EKP project update, June 2015

Project update (summary info)

This project was started in response to a lack of detailed information about the links between Eastern King Prawns (EKP) and estuarine habitat in NSW.

It has been known for some time that EKP spawn at sea, the larvae drift south on the East Australian Current before moving into our coastal estuaries. The tiny prawns spend some months growing in the estuary, before heading out to sea and swimming back up north; where they continue their growth to full maturity and complete the breeding cycle.

However, until now there has been little detail about which parts of the estuary are more important to young EKP. Where do they live? What do they feed on? Are mangroves, seagrass, salt marsh or unvegetated habitats more important; or are they all just as critical? Are some river systems more important than others?

The project is a three year study at sites in the Hunter River, Lake Macquarie and the Clarence River. The Fisheries Research and Development Corporation is funding this exciting body of work, with the interest, involvement and support of the commercial EKP fishing industry. There are also a number of other project partners who are providing additional support to the research program. These include Hunter Water, Newcastle Ports Corporation, Port of Newcastle Authority, Origin Energy, Hunter Local land Services, North Coast Local Land Services, Professional Fisherman's Association, University of Newcastle, Griffith University, Murdoch University and Oceanwatch.

Key research findings

• The project is now just over 1½ years since commencing and has completed its second research season.

• Over 840 trawls have been completed across the study area and 100% of the samples have been sorted.

• Data analysis is ongoing and will be reported in full over the next 12 months.

• Experiments were carried out to provide an estimate of the percentage of prawns being captured by the research trawls, compared to those that avoided capture.

• Catches were more efficient in bare substrates, compared to vegetated ones.

• Catches were much larger at night, compared to those made during the day time.

Hunter – EKP densities

• Results from the previously reported first research season (stable isotope data) indicated that most EKP running from the river originated from near the mouth of Fullerton Cove and the mangrove-lined habitats that dominate the lower reaches of the north arm of the river.

• The latest results from the second season used detailed EKP abundance data and confirmed the origination trend from the first season. The mouth of Fullerton Cove and "the dykes" had the highest densities of EKP recorded.

• The mangrove-lined habitats slightly further upstream on the north arm had very low densities of EKP.

• In addition, Ironbark Creek (Hexham Swamp) had high densities of EKP (but a relatively smaller area available to contribute prawns).

• The density of EKP and their contribution to the adult stocks likely depends on:

1. Vicinity to the river mouth;
2. Salinity;
3. Area of habitat available.

EKP use of inter-tidal habitats

• Small creeks draining saltmarsh areas (at Hexham, Tomago and Kooragang) were sampled to examine the use of these areas by EKP.

• The results from both sampling trips have been sorted, and showed that EKP density in these areas was surprisingly low. However, there were significant catches of a diverse range of other prawns and fish.

• 25 species of fish and prawns were captured with Glass shrimp and Glass gobies dominating the catches.

Salinity experiments

• Until now, the effects of freshwater on growth and mortality of juvenile EKP have been largely unknown in the scientific literature.

• Experiments are now underway to work out the effects of salinity changes (and the speed at which those occur) on EKP respiration rates and survival. Preliminary outcomes show a sharp decrease in survival at a salinity of ~14.

• This helps explain why commercial fishers notice fewer EKP, and smaller EKP, in wetter years.

Habitat loss

• Historic aerial photos have been organised and digitised for analysis for the Clarence River.

• Results from all three study sites (including Lake Macquarie and the Hunter) will be published over the next 12 months.

Communications

The following presentation has been conducted:

Russell, K. (2014) Why understanding Eastern King Prawn habitat is important. Update for the Chairs of the NSW Fishermans Coops, Sydney Fish Market, 21^stNovember 2014;

Detailed Progress report (submitted April 2015)

Objective 1: Determine to what extent young eastern king prawns (EKP) are using natural, degraded or rehabilitated habitat in estuaries

The second field season for the project has been completed, and over 80% of the samples sorted. To date, over 840 trawls have been completed across the three study estuaries, and in Season 2 approximately 20,000 juvenile prawns have been captured in research trawls across the Hunter River and Lake Macquarie, and approximately 5,000 paired length-weight measurements have been collected from these samples. Analysis of this data is ongoing and will be reported over the next two milestones. Following on from research conducted during the first field season, sampling in the second field season concentrated on obtaining quantitative estimates of abundance, examining differences in somatic (body) condition among different habitats, and looking in more detail at what saltmarsh habitats may provide for EKP.

When any trawl is used, it will usually only sample a proportion of the population present over the trawled area at the time of sampling. In order to convert numbers of prawns captured in trawl nets during our study to absolute numbers of prawns, an estimate of this efficiency is required. Efficiency is generally tested by repeatedly trawling a track, and examining the decay in numbers captured over the series of hauls. As mentioned in the first milestone report, this project developed a new research trawl net to conduct surveys of post larval, juvenile and adolescent prawns across a range of size-classes. The net is known as a 26B-5C net, and has a 26 mm diamond mesh in the body, and 5 mm octagonal mesh in the codend.

A removal/depletion experiment similar to that outlined in Loneragan et al. (1995) was conducted to derive an estimate of efficiency, during the day and night for king prawn, over vegetated and unvegetated habitats (i.e. the two broad habitat types being studied). Three replicate trials were conducted in each habitat type during both the day and the night, with each trial involving a series of 5-6 sweeps along a track. The Maximum-likelihood method of Schnute (1983) was used to determine the efficiency, with analysis conducted in the R-package FSA (Fisheries Stock Assessment). The efficiency of the net over unvegetated habitats during the night was determined to be 0.48 ± 0.06 (mean ± 95% C.I.). The efficiency of the net over vegetated habitats during the night was slightly lower than for unvegetated habitats and more variable, at 0.33 ± 0.13 (mean ± 95% C.I.). This is to be expected, as seagrass is known to affect the efficiency of the trawl. These efficiency estimates closely aligned with those in Loneragan et al. (1995), which used a similar sized net but with a different mesh size. The gear did not function well during the day (not enough prawns could be captured to estimate efficiency); this supports previous findings which indicate EKP are only really efficiently sampled during the night (Guest et al. 2003), due to nocturnal activity patterns.

As described in the previous milestone report, stable isotope data from Season 1 was used to inform the design of the quantitative sampling regime in Season 2. The majority of emigrating Eastern king prawns analysed in Season 1 appeared to be originating from Fullerton Cove and also the vast mangrove lined habitats that dominated the lower reaches of the north arm of the river. Almost no emigrating king prawns originated from the extensive mangrove and saltmarsh habitats further up the north arm of the estuary. The highest contributing habitat sites were all within the salinity range 25 – 33; a window that closely reflects the range through which eastern king prawns can best osmoregulate (the isosmotic point for juvenile Eastern king prawns is 28, Dall 1981). The agreement between the isotopic data from Season 1 and the quantitative abundance data from Season 2 is remarkable. Overall, estimates of absolute EKP abundance in the Hunter River ranged from 0 – 5.8 prawns m^-2. At the time of writing, a surface representing EKP abundance was interpolated from November 2014 and January 2015 field data, and the proportional contribution of putative nursery habitats to running prawns from Season 1 was overlaid (Figure 1). This produced some very clear patterns. The area around the mouth of Fullerton Cove and "the dykes" on the north arm of the river was identified as the primary nursery area contributing to emigrating EKP captured at the mouth of the river. These areas supported the highest densities of EKP in the river, and represented a reasonable area of shallow, unvegetated benthic habitat (i.e. large area, high densities). On the south arm of the river, Ironbark Creek (the main habitat rehabilitation site in this project), supported similarly high densities of EKP. This location had a moderate contribution to emigrating prawns (smaller circle in Figure 1), however, due to relative area of this habitat (smaller area, but with high densities). Isotopic data also indicated that the area inside Fullerton Cove had a moderate contribution to EKP, but much lower densities were compensated by much greater area in this region (i.e. large area, low densities). When all these data are taken together in the context of the salinity gradient in the river (Figure 2), the density of prawns and their contribution to the emigrating cohort is likely a function of the several factors, including 1) vicinity to mouth (important for inward-moving coastally spawned larvae); 2) salinity (important for a stenohaline species); and 3) area of habitat. These relationships will be described in a quantitative fashion to support modelling under Objective 4. Data analysis for the Clarence River and Lake Macquarie is ongoing, as is analysis of somatic condition among different habitats. In addition, data (size and abundance) has been collected for all commercial prawn AND fish species captured in this study, and will be worked up and published over the next 18 months.

A key underlying theme of this project is to understand how rehabilitated habitats (such as marsh and mangrove habitats) may benefit Eastern king prawns. As flagged in the last milestone report, a new sub-component of the broader project was commenced in this milestone period to evaluate the use of intertidal marsh and sub-tidal creeks in the Hunter Estuary. While much of the Hunter Estuary is considered to be in a degraded condition, it does contain significant areas of intertidal mangrove and saltmarsh habitat, much of which is the result of rehabilitation works. In particular, the north arm is lined with dense stands of mangroves which are drained by a complex network of intertidal creeks. While marshes become inundated only during spring high tides at water levels 1.5 m above ISLW (approximately 20 hours per month), given their size and potential productivity, they may provide other "services" for Eastern king prawns that benefit either their growth or survival. A number of sub-tidal creeks flow through the mangrove habitat that have very steep banks held together by pneumatophores, while the bottom is mostly bare silty sediment. This project component aimed to evaluate whether prawns may utilise the creek-ward marsh edge (Figure 3) as refuge from current flow and/or predation (this has never been evaluated for prawns in temperate Australia).

Three locations (Tomago marsh, Kooragang marsh, and Hexham marsh) within the Hunter Estuary were included for the study, with each location representing a rehabilitated habitat containing two sub-tidal creeks that drain the surrounding intertidal marsh/mangrove habitat (Figure 1). One sampling trip was conducted during the last quarter of the lunar month in January, and another in March (not yet sorted). An additional third sampling trip is planned. At each location three, double-wing Fyke nets were deployed to qualitatively examine whether Eastern king prawns were moving up onto the marsh surface (and directly using the intertidal mangrove/marsh habitat) during the nocturnal high tide. High tides inundate the high marsh habitat during the night in the last quarter of the lunar month, and Fyke nets (Figure 4) were set at dusk and were retrieved at dawn the following morning. The other sampling nets used in this sub-component were 3/16" mesh monofilament cast nets. These nets were used to sample the creek-ward marsh edge (refuge) and middle (non-refuge) of the creek, as they allow high accuracy of placement, high levels of replication, and effectively sample prawns (Figure 4, Baker & Minello 2011; Stein III et al. 2014).

Within each of the three locations (Tomago marsh, Kooragang marsh, and Hexham marsh), two sub-tidal creeks were sampled using cast nets. Creeks were divided into 'adjacent' areas, which comprised of the edges of the creek within 2 m of the bank, and 'middle' which included creek areas at least 2 m from the bank. For each of the 6 creeks, 8 paired replicate cast net hauls were made in adjacent and middle areas, yielding 96 samples for each of the 3 sampling events (almost 300 samples in total, but only data from the first sampling event is presented here). All sampling was conducted at night.

Interestingly, only seven Eastern king prawns were collected in the intertidal marsh habitat sampled with the Fyke nets, with 5 of these captured in recently rehabilitated Hexham marsh. Catches in the Fyke nets otherwise included 25 other species of commercial and non-commercial fish and invertebrates, but were dominated by Macrobrachium intermedium (4058 individuals) and Gobiopterus semivestitus (1574 individuals).The low numbers of Eastern king prawn in relation to other species appears to indicate that they do not directly utilise intertidal habitat to a great extent, although it should be noted that two more sampling events are still to be analysed/completed.

Within the sub-tidal creeks, cast nets also sampled only a small number of Eastern king prawns (noting that each cast covers about 3-4 m². The highest number of Eastern king prawn was collected in the Hexham marsh (15 individuals), followed by the Tomago marsh (6 individuals) and the Kooragang marsh (5 individuals). When the data from this first sampling event was pooled across locations and Adjacent and Middle areas of creeks compared, there was a slight increase in abundance of Eastern king prawn in the middle areas (noting the caveat that is only a single sampling event; Figure 5). Although the marsh edge does contain "complex" structure (and high densities of small fish are often seen congregating in this habitat), the data does not appear to indicate that the edges of mangrove/marsh-lined creeks support a greater abundance of prawns than within the middle of the creeks.

The greater abundance of EKP on intertidal habitat in the Hexham marsh is an interesting finding, and supports the data presented in Figure 1 (negligible prawn densities were detected in the Tomago and Kooragang marshes). Although the overall numbers from this section of the study are small, patterns indicate that prawns are moving into this recently rehabilitated habitat despite the lack of established mangrove forest or saltmarsh plant species (the marsh is still undergoing transition from a Phragmites dominated system). The low numbers of prawns adjacent to sub-tidal creek banks was unexpected; while it was hypothesised these areas provide food and shelter, these results may arise due to a greater density of fish predators also utilising these edge habitats as a flow refuge. Further planned sampling will help determine if these preliminary findings are indicative of marsh-edge habitat use within the rehabilitated marshes of the Hunter Estuary.

Objective 2: Determine the hydrographic conditions which provide for maximum growth and survival of EKP within nursery habitats

The effect of abiotic conditions on Eastern king prawns, principally, low salinity arising from freshwater inundation of estuaries, are relatively unknown. As highlighted in the project proposal and previous milestone reports, professional fishers have observed and reported decreases in Eastern king prawn catch rates, and also slower growth, in years of high rainfall. Eastern king prawn have been proposed as being relatively stenohaline when compared to other commercial prawn species in NSW (i.e. greasyback and school prawn); however, Dall (1981) demonstrates that small prawns have the physiological capacity to deal with hyposaline conditions as low as 3 psu through osmoregulation. Juvenile eastern king prawns can osmoregulate down to about 10 psu; however, as they grow to adult sizes they lose this ability and are poor osmoregulators across the spectrum of salinities (Dall 1981). Evidence exists that another penaeid prawn, Metapenaeus bennettae(Greentail prawn), can acclimatise to a new salinity within 48 hours (Aziz & Greenwood 1981); however this species is much more euryhaline that Eastern king prawn so acclimation of this species may take longer, if it occurs at all. The length of time they take to acclimate (if they do so) is one of the outputs of this experiment (mortality rate vs time).

To determine the rates of mortality, respiration and activity of Eastern king prawn to salinity changes, Eastern king prawn will be exposed to salinity declines to 18 endpoints over 24 hours, from a starting salinity of ≈35 (Figure 6). These salinities were chosen as they reflect what may be considered as 'infrequent', 'rare' and 'extreme' events with regard to salinity data measured in the field. After the 24 hr transition period, the prawns will then be kept for an additional 5 days to allow them to acclimatise to the new salinity levels. Those prawns which survive the drop in salinity and subsequent 5 day acclimation period will then be subjected to respiration experiments. This will assess the amount of energy they may be expending to osmoregulate after a change in salinity. Specifically the aims of this experiment are:

1. Estimate mortality of EKP experiencing different rates of salinity decline
2. Establish whether prawn mortality and/or are stress (respiration) is linked with the rate of change in salinity or the salinity itself.

Approximately 250 Eastern king prawns were collected from the Hunter Estuary in February and March 2015 for use in the experiments, and are currently held in holding tanks at the Port Stephens Fisheries Institute (biosecurity restrictions prevented importing hatchery-reared prawns from Queensland as previously proposed). Having completed an essential pilot study (see below), treatment experiments are now being conducted in the newly completed Port Stephens Fisheries Institute Fisheries and Aquatic Research Aquarium. Treatments are being staggered, with a new treatment commencing each day over 18 days, and no more than five treatments and a procedural control running at any one point in time. Each treatment tank starts with 10 L of seawater at ambient salinity (≈35, equal to the control), a thin layer (5 cm) of sand on the bottom of each tank, and contains ten randomly allocated prawns. Treatments are also randomly allocated across start days.

The appropriate amount of fresh water is added to each treatment tank using computerised dosage pumps (Grundfos), so that the required salinity is reached by the end of the 24 hr transition period (Figure 6). The control tanks are set to receive the same dosage rates as the largest (salinity of 34) and smallest (salinity of 2) transition treatments, but are dosed with seawater at a salinity of 35. Once the endpoint salinity is reached, tank volume and salinity will be maintained for an additional 5 days of acclimation/experimental time. At the conclusion of each experiment, up to nine prawns (less if high mortalities occur) are removed from each treatment and placed into sealed respiration chambers (Figure 7) and their oxygen consumption is measured for 5 hours using an oxymeter (PreSens oxy-10 mini, Figure 7, measuring oxygen levels each minute). A 5-hr window allows for measureable oxygen consumption to be calculated and the effect of salinity changes on Eastern king prawn metabolism to be determined. Pilot studies suggest that Eastern king prawn metabolism remains independent of the oxygen saturation levels prawns will experience in the chambers over a 5 hour period (see pilot study below). At the conclusion of each experiment, prawns are sacrificed and dried in an oven (100°C) and their dry weight determined for use in metabolic rate calculations.

The oxygen dependence status of the metabolism of Eastern king prawn has not yet been expressed in the published literature. Knowledge of this is critical in the design of experiments that asses and/or compare the metabolic rate in response to experimental treatments (e.g. salinity changes). Specifically, the determination of a critical oxygen saturation point at which this species may switch modes of metabolism in response to surrounding oxygen availability is useful for the design of respiration-based experiments (Dall 1986). This is essential to defining the range of oxygen saturation levels within which experiments can be run without affecting the response of the participating prawns due to this potential additional factor. By plotting oxygen consumption against the background oxygen saturation, a switch in the metabolic mode of the prawns, if it occurs, can be identified by a negative inflection in the graph (Dall 1986). Before the effect of salinity declines on Eastern king prawn metabolism was evaluated using respiration chambers (as described above), the point at which the metabolism of Eastern king prawn remains independent of oxygen saturation was determined by completing a pilot study.

Three prawns were randomly selected from the holding tanks and placed into respiration chambers which were filled with water at a salinity of 35. A fourth chamber which contained no prawn acted as a control. The chambers were placed into a dark box to eliminate the potential of light affecting the probes (which are light sensitive). The prawns were held in the chambers for 8 hours and the oxygen levels of the water measured each minute. At the conclusion of the 8 hour period, prawns were removed from the chambers and their wet weight measured. The data was converted using the weight of the prawn to oxygen consumption (mg O₂ g-of-prawn^-1 h^-1). There was no clear inflection in the metabolic rate for each of the three prawns as oxygen saturation within the chambers declined (Figure 8). For all prawns, even at low oxygen saturation levels (≈40%), metabolic rate remained similar to higher levels of oxygen (≈90%). It took 8 hours for oxygen levels within the chambers to drop below 50%, with no change in metabolic mode of the prawns, indicating their metabolism is oxygen-independent across this range. Given this result, we are confident that for the main salinity experiments, measuring metabolic rate at various salinities over a five hour period will provide robust data on the effects of salinity changes on Eastern king prawn which is not affected by changes in the ambient oxygen levels within the chambers.

Objective 3: Determine the extent of key EKP habitat lost and remaining in a number of key estuaries between the Tweed and the Hawkesbury

Progress was reported against this objective in the last milestone report. Mapping is ongoing according to the prioritised workplan outlined previously. Spatial analyses of habitats will be ongoing in the Hunter River and other study estuaries over the next 12 months. Work during this milestone period has concentrated on digitising and rectifying old aerial photo imagery ready for spatial analysis for the Clarence River (Figure 9).

Objective 4: Outline the potential improvements to the EKP fishery that could be achieved through targeted wetland rehabilitation and freshwater flow management

We are not due to report against this Objective until the next milestone, but I wish to note that we have appointed a PhD student (Mr. Andrew Broadley) through Murdoch University (where CI Taylor is an Adjunct Associate Professor), to pursue the modelling associated with this objective using the program Enhancefish (http://fisheriessolutions.org/projects/enhancefish/). This work will be undertaken in conjunction with Prof. Neil Loneragan and Dr. James Tweedley, as a similar approach is being developed by Mr. Broadley for work undertaken in FRDC 2013/221 "Stock enhancement of the western school prawn (Metapenaeus dalli) in the Swan Estuary; evaluating recruitment limitation and release strategies".