Advances in Larval Development and Feed Regime Optimisation for Giant Kōkopu (Galaxias argenteus)

Abstract

Giant kōkopu, Galaxias argenteus, is an amphidromous fish species that is endemic to New Zealand. In its larval form, giant kōkopu are one of five species that are harvested in the whitebait fishery which holds significant social, cultural and economic value in New Zealand. Due to the high price per kilo ($100-160 kg-1 ) and limited supply of whitebait from the wild fishery there is interest in developing whitebait species, including the giant kōkopu, for aquaculture production. However, there is little published literature on the giant kōkopu on which to base their development for aquaculture. Therefore, the research presented in this thesis aims to describe aspects of the morphological development and larval feeding requirements of giant kōkopu to assist in advancing their aquaculture. Morphometric characteristics of larval giant kōkopu that are critical to the success of larviculture were measured over the 77 day production period, from hatch to harvest. The initial slow growth of the larvae appears to be a result of rapid depletion of endogenous energy reserves prior to initiation of first feeding, while the extent of eye pigmentation and the presence of an open mouth upon hatching indicate that feed provision should be brought forward from 3 DAH. Mouth gape width at hatch was 348 µm and increased rapidly in the first four weeks to 651 µm suggesting that larger live feed items could be used as the initial feed, such as instar-II Artemia. At harvest, the larvae varied enormously in size and condition, with larger fish showing early signs of metamorphosis, such as increasing opacity of flesh and colouration as organs. The results confirm that an improved understanding of the developmental biology of the larvae can assist in providing a more effective feeding regime, particularly for determining the timing of first feed provision and the feed particle size requirements throughout larval development. Consequently, heterogenous growth performance is an aspect of production that needs further attention for developing the efficient aquaculture production of this species. A feeding experiment with early stage larval giant kōkopu compared the growth performance and survival of the larvae over a four week period when fed different proportions of instar-I and enriched instar-II Artemia for different durations. Larvae in the treatment group which received the greatest proportion of instar-I Artemia for the longest duration had the lowest mean wet weight and survival, leading to the lower total production at harvest when compared to other feeding treatments. The feed treatment that received only instar-II Artemia (i.e., no instar-I), achieved equal best total productivity. While the treatment fed a short initial duration of mixed instar-I

and instar-II followed by only instar-II Artemia was the best performing, producing larvae with the greatest mass, length, survival and total production. This feed treatment appears to offer the best balance between availability of feed particles in the optimum size ranges due to the initial combination of Artemia instars, with a higher level of digestibility. Observations of larval ability to consume artificial diets across four ages (18, 21, 25 and 28 days after hatching, DAH) at each of the three daily feeds (Morning, Midday, Afternoon), demonstrated a potential to shorten the expensive live feeding period. At 18 DAH little to no artificial feed was consumed at either of the three feeding events throughout the day. At 21 DAH mean gut fullness increased and the number of fish with empty stomachs reduced by the end of the Morning and Afternoon feeding events. Both variables improved further for larvae at 25 and 28 DAH. Formulated feed consumption did not differ between larvae aged 25 and 28 DAH indicating that three days of feeding live feed could be removed from the feed regime. This change would save 13% on the cost of live feed, however, further research is required to determine the potential impacts on growth and survival. Three artificial dry feed treatments (Otohime, Artemac and O.range ) were experimentally tested with their effects on larval giant kōkopu growth performance compared. Larvae fed with Otohime outperformed both Artemac and O.range treatments by achieving the highest wet weight after 67 days, greater by at least 47 % on average than both Artemac and O.range. These differences in larval performance are likely to be due to the higher protein:energy ratio and EPA content of Otohime. High DHA and ARA in the diets in absolute terms or in relation to EPA did not result in added benefit for growth performance. This study provides an important first step in identifying the nutritional needs of larval giant kōkopu which can assist in improving their commercial aquaculture production Collectively the outcomes of this research provide valuable information to assist in improving the development of giant kōkopu aquaculture. The results confirm that an improved understanding of the developmental biology of the larvae can assist in providing a more effective feeding regime, particularly for determining the timing of first feed provision and the feed particle size requirements throughout larval development. Furthermore, significant productivity gains and cost reductions (i.e., up to 27 % live feed cost) are likely to be achieved through the adoption of early shifts to instar-II Artemia and weaning from live food three days earlier than is current

practised in commercial production of giant kōkopu. Overall, the results from this study confirm the benefits of the close evaluation of larval morphology and feeding in improving the performance of the larval culture of fish.