The Namibian orange roughy fishery: lessons learned for future management

B. Oelofsen^[332] and A. Staby^[333]

1. HISTORY OF THE NAMIBIAN ORANGE ROUGHY FISHERY

During the pre-independence period of Namibia the South East Atlantic Fisheries Commission (ICSEAF), to which 17 states were party, managed the waters off the coast of Namibia. During this period distant water vessels did most of the fishing and there was a fairly clear-cut specialization by vessels from different countries or groupings of countries in targeting certain species. It so happened that the east block (former Soviet Union) countries concentrated mainly on horse mackerel while the Iberian countries targeted hake using bottom trawls. Having used powerful, large freezer vessels and thus been able to fish at fairly great depths (500 metres plus) their vessels, from time to time landed alfonsino and orange roughy as bycatch. The fact that the larger, more valuable, hake size classes tend to occur in deeper waters naturally enticed these vessels to fish at greater depths increasing the probability of landing orange roughy.

The development of orange roughy, e.g. off New Zealand and Australia and in the north Atlantic, and high-price niche it found in the market, naturally led to an interest in investigating the potential of the orange roughy resource off Namibia. As a result a large Spanish group made a search for orange roughy off Namibia, but with discouraging results. It turned out that this particular search did cover some of the grounds that later on, yielded orange roughy. It is therefore puzzling that the first attempt to establish the presence of an orange roughy resource was unsuccessful.

Soon after independence in 1992 a Namibian-based company approached the Ministry of Fisheries and Marine Resources for permission to fish experimentally for orange roughy. After an agreement was reached the company was given the approval to start fishing. One of the interesting conditions of the agreement was that should the company succeed in finding a resource it would be allocated 50 percent of any future TAC in compensation for putting at risk its capital and for its effort to find the resource. The company, Gendor, did find a viable orange roughy resource and up to now the Ministry has honoured the undertaking by allocating 50 percent of the TAC to them to reward their pioneering work.

The Ministry got involved in the research of the resource in 1995 with the formation of the Deep Water Fisheries Working Group. By mid-1996, four spawning areas had been discovered (Figure 1). The first swept-area and acoustic biomass surveys of three of these four areas were done in 1997. The Ministry, recognizing that the fishery had the potential to become a viable commercial fishery, announced that the experimental phase of the fishery would be terminated. At the same time the then Minister of Fisheries, the Hon. H. Pohamba, invited the public to apply for rights in the orange roughy fishery.

In adjudicating the applications, emphasis was placed on the access to expertise and access to, or possession of, suitable vessels and gear to ensure that the successful applicants would be able to successfully participate in the fishery. Not surprising, therefore, that of the five successful companies, two had ties with companies active in the New Zealand and one with a French company experienced in orange roughy fisheries. This illustrates the importance the Ministry attached to having companies that would have access to the necessary expertise in orange roughy fishing. As it turned out, three companies were granted quotas while two companies were not. The granting of rights to five companies, however, bears testimony of the high expectations that existed in Namibia of the orange roughy fishery becoming a major fishery and money earner.

In 1995 an orange roughy working group was formed with participation by Namibian Resource Management staff and the three companies that received catch quotas. This working group decided to solicit inputs from the stock assessment group of the University of Cape Town in South Africa led by Professor D. Butterworth, and Dr. M. McAllister from the Imperial College of London, because the fisheries research institute in Namibia lacked personnel with sufficient numerical skills to develop suitable stock assessment models. The three orange roughy quota holders created a trust account to which they each contributed on a pro rata basis to the size of their quotas to fund the involvement of the Butterworth group. The Ministry also contributed to the costs and the co-funding removed the potential perception that advice by the Butterworth group could be influenced by the "he who pays the piper" principle. In the working group, management advice to the Ministry was generated that aimed at:

• establishing a proper management regime for the individual fishing grounds and

• putting in place sufficient incentives for the five companies to keep up with exploratory fishing outside the known areas. The known areas were referred to as QMAs (quota management areas).

The TAC that was announced was subdivided and allocated to be fished on the individual grounds. Any fish landed outside the QMAs were not regarded as part of the quotas (and therefore were outside the TAC). Trigger levels were set at 100 t of fish landed from an area outside a QMA after which a company could request the area to be declared an exclusive zone, which would reserve the area exclusively for the finder. After 500 t had been landed the area would be declared a QMA and the finder would be treated preferentially in regard to any quotas allocated for such an area in future. It is therefore clear that ample rewards were put in place to entice companies to spend time on exploratory fishing outside the QMAS. This was done in particular because Monte Carlo simulations indicated there to be up to ten possible fishing grounds worthy of QMA status. To date, however, only four spawning grounds have been found (Figure 1), and the likelihood of more grounds to be discovered in future is slim.

Despite the seemingly excellent incentives that were put in place, the two companies that were not granted quotas made no effort to get involved in the exploratory fishing arguing that "the best grounds would have been found first". For the companies with quotas, to land their quotas during the winter spawning period when the CPUE is high and landing cost per ton is at its lowest, was more important than searching for new grounds. As a result not as much effort as was hoped for was directed to further exploratory fishing.

2. THE HISTORY OF CHANGES IN THE RESOURCE ABUNDANCE

When the first stock assessment was done in 1997 the biomass calculated for the QMAs was close to 300 000 t and had an associated high CPUE. In the following year there was already a reduction in the biomass estimate as indicated by the survey and the CPUEs were starting to fall (Fig. 2). This downward trend continued despite the very conservative approach in management aiming at not fishing the biomass down to less than 50 percent of pristine biomass. The fishing down was planned to occur over 14 years with a "soft landing at the end when the sustainable level was to be reached.

The much reduced biomass estimates obtained by subsequent surveys caused some parties to doubt the accuracy of the 1997 biomass estimate for it was clear that the volume of fish taken out of the estimated 1997 biomass by fishing mortality could not account for the reduction in subsequent biomasses determined by survey.

If however the 1997 biomass estimate was correct and fishing mortality alone could not account for the drop in biomass, alternative hypotheses had to be formulated to explain the phenomenon. The three competing hypotheses were proposed that attempted to explain the observed drop in biomass were:

i. fishing down
ii. disturbance of the fish aggregations and
iii. intermittent spawning behaviour, i.e. fish would congregate intermittently during environ-mentally favourable years to spawn.

To test the competing hypotheses the decision was made to close one of the grounds (Frankies) to fishing in 1999 and to monitor the abundance of fish at that location. As can be seen from Figure 3 in 2002 the biomass resurged and was back almost to the level it was at during the 1997 survey. This result almost definitely ruled out overfishing as the single cause of stock biomass reduction. These grounds are about 120 nm apart and are treated as separate populations although this could not yet be verified by detecting sufficient genetic differences between the populations.

Differentiating between the remaining two hypotheses is not easy. The fact that the 1990 search by the Spanish company did not find orange roughy may be taken as circumstantial evidence that the fish were not aggregated on the spawning grounds during that year. The return of the fish to Frankies after three years of "rest" however rather strongly points to disturbance as being the major causative factor in the much lower biomass estimates before 2002. If the return of the fish to the Frankies was mainly the result of attractive spawning conditions it should also have been evident in higher than the normal presence of fish on the other grounds. A swept area survey of the most southern ground, Johnies showed a two to three times higher biomass estimate than in previous years.

3. BEST MANAGEMENT OPTIONS

Retrospectively, it seems clear that orange roughy protects itself rather effectively from being fished out by either aggregating infrequently or by moving away from frequently trawled grounds. Inevitably, therefore, the CPUE will drop on new grounds after a few seasons. This should therefore not because alarm. It also renders such grounds difficult to mange from the point of view of ensuring that landings are consistent, and profitable. If it turns out that spawning is intermittent and is the major factor causing fluctuations in fish abundance the problem will become more difficult.

In such a scenario a single vessel may monitor the grounds and, if aggregating fish are found, the "make hay whilst the sun shines" principle may be the way to go. This is not a scenario that will be desired by industry for markets will be supplied with fish sporadically and fishing fleets and processing capacity will be idle for years unless it can be used for other fisheries in the meantime. This strategy, for obvious reasons, is therefore not to be seriously considered.

If disturbance, or a mixture of disturbance and intermittent spawning, turns out to be the major cause of biomass reduction, management may be based on the closure of the grounds to allow the fish to re-aggregate. Assuming a rest period of three years, at least four grounds need to be available so that a single ground can be fished each year and rest for three years. In this way relatively high CPUEs can, one hopes, be maintained and a steady, albeit small volume of fish will flow to the markets. An added advantage will be that a small number of vessels or even a single vessel will be sufficient to fish and thus the fishing can be cost effective.

This approach will be possible only once the potential of fishing grounds is known and stock assessment is sufficiently sophisticated to enable the setting of individual TAC’s for the grounds. There is also an additional cost saving advantage in that once a fairly extensive data series on the grounds has been compiled a survey only needs to be done on the ground before it is opened the following season. This assessment approach may be considered, e.g. in the south Indian Ocean once a Regional Fisheries Management Organization has been established for that region. The orange roughy grounds of Namibia and the grounds presently being developed off Chile may also benefit from such an approach.