Paradigms for planktonic assemblages: 50 years of contributions from the Leigh Marine Laboratory, Northland, New Zealand

Abstract

Plankton includes the primary producers and consumers that are critical for healthy ecosystem function in the marine realm. My objective was to identify the major contributions of the Leigh Marine Laboratory to our broader understanding of planktonic assemblages. Significant contributions were made prior to 1990 on the ecology of larval fishes. From the 1990s the focus changed to the sensory biology of larval invertebrates and fishes, with a strong emphasis given to the role of reef sound in attracting potential settlers. Both early and post 1980s research has been highly influential in a paradigm shift away from passively drifting larvae that have little control over their fate. Important contributions have also been made on the dynamics of nutrient–phytoplankton interactions, larval development and aquaculture. Opportunities abound for future research on the dynamics of planktonic assemblages in shelf waters and in changing seas.

Keywords:

• plankton
• phytoplankton
• zooplankton
• presettlement fishes
• larval development
• sensory behaviour
• oceanography

Introduction

Plankton is the trophic base of pelagic ecosystems and has a critical role in the cycling of atmospheric CO₂ and important nutrients (Chisholm et al. 2001). Macronutrients and trace elements support autotrophs that range in size from picoplankton to large diatoms and dinoflagellates (e.g. Finkel et al. 2010). There is considerable recycling of nutrients through the microbial loop (Fenchel 1988). Heterotrophs depend on autotrophs where energy and nutrients are passed through multiple levels of consumers from tiny organisms, that are part autotroph/part heterotroph, to obligate heterotrophs that include microplankton, meso and macroplankton as well as larger nekton (i.e. planktivorous fishes and invertebrates to macrofauna such as tuna, sharks and whales). Some plankton spend their entire life history in the plankton (sensu holoplankton), while some are there only as larvae (sensu meroplankton). Most organisms that inhabit soft and hard benthic environments as juveniles and adults have planktonic life stages. In temperate reef environments, for example, ‘habitat formers’ such as kelp plants have spores which disperse in the plankton (habitats classification, sensu Jones & Andrew 1993), and ‘habitat determiners’ or ‘habitat responders’ such as sea urchins, ascidians and many fishes have larvae and spend hours to weeks in the plankton as larvae. Knowledge of life cycles is critical to understanding population dynamics, and this is exemplified by jellyfishes that have benthic and pelagic components to their life history with polypoid benthic stages and pelagic medusoid stages. Moreover, they have sexual reproduction and multiple modes of sexual reproduction (Arai 1997). A major paradigm in plankton biology is that most larvae die before they have completed a larval phase, and estimates of 5%–10% mortality per day are common (Cushing 1975); 50% of larvae of some species are lost before the yolk is absorbed (Fortier & Leggett 1984, 1985). Historically, key questions in larval studies are: how do larvae disperse; how are they distributed both horizontally and vertically; and are they just passive particles?

My objective is to review research on planktonic assemblages carried out at the Leigh Marine Laboratory (LML) in the broad context of the acquisition of knowledge and contributions to paradigm change.

It was not my intention to list all pelagic paradigms, as that would be beyond the scope of a short review. I have, therefore, focused on areas that have been addressed at least in part at the LML, and for some there are clear and significant contributions to paradigm shifts. The approach I have taken is to give a brief background to the initial building blocks of plankton research at the LML; here, I reflect on the strong influence of quantitative benthic ecology at Leigh. A summary of papers of high impact follows. My measure of the LML's impact is based on citations by publication and the contributions of clusters of papers toward major paradigms of planktonic assemblages in pelagic ecosystems. Focus is then given to LML and related international contributions in the following areas:

•	innovative methods and sampling designs for studying plankton, and an increasing appreciation of the physical drivers of plankton distributions with an emphasis on fine-scale features (e.g. internal waves);
•	descriptions of the development of larvae, and their condition and growth, both from rearing and field-related studies. Some of this work was carried out in the context of aquaculture and overlaps with the following area of research:
•	an enhanced appreciation that fish larvae, in particular, are not passive particles has come from a greater knowledge of larval sensory capabilities and swimming abilities. Daily increments in otoliths have also allowed estimates of the time spent in the plankton; and
•	a broader understanding of bottom-up and top-down phenomena based on the concurrent studies of the continental shelf of Northland, New Zealand and adjacent to the LML. Here, I comment on contemporaneous studies by other research groups in Northland (e.g. the National Institute for Water and Atmospheric Research [NIWA]).

Finally, I comment on growing contributions and opportunities for future research on plankton, focusing on developing areas such as: ecophysiological traits of phytoplankton and our understanding and predictions of pelagic food chains in warming and increasingly acid seas; expanding on historical platforms; collaboration with other institutions (e.g. NIWA) and the vulnerability of losing specialist skills required for the study of plankton, especially taxonomy.

Background

The LML was built in 1962, located on the northeastern side of New Zealand in temperate to subtropical waters (36°16 S; 174°47 E). Located at the boundary of the inner and outer Hauraki Gulf, Leigh is about 70 km from the shelf break (Fig. 1). Early research on plankton at Leigh, and by Fisheries research, initially focused on the identification of holoplankton (phytoplankton; e.g. Taylor 1974a, b) and meroplankton (e.g. Cassie 1956) (Table 1). Jillett (1971) published a memoir on seasonal variation in plankton abundance and physical conditions by season (e.g. temperature) in the Hauraki Gulf and Jellicoe Channel near Leigh.

Figure 1  Map of coastal waters and the shelf break within c. 100 km of LML on the east coast of New Zealand. The star indicates the Jellicoe Channel site used by Jillett (1971) in his plenary work on seasonality of plankton. The majority of work on plankton took place near to the LML, some studies extended to Great Barrier Island and work on the interactions between planktivorous reef fishes and plankton was done at the Poor Knights Islands. Research by New Zealand Fisheries and NIWA (especially by J Zeldis) and others has encompassed the Hauraki Gulf and outer Hauraki Gulf (north of the star). Oceanography of the Gulf is dominated by wind effects and periodic flows of the East Auckland Current (Zeldis et al. 2004). The Okakari Point to Cape Rodney Leigh Marine Reserve extends ~2.5 km either side of LML.

Table 1  Examples of studies on the identification and distribution of plankton with a brief explanation of the focus of the study, location and the sampling design.

References	Topic	Group	Design/task
Cassie 1956	Early development of snapper	Fish larvae	Larval description
Taylor 1970	Diatoms	Phytoplankton	Taxonomy, Goat Island Bay, Leigh; two stations
Jillett 1971	Zooplankton and hydrology	Zooplankton	Abundance, Hauraki Gulf; two stations regularly sampled over 2 years
Baker 1972	Reproduction of New Zealand pilchards	Fish larvae	Abundance, Marlborough Sounds–Tasman Bay area, haphazard stations, 3 years
Taylor 1973	Phytoplankton and nutrients	Phytoplankton	Taxonomy, Goat Island, Beach, Leigh
Taylor 1974a	Dinoflagellates and diatoms	Phytoplankton	Check list
Taylor 1974b	Microalgae other than …	Phytoplankton	Check list
Robertson 1975	Key to eggs	Fish larvae	Multiple sources around New Zealand
Taylor 1978	Phytoplankton records	Phytoplankton	Taxonomy, Goat Island Bay, Leigh, two stations
Taylor & Durbin 1978	Phytoplankton records	Phytoplankton	Taxonomy, four stations (Omaha Bay; Whangateau Harbour), 3 years
Crossland 1980	Snapper abundance	Fish larvae	Hauraki Gulf, spring–summer tows over 3 years at grid stations
Taylor 1981	Phytoplankton and nutrients	Phytoplankton	Abundance, Goat Island Bay; two stations; 4 years
Crossland 1981	Hauraki fish eggs and larvae	Fish larvae	Hauraki Gulf, spring–summer tows over 3 years at grid stations
Maddock & Taylor 1984	Phytoplankton count stats	Phytoplankton	Hauraki Gulf, single site, temporal abundance, 5 years, multiple depths

The identification of many planktonic groups in New Zealand was based on publications from Australia (e.g. Dakin & Colefax 1940) and Europe (e.g. Rose 1970), with a small number of local descriptions on specific taxa (e.g. Robertson 1975; Bradford-Grieve & Jillett 1980; Bradford et al. 1983). A series of cruises by Baker (1972) and Crossland (1980, 1981, 1982) were completed in the inner and sometimes outer Hauraki Gulf. Samples were generally taken at grid stations, or more haphazardly located stations, and data were presented on the egg and larval abundance of multiple species of fish. In some cases, larval descriptions were made and these were augmented by a number of unpublished guides and theses (Kingsford 1985; review and illustrations in Kingsford 1988).

The focus of research and impact

The research reputation of the LML has generally focused on the ecology of rocky intertidal and subtidal reefs (e.g. Ayling 1981; Andrew & Choat 1982), and research that was unrelated to plankton predominated. This quantitative approach to ecology influenced the approach that was taken to study plankton and the types of questions that were asked. For example, while Choat and Kingett (1982) estimated the influence of predation by small snapper (Pagrus auratus) on the algal invertebrate fauna a similar type of study was done on planktivorous fishes and plankton by Kingsford and MacDiarmid (1988). In this case it was demonstrated that large aggregations of planktivorous fishes at the Poor Knights Islands (Figs. 1, 2) had a measurable effect on the abundance and composition of mesoplankton. Plankton was more abundant upcurrent of aggregations of fish than within or downstream of aggregations. Sampling at the same locations at night did not detect low concentrations of zooplankton where diurnal planktivores had earlier fed.

Figure 2  Planktivorous fish, Caprodon longimanus, feeding in Northern Arch, Poor Knights Islands. Kingsford and MacDiarmid (1988) demonstrated that large aggregations of fish could reduce densities of plankton by up to 5×that of areas without fish. Photo: MJ Kingsford.

The ability to work on plankton was partly limited by the size of vessels at the disposal of staff and students. Unless collaborations were done with Fisheries, or the Oceanographic Institute of New Zealand (e.g. Foster & Battaerd 1985), sampling was undertaken from the shore (e.g. Martin & Foster 1986) or from small boats (Taylor & Taylor 1985). During the late 1970s the only boats were 3.7 m dinghies (e.g. Roper 1986). The capability to study plankton was enhanced with a 5 m Avon inflatable that was used to deploy a plankton mesh purse seine net (Kingsford & Choat 1985). Although modest in size it gave researchers an opportunity for a ‘between the grid lines’ approach (Table 2; e.g. the study of drifting algae [Kingsford 1992] and how it can affect the distribution of meroplankton). The Avon was also used to investigate the influence of small-scale oceanographic features (e.g. the slicks of internal waves) on the distribution and movements of plankton (Kingsford & Choat 1986). In 1982 the 12 m R. V. Proteus was launched and this allowed larger ichthyoplankton nets to be towed to the side of the vessel (Kingsford 1988; Tricklebank 2000) or oblique tows to depths of up to 60 m (Kingsford & Choat 1989). The most recent vessel, the 14.9 m R. V. Hawere, was launched in 2001 as a multi-task platform that could give quick access to the offshore islands.

Table 2  Examples of between the grid lines research from LML.

Reference	Title	Group	Design/task
Kingsford 1985	Eggs and larvae of Chromis dispilus	Larval fish	Description and egg hatching
Kingsford & Choat 1985	Drift algae fauna	Larval fish	Abundance, experiments
MacDiarmid 1985	Sunrise release of larvae from lobster	Crustaceans	Vertical tows
Kingsford & Choat 1986	Surface slicks	Larval fish	Oceanography and larvae
Kingsford & Milicich 1987	Presettlement phase of Parika scaber	Larval fish	Abundance, development, growth
Kingsford & MacDiarmid 1988	Planktivorous reef fish and zooplankton	Larval fish	Impact of planktivores
Kingsford & Choat 1989	Horizontal distribution patterns	Larval fish	Cross shelf distributions, visual counts of larvae
Kingsford & Tricklebank 1991	Ontogeny and behaviour of mullet	Larval fish	Behaviour
Kingsford 1992	Drift algae and small fish	Larval fish	Abundance, experiments
Tricklebank et al. 1992	Neustonic ichthyoplankton	Larval fish	Tows, distance from shore

The output of publications and the type of research at LML has greatly changed with time. From 1966 to 1970 only 19 papers were published; but, by the early 2000s, there were over 100 papers per 5-year period (Fig. 3). Publications on plankton made up 6%–23% of total publications in a 5-year period. The greatest contributions have been made in the areas of the ecology of larval fishes, and the sensory biology and development of fishes (Fig. 4). A classification of publications into broad categories before and after 1990 demonstrated great changes in the type of plankton research that was done over the 50 year history of the LML. The proportion of papers on the ecology of fish larvae dropped, and research on the sensory biology of invertebrate and vertebrate larvae, and on aquaculture increased after 1990. Research on invertebrate larvae (e.g. crab megalopae) and experimental work increased with time, and there was greater engagement of staff and students in tropical research that was seeded from LML, especially on the Great Barrier Reef (GBR). This phase-shift in emphasis created new opportunities but, in some cases, bypassed the original research building blocks. The substantial ‘other’ category in 1966 to 1990 included general plankton studies, taxonomy and other minor contributions.

Figure 3  The contribution of plankton-related publications to the total LML publications from 1966 to 2012 (5 year bins).

Figure 4  Research categories covered by plankton publications 1966 to 1990, and 1991 to 2012; a single paper could score more than once by category.

The impact of papers has varied greatly (Fig. 5). Not unexpectedly, taxonomy papers were generally poorly cited. The two most cited papers on plankton are those of Kingsford and Choat (1985, 1986); the first, an experimental study on drift algae, presettlement fishes and invertebrates, and the second on the influence of the surface slicks of internal waves on the distribution and transport of presettlement fishes and plankton. With the exception of Babcock's work on coral spawning in Western Australia, the papers fall into three general categories: reviews; presettlement fishes; and invertebrate distributions as influenced by drift algae and oceanography and sensory biology of larvae. Kingsford (1988) reviewed research on the early life history of fishes, Montgomery et al. (2001, 2006) on active habitat selection by presettlement fishes and Finkel et al. (2010) on phytoplankton in a changing world. Papers of high impact for sensory biology were Tolimieri et al. (2000) with their ground-breaking work on sound as a cue for navigation. Simpson et al. (2005) consolidated this position of leadership with similar work on a tropical setting – Lizard Island (GBR). Although some of the old papers are still getting four to five citations per year, some more recent papers (date, cites per year) greatly exceed this rate as follows: Finkel et al. (2010; 26.5); Simpson et al. (2005; 10.1).

Figure 5  A, Total citations by year for LML plankton publications. B, Average number of citations by year for LML plankton-related publications. Source: Web of Knowledge, November 2012.

External benchmarking with the other major marine laboratory in New Zealand, Portobello Marine Laboratory (University of Otago) indicated that LML has been competitive with respect to publications on plankton. Portobello had maximum plankton-related outputs of about 15 papers within a 5-year period (especially relationships between oceanography and plankton and aquaculture/developmental biology). Most cited papers include (date, total citations); Wing et al. 1995a; 138, 1995b; 145) on the transport and settlement of benthic invertebrates, Zeldis and Jillett's (1982; 89) aerial views of Munida gregaria aggregation in fronts and internal waves, McGehee (1998; 78) on acoustic scattering layers; Byrne and Barker (1991; 52), Lamare and Barker (1999; 57) on larval development. In contrast to the LML metrics I could not find any high impact reviews from Portobello.

Innovation in approaches to sampling plankton and oceanographic features

Innovation in methodology and approaches to the study of plankton has resulted in paradigm shifts worldwide. Direct observations of gelatinous zooplankton (Hamner et al. 1975) and aggregate material (e.g. Silver et al. 1978; Alldredge & Gotschalk 1989) greatly increased our understanding of the role of delicate zooplankton and aggregate material that would have previously gathered as an amorphous mass in the cod-end of plankton nets. Towed particle counters (e.g. Suthers et al. 2004) and video-based identification have allowed continuous data (rather than discrete data from nets) to be collected among water masses and clear oceanographic features such as gyres. Some methods of antiquity prevail. For example, continuous plankton records from Longhurst Hardy Plankton Recorders (LHPR) are towed on ships of opportunity and have been since the 1930s. As a result they have been the sentinels of climate change (e.g. Richardson & Schoeman 2004).

New approaches from LML contributed to changes in our view of larval distributions and behaviour. For example, the presence of larval aggregations in complex near-shore environments, inaccessible to nets, were identified by direct observation on snorkel or scuba (Kingsford & Choat 1989; also see Brietburg 1991). These studies suggested that some taxa may not disperse far from natal reefs and/or could have a prolonged phase as a ‘potential settler’. Avoidance of plankton nets by agile plankton such as presettlement fishes blinkered our view of what was happening in the plankton. Light traps (Doherty 1987), purse seines made of plankton mesh (Leigh; Kingsford & Choat 1986) and very large multiple opening closing nets (MOCNESS; Huebert et al. 2011) allowed a greater level of accuracy in determining temporal and spatial variation in the abundance of late stage presettlement reef fishes.

Spatial scales of variation in plankton abundance have been the essence of pelagic ecology, and there are strong relationships between oceanographic features of different spatial scales and the distribution of plankton (reviews in Haury et al. 1978; Kingsford & Murdoch 1998; Mann & Lazier 2003). A hierarchical understanding of oceanographic features and their relative importance for influencing plankton distributions is essential. Small-scale phenomena such as internal waves, wind driven circulations and topographic effects can be of great biological importance for determining distributions and for transport (Zeldis & Jillett 1982; Pineda 1991; Pineda et al. 2007). Stratified sampling in and out of the slicks of internal waves near Leigh demonstrated aggregation and the movement of larvae toward coastal environments (Kingsford & Choat 1986). Further, variation in the accumulation of drifting macroalgae and associated fishes near shore was attributed to wind and transport by internal waves (Kingsford 1992). Data on the velocity of internal waves was presented by Kingsford and Choat (1986) and further data on internal tide activity was provided by Sharples et al. (2001).

Larval development and aquaculture

There have been significant contributions from LML in the areas of aquaculture and larval development (Table 3). Research on Pagrus (=Chrysophrys) auratus (Sparidae) has included larval identification, larval development (including senses) and early growth (Pankhurst et al. 1991; Pankhurst 1994). Important work has also been done on the factors influencing first feeding in kingfish larvae (e.g. Carton 2005; Carton & Vaughan 2010), digestion and growth in lobster larvae (Simon & Jeffs 2011) as well as the larvae of green lipped mussels and patterns of settlement (e.g. Fitzpatrick et al. 2011) (Table 3).

Table 3  Examples of research on aquaculture from the LML aquaculture; measures of impact are also provided.

Reference	Short title	Number citations	Citations per year	Group	Experimental	Settlement/Recruitment	Larval aquaculture	Oceanography and plankton
Pankhurst et al. 1991	Growth etc of larval snapper	24	1.14	Fish	Yes		Yes
Kingsford & Atkinson 1994	Otoliths and scales of Pagrus auratus	11	0.61	Fish	Yes		Yes	Yes
Alfaro et al. 2004	Bottom drifting algae/mussel spat	18	2.3	Shellfish	Yes	Yes
Carton 2005	Impact of light intensity on kingfish	12	1.7	Fish	Yes		Yes
Gribben 2005	Razor clam development and spawning	2	0.3	Shellfish
Alfaro et al. 2010	Beach cast green-lipped mussel spat	18	9	Shellfish		Yes		Yes
Carton & Vaughan 2010	Kingfish larvae	0	0	Fish	Yes
Gribben et al. 2011	Settlement of green lipped mussels	5	5	Shellfish		Yes
Simon & Jeffs 2011	Diet of early stage spiny lobsters	0	0	Crustaceans	Yes		Yes
Fitzpatrick et al. 2011	Green-lipped mussel larvae	0	0	Shellfish	Yes		Yes

Assessments of holoplankton and meroplankton quality have had significant impacts on the questions that could be addressed concerning the feeding environments that plankton encounter. In short, the health of larvae will influence survival rates (Purcell 1985). The condition of larvae and relative growth rates have been measured using morphometrics, total lipid and classes of lipid (e.g. Fraser 1989; Westerman & Holt 1994) as well as RNA:DNA ratios and the widths of daily increments in otoliths (Suthers et al. 1989); the latter is used as a proxy for growth. There are few studies from LML that have focused on quality-related assessments of larvae. The spacing of increments was assumed to be a proxy for growth in presettlement monacanthids by Kingsford and Milicich (1987) and this assumption was tested experimentally by Milicich and Choat (1992). Sim-Smith et al. (2012) used increment spacing as a proxy of growth and found that snapper that grew fast as larvae and had a short planktonic larval duration (PLD) of ≤20 d generally grew faster as settled juveniles. Kingsford and Atkinson (1994) suggested that circuli in the scales of recently settled snapper could provide information on growth late in the presettlement phase and be a predictor of behavioural changes of larvae. Lipids (especially triglyceride) were used as a measure of condition of snapper larvae in the unpublished thesis of Atkinson (1987).

How long are fish larvae in the plankton?

The end of the larval stage for pelagic species is when they metamorphose into the form of a juvenile. In contrast, in the case of taxa that are benthic as juveniles and adults, the event of ‘settlement’ (adapted from the study of invertebrate larvae, e.g. Rodriguez et al. 1993) marks the end of the presettlement (sensu Kingsford & Milicich 1987) or planktonic phase. The duration of this phase has been referred to as the PLD (Victor 1991). In the case of fish, this can be estimated from daily increments in otoliths and a change in the spacing of increments that coincides with the time of settlement. Validation that increments are deposited daily in otoliths has been demonstrated for pelagic juvenile monacanthids (Meuschenia (=Parika) scaber; Kingsford & Milicich 1987) and larval and juvenile sparids (P. auratus; Kingsford & Atkinson 1994). PLDs were estimated as c. 22–44 d in M. scaber and c. 20–30 d in P. auratus (Sim-Smith et al. 2012).

Larvae are not passive particles: major international paradigm shift

The primary contributions to global paradigm shifts have been in the areas of larval reef fish biology and ecology, and the sensory biology of fish larvae and invertebrates. Historically, reefs were considered to be ‘open systems’ where larvae were thought to disperse widely and there was little likelihood of larvae returning to the reef from which they were spawned (=natal reef). This view was partly based on a lack of research on the larvae of reef fishes during the 1960s and 70s. Comprehensive work had been done on high latitude larvae by Blaxter (1986) and others. The larvae of clupeids, gadiformes and pleuronectids generally developed slowly and took many days to gain notable swimming abilities, to see further than their own body length, and to gain resilience to starvation. Small wonder that, despite early contributions on the distributions of tropical larval reef fishes (e.g. Leis & Goldman 1984), models that predicted dispersal and likely locations for settlement treated larvae as passive particles in mainstream currents (e.g. Williams et al. 1984). The mismatch between what was caught in nets, and the size and form of new settlers arriving on reefs was clear.

The term ‘larvae’ implied poorly-formed fish of limited swimming abilities. This did not match what was caught off Leigh (i.e. large pelagic juveniles of fishes such as monacanthids, tripterygiids and mullids that looked very similar to the fishes caught on reefs). The use of plankton-meshed purse nets (Kingsford & Choat 1985; Kingsford & Murdoch 1998) to encircle drift algae, and stratified sampling in and out of convergence zones resulted in catches of these fishes. In parallel with this work, light traps that were pioneered in Australia (Doherty 1987) caught pelagic juveniles of many taxa that resembled the form of newly-settled juveniles.

It was also clear from underwater observations that some larvae, such as tripterygiids, may spend a considerable part of their presettlement phase near suitable settlement habitat (Kingsford & Choat 1989). High catches of tripterygiid larvae have been obtained kilometres from shore (e.g. Kingsford & Choat 1986; Tricklebank et al. 1992). These larvae have a PLD of about three months and if they encounter suitable habitat during this period then they have the swimming abilities to stay close to shore. Larvae of the Gobiesocidae were also found very near shore (in less than 2 m of water) and given the small size difference between the size at which they hatch and settlement it is likely they never move far from their natal reefs (Kingsford & Choat 1989).

The larvae of different species were clearly doing different things (Kingsford 1988). The monacanthid Meuschenia scaber was found throughout the water column as larvae. Once they reached a juvenile form they were attracted to floating projects such as drift algae in surface waters. Mullids (Upenichthys lineatus) were found in surface waters as larvae and pelagic juveniles. Moreover, aggregations of juveniles were often found around drift algae. In contrast, snapper (Pagrus auratus) were found throughout the water column through development and may be found in concentrations where there are clear thermoclines, while the aforementioned gobiesocids probably don't move far from natal reefs. Although swimming speeds of these taxa were not estimated it was clear from observing larvae swimming near reefs or in purse seines that they have strong swimming abilities.

A series of papers on tropical reef fishes in Australia (Stobutzki & Bellwood 1994, 1997; Fisher & Bellwood 2001; Fisher & Wilson 2004) demonstrated that presettlement fishes could swim for long periods of time in waters that were flowing at 35 cm/s. Great differences in performance were found among taxa, and swimming speed increased greatly with size of fish, especially after flexion of the notochord to form a caudal fin and while fish were in the form of a pelagic juvenile. Swimming was one part of the discovery process, but what senses determine the direction of swimming? Directional swimming had been demonstrated in salmon and other taxa, but the literature had historically focused on juveniles and adults (review in Kingsford et al. 2002). A range of senses had been suggested including the detection of smell, sound, vibration, visual cues, magnetism, acceleration, electricity and sun compass; many of which influence the orientation of birds and other taxa.

Contributions from LML have been pivotal in terms of demonstrating how presettlement fishes and invertebrates can respond to environmental stimuli, particularly sound (Table 4). Tolimieri et al. (2000) used light traps, some that emitted sounds (recordings of reef noises), while others were controls. Tripterygiids, in particular, were about four times as likely to be attracted to traps with sound compared to those without. This discovery was then tested in the tropics by Simpson et al. (2005), who demonstrated that damselfishes (Pomacentridae) and cardinalfishes (Apogonidae) were more likely to be attracted to high frequency or low frequency noises than to controls. In situ measurements of the distances that reef sounds could potentially attract presettlement organisms to reefs were calculated (Radford et al. 2011); estimates of up to 4 km can allow modellers to include larval behaviour in their models.

Table 4  Examples of investigations on the sensory biology and behaviour of larvae.

Reference	Short title	Number of citations	Citations per year	Group	Field expt	Lab expt	Sense	Attraction to sense	Notes
Kingsford & Tricklebank 1991	Ontogeny and behaviour of mullet	5	0.24	Juvenile fish	Yes		Disturbance	Positive
Pankhurst 1994	Age-related changes in the visual acuity of larvae	24	1.33	Juvenile fish		Yes	Visual
Tolimieri et al. 2000	Ambient sound as a cue	81	6.75	Juvenile fish	Yes		Sound	Positive	Light traps with speaker
Montgomery et al. 2001	Habitat selection by presettlement reef fishes	59	5.36	Juvenile fish					Review
Baker & Montgomery 2001	Species specific attraction of kōkopu	22	2.00	Juvenile fish		Yes	Olfactory	Positive	Whitebait pheromones
Tolimieri et al. 2002	Ambient sound as a cue	28	2.80	Juvenile fish	Yes		Sound	Positive
Jeffs et al. 2003	Crabs on cue for the coast	20	2.22	Crabs	Yes		Sound	Positive
Montgomery et al. 2006	Sound as an orientation cue	53	8.83	Fish and crustaceans					Review
Mishra et al. 2007	Eye structure of rock lobster phyllosoma	5	1.00	Crustaceans			Visual		Histological
Stanley et al. 2010	Induction of settlement in crab megalopae	10	5.00	Crabs		Yes	Sound	Positive	Temperate and tropical
Stanley et al. 2011	Crabs and noise	3	3.00	Crabs		Yes	Sound	Positive
Wilkens et al. 2012	Vessel noise, mussel larvae and settlement	0	0.00	Mussels		Yes	Sound	Positive

Jeffs et al. (2003) and Stanley et al. (2010, 2011) demonstrated that crab megalopae are also attracted to sound, and that favourable sound can speed up metamorphosis into the form of a juvenile crab. Most recently, sound has been shown to influence the settlement of mussels (Wilkens et al. 2012). Other significant contributions from LML have been descriptions of changes in the visual capabilities of both invertebrate larvae (Mishra et al. 2007) and fish larvae (Pankhurst 1994). Furthermore, Radford et al. (2012) have demonstrated that larval snapper can use olfactory cues to orient to suitable settlement habitats such as sea grass beds. The next important step in this research will be to determine the importance of sound relative to other cues such as smell, sun compass and others. Furthermore, how does the importance of different senses change with distance from the reef? For example, presettlement forms may be attracted to reefs by sound and olfaction, but the choice of settlement site can depend on the olfactory cues from conspecifics and habitat (Kingsford et al. 2002).

Oceanography and ‘bottom-up’ and ‘top-down’ phenomena on the continental shelf

Pelagic ecosystems are often viewed as ‘bottom-up’ in that the supply of nutrients influences biomass of primary producers, which in turn determines biomass of primary, secondary and tertiary consumers (i.e. from microscopic grazers to the charismatic megafauna such as whales). However, there can also be significant ‘top-down effects’. For example, high densities of predatory cnidarian jellyfishes can have a great effect on small grazers, which in turn affects abundance of producers (e.g. Pitt et al. 2007). The grazing activity of salps can also have a great influence on the abundance of microscopic prey (Zeldis et al. 1995). Gelatinous zooplankton are considered to be a major causal factor in ‘phase shifts’ of pelagic ecosystems in many parts of the world (e.g. west coast of North America, Brodeur et al. 2002; Africa, Lynam et al. 2006).

Production in the photic zone has a great impact on the benthos because dead plankton, faeces and aggregations of material (sensu ‘marine snow’; Silver et al. 1978) sink to the substratum. This component of ‘benthic-pelagic coupling’ provides deposit feeders with nutrition, in turn supporting a range of other consumers (Graf 1989). Other important aspects of benthic–pelagic coupling include biochemical coupling of nutrients from the water column (Marcus & Boero 1998; Zeldis 2004), resting phases of planktonic organisms that spend part of their life on the substratum and mortality events on the benthos from plankton blooms (Taylor et al. 1985). Furthermore, benthic organisms also have links with the water column through suspension feeding and/or larvae that are planktonic. The survival of larvae can be highly dependent on pelagic processes be they bottom-up or top-down. For example, variation in recruitment and subsequent population size may be determined by upwelling-related variation in feeding conditions for larvae (e.g. Navarette et al. 2005). In contrast, direct predation on larvae (Moller 1984), or competition for suitable larval food (Zeldis et al. 1995, 2005; Lynam et al. 2005), are top-down planktonic processes that can influence survival to settlement out of the plankton.

There have been few contributions from LML on shelf-wide oceanographic processes and variation in plankton dynamics as well as benthic–pelagic coupling; this has been the domain of NIWA. In a series of papers it was demonstrated that warm incursions by the East Auckland Current (EAC) and upwelling events occur on the shelf late in spring and leading into summer (Sharples 1997; Sharples & Greig 1998). Between early and late summer upwelling generally ceases, and there was a transition to downwelling, due to a combination of forcing by local winds and the influence of the EAC (Zeldis et al. 2004). Upwelling caused extensive loading of the inner shelf and Hauraki Gulf bottom waters with nitrate. However, the system moved towards a net heterotrophic system where it was concluded that remineralisation on the shelf and regenerated nutrient supply was important for sustaining stocks of chlorophyll a (Zeldis 2004; Bury et al. 2012). Furthermore, calculations on the percentage of primary production that was remineralised in sediments of the northern shelf-slope region, Hauraki Gulf and Firth of Thames (13%–34%) were provided by Giles et al. (2007).

Zeldis also published a number of important papers that demonstrated the importance of upwelling and abundance of gelatinous zooplankters on the abundance of snapper larvae (P. auratus). The larvae of this important species suffered highest mortality rates when salps were abundant. They altered the distribution of potential prey or competed for suitable larval food (Zeldis et al. 1995, 2005). Zeldis and Francis (1998) also used plankton surveys (i.e. the daily egg production method) to estimate the biomass of snapper in the Hauraki Gulf.

Phytoplankton: an area of growth in research

Research on phytoplankton moved from identification to nutrient uptake kinetics (Rees & Allison 2006; Rees 2007; Beardall et al. 2009; Flynn et al. 2010). This has received increased attention due to potential climate-induced changes to marine food chains (Finkel et al. 2010). In this well-cited review, it was stated that ecophysiological traits in phytoplankton can be used to predict phytoplankton growth rates as well as chemical recycling and changed interactions between producers and consumers and loss processes. This in turn can vary with changes in key environmental factors. There is great potential for the LML to contribute to paradigm shifts in this area.

Other contributions and opportunities

With a focus on sensory biology over the last 10 to 15 years, many larger-scale questions have not been addressed. For example, there are great differences in the distribution and abundance of planktivorous fishes across the continental shelf (Kingsford 1989). Offshore islands have a fauna dominated by Chromis dispilus and other taxa such as Caprodon longimanus and Callanthias allporti, whereas Scorpis lineolata is abundant near the mainland. These patterns are reminiscent of debates that have been had in Australia on the factors determining cross-shelf distributions of fish such as: oceanography; plankton abundance and composition; turbidity; and larval connectivity. Benthic feeding fishes also show great variation in abundance across the shelf (e.g. Notolabrus celidotus, Pseudolabrus luculentus, Parma alboscapularis) and new tools for addressing issues, such as connectivity, are available and can provide invaluable information to marine managers (Jones et al. 2009; Jones 2013).

Marine laboratories generally have many theses with valuable unpublished information; LML is no exception. Cole (1987) worked on the depth stratification of pilchard larvae while Coupe (1993) studied plankton in the Leigh Marine Reserve and Parr (1994) worked on bivalve larvae in Whangateau Harbour. Despite Jillett's (1971) major contribution on the seasonality of plankton there are few other published data. The theses of Kingsford (1980) and MacDiarmid (1981) contain 18 months of valuable plankton data (their seasonal data on Penilia avirostris are graphed in Kingsford & Murdoch 1998). Thompson (1983) described seasonal variation in fish larvae near Goat Island. Although seasonal plankton data appear pedestrian, they are critical for understanding plankton trophodynamics at different times of the year. The importance of baselines of plankton composition and abundance is now considered critical in times of changing water temperatures, currents, ocean pH and oxygen profiles (Richardson & Schoeman 2004; Brierley & Kingsford 2009).

In conclusion, research at the LML has contributed to paradigm shifts in a few discrete areas, but opportunities abound. Aquaculture and research on uptake dynamics of phytoplankton shows promise. The impact of climate change on phytoplankton and the cascade effects that cause broader impacts on marine food chains is a sensible area to progress (e.g. Finkel et al. 2010). Patterns that were found pre-1990s have not been pursued. Our knowledge of the senses that larvae use prior to settlement has increased greatly. However, multifactorial experiments are required for a more integrated understanding of the spatial and temporal hierarchy of senses used. New tools (e.g. intergenerational tagging and improved genetics) can allow old questions, such as larval connectivity, to be addressed. Research on cross-shelf phenomena have been in the NIWA domain and value could have been added with more collaborative ventures; the Auckland/NIWA Joint Graduate School in Coastal and Marine Science should help. The rich broth of deep-sea plankton that rises from the shelf break beckons for both taxonomic and ecological investigation.

Acknowledgements

I would like to thank Mark O'Callaghan for assisting in the data mining required to write this review. Furthermore, thanks to Kendall Clements and John Montgomery for organising the collection of reviews in this special issue. I apologise to the authors of any papers that have been missed, but I hope you agree with the thesis of this review. Thanks to John Zeldis and an anonymous referee for helpful comments on the manuscript. I believe telling the stories as we have done reminds us of how far we have come and will realise new opportunities for research. The LML has been a special place for inspiration, camaraderie, unique characters, life-long friendships and opportunity; it has been a privilege to be a part of the Leigh story.