loggers were used during the experiment to monitor changes in CPUE over a wide range of pāua abundance. Fishing was also carried out at Boat Harbour, a nearby fished site.

Catch per unit effort was measured using the loggers in two different ways: firstly, from the time that divers spent underwater (20 cm or more below the surface), and secondly by summing the half-hour blocks during which fishing activity is recorded on the boat loggers. The first measure was derived from the dive data and would allow effort to be mapped at a fine spatial scale, while the second measure is only dependent on data from the boat loggers. Over the course of the fishdown, a linear relationship between CPUE and cumulative catch was seen for both measures. This is consistent with a constant catchability. Depletion models were fitted to the data, with a separate catchability for each diver. The resulting models fitted the data well, with the c.v.’s of the model fitted using the underwater and fishing-time measures of effort being 0.17 and 0.10, respectively. It is striking that the simpler measure of effort fitted the catch data more closely. During the fishdown, 8 t of pāua were harvested from Fighting Bay. Using the fishing-time measure of effort, the depletion model estimated that there were 3 (2.3 to 3.9) t of harvestable pāua remaining at the end of the experiment.

Exploration of the data shows the power of the loggers to quantify the detailed distribution of the catch. The catch over the course of the experiment was mapped in 10 m² squares, with one of these 0.01 ha squares having a catch of 221 kg. In Fighting Bay, the hectare with the highest catch yielded 4635 kg (the actual yield would have been higher as, due to problems with the dive logger GPS units, only 64% of the catch was spatially mapped). It was expected that during the fishdown divers would move to deeper water. No consistent change in the depth or duration of the dives was seen. Swim distances were variable, and no consistent change in the swim distance associated with dives (either on the surface or underwater) was seen. The other indicator that showed a marked change was shell length-frequency. Harvesting was size selective, with the larger pāua being preferentially harvested. On the first day in Fighting Bay, 2.7% of the shells had a basal length less than 130 mm, and on the final day 28.8% of the shells were less than this threshold.

The Fighting Bay data will be a valuable dataset for the future development of indices of pāua stock status. In this report we have carried out an exploratory analysis. There are more avenues that could be fruitfully explored. The next priority for analysis of the logger data, is to develop indices of spatial movement of the divers. There was evidence during the experiment that divers moved more slowly when catches were higher, and more rapidly when catches were lower. These spatial patterns have not yet been quantified.