Search

Highlights • Dissolved iron limits primary production in offshore South Pacific surface waters. • Phytoplankton is elevated in high-iron filaments in mesoscale eddies east of New Zealand. • Iron in these eddies is due to entrainment of high-iron coastal water into offshore water. • The resultant eddy-driven flux of iron supports production in the Subtropical Front. Abstract Subtropical and subantarctic waters either side of the southern hemisphere Subtropical Front are considered iron-limited, suggesting production within the front is dependent on a supply of iron from atmospheric deposition, zonal advection of coastal water, or upwelling. We present the results from a one-day biogeochemical survey in Subtropical Water east of the North Island, New Zealand, in a region where mesoscale cyclonic and anticyclonic eddies entrain chlorophyll in filaments around the eddies. There was no significant relationship between upper mixed layer chlorophyll and any physical or macronutrient quantity. However, chlorophyll was significantly positively correlated with dissolved iron. A simple model suggests that while vertical entrainment of iron into the upper mixed layer occurred, most of the dissolved iron in the eddy was due to entrainment of high-iron coastal water into low-iron offshore Subtropical Water, and that this iron supports primary production in otherwise iron-deficient water. We suggest that a significant component of the total primary production within the Subtropical Front may be determined by mesoscale eddy induced lateral advection of iron.

We present annual cycles of chlorophyll a, phytoplankton carbon, nitrate and oxygen for Subtropical (STW), Subantarctic (SAW), and Subantarctic Mode (SAMW) waters near Aotearoa New Zealand from data collected by two Biogeochemical (BGC) Argo floats. We develop two simple models of depth-integrated net primary production (NPP), tuned against 14C-uptake measurements, to compare with Vertically-Generalised Production Model (VGPM) satellite-based estimates of NPP. One model is the simplest possible, and assumes production is proportional to light multiplied by chlorophyll a concentration. The second model modifies the light response profile to account for photoacclimation. In STW at 30–35°S, enhanced production is initiated in austral autumn when the mixed layer deepens to entrain nutrients into the photic zone. For about half the year, there is substantial production within a deep chlorophyll maximum that sits below the mixed layer. Consequently, depth-integrated NPP is only loosely related to surface biomass as imaged from satellite remote-sensing, and BGC Argo-based model estimates of depth-integrated NPP are about double VGPM estimates. In SAW at 45–55°S, production is initiated when vertical mixing decreases in austral spring. Production is largely within the mixed layer, and depth-integrated phytoplankton biomass and depth-integrated NPP follow surface phytoplankton biomass. Model estimates of depth-integrated NPP based on BGC Argo float profiles are comparable with VGPM estimates for the southern water masses., SI

Aotearoa New Zealand is a marine nation. The islands are 1700 km from the nearest continental land mass, its northern coasts are bathed by warm waters arriving from the tropics, and its southern co...

Satellite-derived estimates of sea surface temperature (SST), chlorophyll (Chl), and fluorescence line height (FLH), from 22 sites ∼20 km offshore from the West Coast of the South Island, New Zeala...

We test the paradigm that in a future warmer ocean, shallower winter mixing will lead to less net primary production (NPP), by investigating whether warming between 2002 and 2018 led to changes in ...

Phylogenetic asymmetry is common throughout the tree of life and results from contrasting patterns of speciation and extinction in the paired descendant lineages of ancestral nodes. On the depauperate side of a node, we find extant ´relict´ taxa that sit atop long, unbranched lineages. Here, we show that a tiny, pale green, inconspicuous and poorly known cicada in the genus Derotettix, endemic to degraded salt-plain habitats in arid regions of central Argentina, is a relict lineage that is sister to all other modern cicadas. Nuclear and mitochondrial phylogenies of cicadas inferred from probe-based genomic hybrid capture data of both target and non-target loci and a morphological cladogram support this hypothesis. We strengthen this conclusion with genomic data from one of the cicada nutritional bacterial endosymbionts, Sulcia, an ancient and obligate endosymbiont of the larger plant-sucking bugs (Auchenorrhyncha) and an important source of maternally inherited phylogenetic data. We establish Derotettiginae subfam. nov. as a new, monogeneric, fifth cicada subfamily, and compile existing and new data on the distribution, ecology and diet of Derotettix. Our consideration of the palaeoenvironmental literature and host-plant phylogenetics allows us to predict what might have led to the relict status of Derotettix over 100 Myr of habitat change in South America. Fil: Simon, Chris. University of Connecticut; Estados Unidos Fil: Gordon, Eric R. L.. University of Connecticut; Estados Unidos Fil: Moulds, M.S.. Australian Museum Research Institute; Australia Fil: Cole, Jeffrey A.. Pasadena City College; Estados Unidos Fil: Haji, Diler. University of Connecticut; Estados Unidos Fil: Lemmon, Alan R.. Florida State University; Estados Unidos Fil: Lemmon, Emily Moriarty. Florida State University; Estados Unidos Fil: Kortyna, Michelle. Florida State University; Estados Unidos Fil: Nazario, Katherine. University of Connecticut; Estados Unidos Fil: Wade, Elizabeth J.. Curry College. Department of Natural Sciences and Mathematics; Estados Unidos. University of Connecticut; Estados Unidos Fil: Meister, Russell C.. University of Connecticut; Estados Unidos Fil: Goemans, Geert. University of Connecticut; Estados Unidos Fil: Chiswell, Stephen M.. National Institute of Water and Atmospheric Research; Nueva Zelanda Fil: Pessacq, Pablo. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Patagonia Norte. Centro de Investigación Esquel de Montaña y Estepa Patagónica. Universidad Nacional de la Patagonia "San Juan Bosco". Centro de Investigación Esquel de Montaña y Estepa Patagónica; Argentina Fil: Veloso, Claudio. Universidad de Chile; Chile Fil: McCutcheon, John P.. University of Montana; Estados Unidos Fil: Lukasik, Piotr. University of Montana; Estados Unidos. Swedish Museum of Natural History. Department of Bioinformatics and Genetics; Suecia

The shelf seas surrounding New Zealand/Aotearoa, a region nominally extending out to the 250 m depth contour, are complex and varied as they sit above the submerged continent of Zealandia. ...

When SST anomalies are defined with respect to a changing baseline and normalised by their 90th percentile, the Tasman Sea is one of the southern hemisphere hotspots of marine heat waves (MHW) and marine cool spells (MCS). There is little evidence that MHW or MCS are increasing in either frequency or intensity, although the duration of MHW has increased from 8 d four decades ago to 26 d now. On average, Tasman Sea MHW/MCS co-occur with MHW/MCS in the Atlantic, Indian, and eastern-Pacific Oceans, in a wavenumber 4 (W4) pattern. Canonical MHW and MCS show they are likely driven by a stalling of the eastward propagation of a W4 atmospheric wave. During MHW, this slow down leads to near-stationary anomalously high and low air pressure areas driving anomalous north-easterly winds over the Tasman Sea. During MCS, similar a slow-down occurs, but shifted by one-half wavelength zonally.

Observations from two research cruises made in 2008 and 2012 to east of New Zealand are put into context with satellite data to contrast and compare surface chlorophyll a evolution in the two years in order to explore mechanisms of phytoplankton bloom development in the southwest Pacific Ocean. In 2008, surface chlorophyll a largely followed the long-term climatological cycle, and 2008 can be considered a canonical year, where the autumn bloom is triggered by increasing vertical mixing at the end of summer and the spring bloom is triggered by decreasing vertical mixing at the end of winter. In contrast, 2012 was anomalous in that there was no autumn bloom, and in early spring there were several periods of sustained increase in surface chlorophyll a that did not become fully developed spring blooms. (In this region, we consider spring blooms to occur when surface chlorophyll a exceeds 0.5 mg m-3). These events can be related to alternating episodes of increased or decreased vertical mixing. The eventual spring bloom in October was driven by increased ocean cooling and wind stress (i.e. increased mixing) and paradoxically was driven by mechanisms considered more appropriate for autumn rather than spring blooms.

We present the results of a combined observational and numerical study to investigate cool plumes of nutrient-rich upwelled water that emanate near the Kahurangi Shoals and extend into Greater Cook Strait. Surface temperature and chlorophyll are mapped using satellite observations to produce surface climatologies, to validate a numerical simulation and to show the utility of using spatial temperature differences as a measure of upwelling. We find upwelling near the Kahurangi Shoals is strongly wind-driven in the weather band. Upwelling occurs at all times of the year, but its surface signature is only visible in summer months. The upwelled nutrient-rich water supports increased primary production compared to surrounding waters, particularly in summer when the water column is more stratified and surrounding surface waters are presumably nutrient depleted.

With the relatively recent development of Global Drifter Program (GDP) drifters that measure the near-surface ocean velocity and Argo floats that can be used to derive both the intermediate-ocean (1000 m) velocity and the mean dynamic height of the surface relative to 1000 dbar, there now exists the opportunity to directly observe the mean velocity decomposition of the ocean. This study computes the mean Ekman velocity by subtracting the mean referenced velocity derived from Argo data from the mean surface velocity derived from GDP data. This Ekman velocity is slightly stronger than previous observations and shows a spatial structure consistent with a vertical eddy diffusivity that is linearly dependent on wind stress. To do this analysis, the author has to deal with the fact that GDP drifters often lose their drogues, and a product of this research is validation of the wind-slip correction applied to GDP drifters that have lost their drogues.

The periodical cicadas ( Magicicada spp.) of eastern North America are subdivided into “broods” or locally synchronized populations that share a common emergence schedule. Twelve extant broods have 17-year life cycles and three have 13-year cycles. Most broods contain three morphologically distinct species. In the late 19th century, Charles Marlatt (1898) published a groundbreaking series of maps and emergence schedules accompanied by a sequential numbering scheme designating the 17-year brood that emerged in 1893 “Brood I” (with other potential 17-year broods numbered up to XVII), and the 13-year brood that emerged in 1894 as “Brood XIX” (with potential broods up to XXX). Later editions of Marlatt's work (1907, 1923) added more distributional records. Marlatt's …

Several hypotheses exist that describe phytoplankton spring blooms in temperate and subpolar oceans: the critical depth, shoaling mixed layer (ML), critical turbulence, onset of stratification and disturbance-recovery hypotheses. These theories appear to be mutually exclusive and none of them describe the annual cycle of phytoplankton biomass. Here, we present a model of the annual cycle in phytoplankton that recognizes that phytoplankton are not always mixed throughout the so-called ML, and that it is important to distinguish between the surface biomass and depth-integrated phytoplankton. Once these important distinctions are made, the annual cycles and blooms in surface and depth-integrated phytoplankton can be described straightforwardly in terms of the physical drivers and biotic responses.

We review the advances in ‘blue water’ physical oceanography of the seas around New Zealand since the last major review in 1985. By 1985, a basic description had been made of the circulation around...

Global Drifter Program (GDP) drifter data and Argo float data from 2006 to 2013 are used to provide quantitative description and comparison of the mean near-surface (15 m) and 1000 m velocity fields in the New Zealand region. These fields are estimated both from bin-averaging the drifter/float velocities and by fitting non-divergent streamfunctions to the velocities. At both levels, our velocity fields are broadly consistent with previous drifter/float estimates, but provide more detail on the local structure of the mean fields. North of the Subantarctic Front there are large differences between the surface and 1000 m flows, with the Tasman Front being largely absent at 1000 m. South of the Subantarctic Front, where the flow is expected to be more barotropic, there is a closer correspondence between the flows at both levels. Several previously undocumented eddies in the mean flows at both levels are described.

National Institute of WaterandAtmospheric Research Ltd, Private Bag 14901, Wellington, NewZealand*Corresponding author: tel: +64 93754505; fax: +64 78560151; e-mail: ian.tuck@niwa.co.nzTuck, I. D., Parsons, D. M., Hartill, B. W., and Chiswell, S. M. Scampi (Metanephrops challengeri) emergence patternsand catchability. – ICES Journal of Marine Science, doi: 10.1093/icesjms/fsu244.Received7 August 2014; revised 9December 2014; accepted 10 December 2014.Catchabilityisoftenakeysourceofuncertaintywithanystockassessment,butespeciallyforburrowingspecies,astheiremergentbehaviourisoftenpoorly understood. Quantification of catchability will provide a major step towards improvements in the assessment for many species. Scampi(Metanephrops challengeri) are widely distributed around New Zealand, and as with Nephrops (Norway lobster), they occupy burrows inmuddysubstrate,andareexploitedthroughtrawlfisheries,butareonlyavailabletothesefisherieswhenemergedontheseabed.Burrowemergenceisknowntovaryoverdailyandlongercycles.Uncertaintyovertrawlcatchabilityassociatedwithemergencepatternshasledtothedevelopmentofphotographicsurveyapproachesforscampi,basedonthecountsofburrows.Bothsurveyapproachesrequireanunderstandingofburrowoccu-pancyandemergenceratestoestimatetrawl/photosurveycatchability,whichisakeysourceofuncertainty.WeusedacoustictaggingtoexaminelevelsofandpatternsintheemergenceofMetanephrops,usinghydrophonereceiversmooredclosetotheseabed.Strongemergencecycleswereapparentinrelationtotidalcurrent(higheremergencewithinshorewaterflowacrosstheslope)andtimeofday(peakingjustafterdawn).Thesedatahavesubsequentlybeenusedwithinalength-basedstockassessmentofNewZealandscampi,whichusesemergencedatainconjunctionwithburrow and animal counts from photographic surveys, for the first time, to inform priors for trawl (mean 0.094) and photographic (mean 0.46)surveycatchability, and for burrow occupancy (mean 49.3%).Keywords: acoustic tagging, catchability, emergence, Metanephrops, scampi.

We present an evaluation of a 19-year numerical hindcast, Bluelink ReANalysis 3.5 (BRAN 3.5), derived from the Australian Bluelink programme. The evaluation is made for the surface and at 1000 m depth using Global Drifter Program (GDP) drifters and Argo floats, respectively. It is aimed at comparing the model’s ability to correctly simulate the Lagrangian mixing processes. These mixing and dispersal processes are often parameterised in terms of an eddy diffusivity, and we use diffusivity as a metric of comparison. Because the data are spatially sparse and non-uniform in time, the hindcast was sampled at the data locations to provide a model velocity corresponding to each observed velocity. Pseudo-Eulerian mean velocity and eddy kinetic energy fields were computed from bin-averaging of the Lagrangian velocities, and eddy diffusivities were computed from single-particle statistics. The mean field at 2.5 m in BRAN is largely dominated by Ekman transport. The mean velocities at 12.5 m compare well with the GDP drifter mean velocities, except that the Antarctic Circumpolar Current (ACC) is stronger in GDP than in BRAN. At the surface, EKE in BRAN tends to be marginally higher than in GDP. Modelled eddy diffusivity compares well with observed diffusivity, except in the boundary currents and near the equator. At 1000 m, BRAN tends to overestimate the mean field, EKE and eddy diffusivity at low latitudes and underestimate them at higher latitudes.

Argo floats cannot be regarded as true Lagrangian drifters because they periodically rise to the surface. Hence, previous estimates of eddy diffusivity at depth using single-particle statistics have been limited to one submerged cycle. However, unless the Lagrangian time scale is significantly shorter than the Argo cycle time, this single-particle calculation can have a large bias. Here, eddy diffusivity computed from single-particle statistics using Argo data is compared to that computed by assuming that Eulerian scales at depth are the same as at the surface, and that the relationship between Lagrangian and Eulerian time scales derived by Middleton is valid. If the methods provide the same answer, one can have confidence in both methods. Eddy diffusivity calculated from the single-particle statistics shows the same spatial structure as that computed from inferred time scale, but is smaller by a factor of about 2. It is suggested that this is because the deep Lagrangian time scale over much of the region is comparable to, or longer than, the 10-day Argo submergence cycle.

Models of larval dispersal in which ocean currents are Gaussian-distributed provide analytical solutions for the dispersal kernel that are usually assumed to be Gaussian-distributed as well. However, such Gaussian solutions are only valid for a single value of the duration of the larval period. In reality, larvae have a settlement competency period that can range from minutes to >1 yr. Dispersal kernels computed for larvae having a competency period of finite length become non-Gaussian, because later settlers contribute proportionately more to both tails of the kernels than early settlers. In species where the variability in larval duration is large, or mean flows are large compared to the variability, the dispersal kernels approach uniform rather than Gaussian distributions.

The critical-depth model for the onset of the spring phytoplankton bloom in the North Atlantic has recently been called into question by several researchers. The critical-depth model considers that the spring bloom starts when the mixed layer shoals to become shallower than a critical depth. Satellite and in situ measurements of chlorophyll are used here to show that the critical-depth model is indeed flawed. It is shown that the critical-depth model does not apply in the spring because the basic assumption of an upper layer that is well-mixed in plankton is not met. Instead, the spring bloom forms in shallow near-surface layers that deepen with the onset of thermal stratification. A stratification-onset model for the annual cycle in plankton is proposed that adheres to the conventional idea that the spring bloom represents a change from a deep- mixed regime to a shallow light-driven regime, but where the upper layers are not well mixed in plankton in spring and so the critical-depth model does not apply. Ironically, perhaps, the critical- depth model applies in the autumn and winter when plankton are well-mixed to the seasonal thermocline, so that the critical-depth model can be used to determine whether net winter production is positive or negative.

Non-indigenous marine species (NIMS) are likely to be introduced into harbours by ballast water discharge or hull fouling, and then spread by natural means in the nearshore/shelf circulation. Reanalysis currents from the Australian Bluelink modelling project are used with particle-tracking to estimate the potential dispersal kernels for major New Zealand ports. A random walk term is added to the particle-tracking algorithm to account for missing variability in the numerical model currents. With few data to validate the model, the choice of added diffusivity is made by comparing simulated dispersal with observed dispersal of Global Program drifters. Histograms of the dispersal times between the ports are computed, along with estimates of the 10 −1 and first percentile dispersal times, which are taken as proxies for the minimum dispersal time between the ports. The 10 −1 percentile times range from 1 day (e.g., Bay of Islands to Whangarei) to several months (Tauranga to Chatham Islands). We identify eight geographical regions, based on the coastal currents and present kernels for a representative port in each region, along with tables of the 10 −1 and first percentile dispersal times for all ports. The regions are: (1) northeast coast of the North Island (influenced by the East Auckland Current), (2) east coast of the North Island (East Cape Current), (3) west coast of the North Island (West Auckland Current/Taranaki Bight circulation), (4) Cook St (circulation in Cook St), (5) east coast of the South Island (Southland Current), (6) west coast of the South Island (Westland Current), (7) Fiordland (Subtropical Front/Southland Current), and (8) the Chatham Islands.

Lagrangian drifters, moored acoustic Doppler current meters and hydrographic observations are combined with wind observations to describe the mean and variable nature of flow around Kapiti Island, New Zealand. Thirteen day-long deployments of up to six Lagrangian drifters show the mean flow is to the southwest, with evidence of stronger flows in the channel separating the island from the mainland, and an island wake in the lee of the island. Vortices in this island wake may be tidally driven. Scaling considerations suggest the flow is strong enough that tidal-generated vortices are shed on each tidal cycle. Both the drifters and mooring data suggest that the d’Urville Current around Kapiti Island has a significant wind-driven component. During north-westerlies, the drifters tend to hug the coast, and south-eastwards flows in the Rauoterangi Channel are accelerated. We suggest the observed correlation is the local expression of a South Taranaki basin scale response to the winds.

Satellite-derived measurements of ocean currents made since 1993 were used to examine whether present-day oceanic circulation can explain 2 observed genetic lineages of the intertidal limpet Cellana strigilis in the New Zealand region. The western lineage comprises the South Island of New Zealand, Auckland and Campbell Islands, and the eastern lineage comprises the Chatham, Antipodes and Bounty Islands. Satellite-derived currents were used to simulate potential larval trajectories and hence estimate dispersal times between the islands. A random walk was added to the model to account for missing high-frequency variance in the satellite-derived currents. Surface drifters were used to calibrate the model. Minimum simulated dispersal times between the islands can easily explain the separation of the 2 lineages, but are too long, except between Antipodes and Bounty Islands, to explain the observed homogeneity within the lineages. It is suggested that rare long-distance dispersal events in the ocean, occurring too infrequently to be observed in 15 yr of simulations, influence the gene flow, and the frequency of these events may be estimated by fitting analytical functions to histograms of simulated larval dispersal. For some island pairs that are genetically connected, the maximum 10 d larval duration of C. strigilis appears at the 10–4 percentile. If this genetic connection occurs via larval dispersal, this suggests that connection may occur even when fewer than 1 in 1 million dispersal times is shorter than the larval duration.

Baroclinic tides can be observed both remotely from the Ocean Topography Experiment (TOPEX)/Poseidon (T/P) altimeter and in situ using current meters. However, it is rare that current meter moorings have high vertical resolution and are located under T/P ground tracks so that a direct comparison can be made between the two methods of observations. Here, data from a current meter mooring directly under a T/P ground track off the Bounty Plateau, New Zealand, are used to obtain energy estimates of the first baroclinic mode. These estimates are compared with those calculated from the T/P surface elevation. The two methods return estimates of the internal tide that are in agreement in phase and direction but have about a factor-of-2 difference in amplitude (and a factor-of-4 difference in energy); the flux estimates are 787 and 170 W m−1, respectively. Uncertainties in these estimates are relatively large, and there is a low but not negligible probability that the differences are entirely due to measurement error. However, there are other reasons that might explain the differences in the estimates. It may be that the baroclinic tide is highly modulated in time and the current meters were deployed during a period of relatively low amplitude.

The relative distributions of mid‐ and late‐stage phyllosomas, and pueruli, of Jasus edwardsii from five research cruises off the east coast of the North Island of New Zealand were analysed. A statistical approach was taken, using a bootstrapping with replacement method to test whether the observed mid‐ and late‐stage phyllosomas could have come from the same distribution. Differences in distribution between mid‐ and late‐stage phyllosomas were very close to significant at 95% confidence for two cruises, significant at 85% confidence for one cruise, and two cruises showed a non‐significant result. When all cruise data were combined into one analysis, the differences between mid‐ and late‐stage phyllosomas were significant with 85% confidence. On average, late‐stage phyllosomas were found 49 km further inshore than mid‐stage phyllosomas. One station during one cruise and one transect during another showed such high abundances that they affected the analyses. When these outlier data were excluded, ...

The rock lobster Jasus edwardsii forms the basis of important fisheries in south‐eastern Australia and New Zealand. Their long pelagic larval phyllosoma phase (12–24 months) raises many questions as to how the larvae are retained and/or recruited into local populations. Recent attempts to model the dispersal of J. edwardsii phyllosoma have had mixed success at reconstructing settlement patterns. However, these models have either ignored vertical distribution or have used that of the western rock lobster, Panulirus cygnus. We report on the vertical distribution and migration of J. edwardsii phyllosomas, collected in March/April 2003 from the Wairarapa Eddy off the east coast of the North Island, New Zealand, and provide a model to describe their vertical distribution. J. edwardsii phyllosoma were primarily recorded within the upper 100 m over similar depth ranges to those reported for other palinurid and scyllarid species. Well‐defined changes in diel vertical distribution were restricted to late‐...

The mean and variability of the circulation off the east coast of the North Island, New Zealand are investigated using shipboard conductivity‐temperature‐depth (CTD) and satellite altimeter data collected between 1993 and 2003. The altimeter data are used to adjust the in situ observations for the mesoscale eddy variability before computing the 11‐year mean in dynamic height. Mean dynamic height shows two anticyclonic eddies, centred near 178.3°E, 41.2°S and 176.2°E, 42.4°S. These locations are consistent with previous historical observations of the Wairarapa and Hikurangi Eddies, respectively. A long‐term trend in both in situ and satellite data shows that dynamic height rose at an average rate of up to 2 dyn cm year‐1 in the centre of the Wairarapa Eddy which is consistent with a strengthening of the eddy over the 11 years. The satellite data show periodic shedding of the Wairarapa Eddy from near East Cape at a rate of between two and three eddies per year. Thus, rather than indicating a perman...

The Wairarapa Eddy appears as a permanent anticyclonic eddy situated off the east coast of the North Island, New Zealand. In April 2001 a spatial survey of the eddy was made on a ship equipped with an Acoustic Doppler Current Profiler (ADCP). The absolute circulation at 100 m was estimated by objective mapping of the ADCP‐derived velocities to produce a velocity field that has enforced non‐divergence. Assuming that enforcing non‐divergence produces the best estimate of the geostrophic flow, this velocity field can be used as a “level of known motion” to reference geostrophic velocities at other depths. In particular, it can be used to estimate the velocity at 2000 dbar, which is otherwise used in this region as a level of no motion. The resulting flow at 2000 dbar has a mean speed of 0.07 m s‐1, and appears to be well correlated with the surface flow.

Two species of rock lobster (Palinuridae) are commonly found in coastal waters around both New Zealand and Australia. Both species have a relatively long larval phase, which raises the question of how populations can be maintained in regions where the mean flow advects larvae away from the coast. Previous work shows that Jasus edwardsii is genetically indistinguishable between Australia and New Zealand, whereas Sagmariasus verreauxi may be genetically different between the 2 countries. Satellite altimeter data are used here to test the hypothesis that Australia acts as a source of larvae for some New Zealand populations, and to investigate if there is a physical mecha- nism for the apparent lack of gene flow for S. verreauxi. A Lagrangian approach is used in which the western Tasman Sea is seeded with numerical drifters to build up a statistical summary of the likely distribution of larval trajectories. Ignoring biological factors, about 9% of J. edwardsii and about 2% of S. verreauxi larvae might be expected to arrive in New Zealand waters within their respective larval lifetimes. An order-of-magnitude estimate shows that even with extremely high mortality, these rates are high enough for trans-Tasman larval flow to contribute to and possibly maintain some New Zealand populations. Larval flow across the Tasman Sea should be sufficient to maintain trans- Tasman genetic homogeneity in both species. We suggest that 1 reason that S. verreauxi may not be genetically homogenous is that in order to arrive in New Zealand within 1 larval lifetime, they have to cross the Tasman Sea north of the productive Subtropical Front where there is not enough prey to sustain them.

Inverted echo sounder data from Station Aloha north of Oahu, Hawaii, are used to determine the absolute energy levels and time-varying nature of the first-mode baroclinic tide north of Hawaii. The semidiurnal tide amplitude and phase are modulated with 60–120 day periodicity, which indicates interaction between the baroclinic tide and intra-annual mesoscale variability seen off Hawaii. Close to the presumed generation site, there is less phase modulation, and amplitude modulation is uncorrelated with the far-field variability. These results are consistent with a model of phase-locked generation, together with phase and amplitude distortion during propagation. The sense of the amplitude modulation is opposite to that previously observed at decadal timescales. Semidiurnal energy flux at Station Aloha ranges from 2.2 to 8.0 kW m−1, with a mean value of 4.4 kW m−1.

Estimates are made of the mean and variability in temperature and salinity across the Subtropical Front along 178°30'E, east of New Zealand. These estimates are made from six sections taken over 5 years, but biased towards spring and autumn sampling. Monte‐Carlo simulations of the potential errors introduced by this sampling suggest that the mean fields can be reasonably well determined, but the annual cycles have large relative error. April 1999 residuals were anomalously warm and saline compared with the other five sections. These anomalies may be linked to an increased strength of the Wairarapa Eddy at that time.

The Wairarapa Coastal Current (WCC) flows along the east coast of the North Island, New Zealand. This current has been relatively little studied, and there are few direct measurements of its mean and variability. Here, an indirect method is used to gain some insight into the WCC variability and its effect on coastal temperature. Eight years of coastal temperature from Napier in Hawke Bay are modelled as a bivariate linear expansion of alongshore windstress and solar radiation. In the frequency domain, the modelled temperatures are coherent with observed temperatures, and in the time domain, they are positively correlated for 80% of the time. The model results suggest both that variations in the WCC are wind‐driven, and that alongshore advection accounts for most of the temperature change at Napier. Periods when the modelled temperature is not correlated with observed temperature could be because of episodes of enhanced upwelling, or large‐scale influences breaking down the alongshore temperature ...

The assumption that variability in the circulation around the east coast of the North Island of New Zealand can be described using satellite‐derived sea level and temperature and two empirical modes is tested. Potential errors are quantified if such an assumption is used to derive the internal fields from surface observations. Data from 303 CTD (conductivity‐temperature‐depth) casts made in the region between 1994 and 1999 are first tested against a much larger XBT (expendable bathythermograph) data set to show that they are representative of the expected variability, and then used to compute the empirical modes. The empirical modes are not baroclinic modes, but probably represent the lateral meanderings and variations in strength of the East Auckland Current, together with the presence of Rossby waves in this region. If surface temperature and height can be measured precisely, 60–70% of the variance in the internal temperature and velocity perturbations can be explained with the model. The stand...

Thermistor data from around Kaikoura Peninsula, New Zealand, show that in the synoptic band, temperature fluctuations at 1 m are coherent with the along-shore wind with a 90° phase differ- ence. At 5 m, however, the coherence is not as high, and the phase relationship is in the opposite sense. Along this coast, wind-driven advection of the coastal current and upwelling have opposite effects on temperature. A simple model is developed which has along-shore advection and upwelling terms parameterised in terms of the wind and two unknown coefficients. Solving for the coefficients shows that at 1 m, temperature variability is determined by along-shore advection, whereas at 5 m, temperature variability is determined principally by upwelling.

Massive mortality of marine life and algal blooms around central New Zealand during January and March 1998 have been attributed to the toxic dinoflagellate, Karenia brevisulcata (= Gymnodinium brevisulcatum). Phytoplankton surveys in January and February 1998 showed the presence of K. brevisulcata both offshore, in the relatively warm, southwards-directed East Auckland and East Cape Currents, and inshore, in the relatively cold, northwards-directed Wairarapa Coastal Current, along the entire east coast of the North Island. Small numbers of this species (200 cells 1−1) were also recorded on the southwest coast of North Island. A bloom of K. brevisulcata occurred in Wellington Harbour during February–March 1998. Reports of human respiratory distress progressed northwards along the Wairarapa Coast at about the same speed (0.23 m s−1) as observed for the Wairarapa Coastal Current, suggesting the passage of a bloom along the coast. We relate the timing of these events to local oceanographic conditions.

The Chatham Rise, which rises to a depth of 400 m immediately to the east of New Zealand, appears to constrain the latitudinal movement of the Subtropical Front (STF). Data from current meters moored over the Chatham Rise are used to estimate the spatial variation of mean and turbulent kinetic energies and Reynolds stresses in and around the STF. Mean currents over 8 months are to the east, and are stronger south of the rise. Mean speeds at 250 m south of the rise are as high as 15 cm s−1 compared to 4.5 cm s−1 north of the rise. In contrast, the eddy kinetic energy is higher to the north at 8.2 J m−3 compared to 3.3 J m−3 to the south. Short‐term current meter moorings over the crest of the rise show no coherence over length scales of 15 km for periods less than 10 days. At 250 m depth, Reynolds stresses are positive north of the rise, and negative south of it, suggesting an export of momentum away from the STF.

Long time series of sea level from tide gauges along the north side of the Hawaiian Ridge and shorter series of dynamic heights inferred from inverted echo sounders moored just north of the main Hawaiian Islands are examined for evidence of internal tides at the M2 frequency. We find that the amplitudes and phases of the M2 tidal components have low-frequency variability, which is consistent with a superposition of an internal tide with the larger barotropic tide. Further, the low-frequency variability is correlated with low-frequency changes in the depth of the pycnocline, which suggests a simple physical mechanism to account for the low-frequency modulations in the internal tidal amplitude. These modulations are coherent for long distances along the Hawaiian Ridge, indicating a coherent generation of the internal tide that is consistent with acoustic observations in the North Pacific and with recent analyses of sea surface heights from satellite altimetry.