RESEARCH TO INFORM SEED-BASED SEAGRASS RESTORATION IN AOTEAROA NEW ZEALAND

Full text

REPORT NO. 3994
RESEARCH TO INFORM SEED-BASED SEAGRASS
RESTORATION IN AOTEAROA NEW ZEALAND

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
RESEARCH TO INFORM SEED-BASED SEAGRASS
RESTORATION IN AOTEAROA NEW ZEALAND
A CAWTHRON UNDERGRADUATE SUMMER SCHOLAR REPORT
DEMI FEARN, IRISA HUDSON, DANA CLARK, ANNA BERTHELSEN,
DAN CROSSETT, MAUREEN HO
Prepared with funding from the Westpac New Zealand Government Innovation Fund, Port
Nelson Limited, OneFortyOne, Friends of Nelson Haven and Tasman Bay and a Royal Society
Catalyst Grant.
CAWTHRON INSTITUTE
98 Halifax Street East, Nelson 7010 | Private Bag 2, Nelson 7042 | Aotearoa New Zealand
Ph. +64 3 548 2319 | Fax. +64 3 546 9464
www.cawthron.org.Aotearoa New Zealand
REVIEWED BY:
Lisa Floerl
APPROVED FOR RELEASE BY:
Grant Hopkins
ISSUE DATE: 6 December 2023
RECOMMENDED CITATION: Fearn D, Hudson I, Clark D, Berthelsen A, Crossett D, Ho M. 2023. Research to inform seed-
based seagrass restoration in Aotearoa New Zealand. Nelson: Cawthron Institute. Cawthron Report 3994.
DISCLAIMER: While Cawthron Institute (Cawthron) has used all reasonable endeavours to ensure that the information
contained in this document is accurate, Cawthron does not give any express or implied warranty as to the completeness of
the information contained herein, or that it will be suitable for any purpose(s) other than those specifically contemplated
during the project or agreed by Cawthron and the client.
© COPYRIGHT: Cawthron Institute. This publication may be reproduced in whole or in part without further permission of the
Cawthron Institute, provided that the author and Cawthron Institute are properly acknowledged.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
i
EXECUTIVE SUMMARY
Seagrass meadows are an important coastal ecosystem in Aotearoa New Zealand and
around the world. They provide habitat for fauna, contribute to climate change mitigation by
storing carbon, and improve water quality by trapping sediments and taking up excess
nutrients. However, seagrass meadows have declined in many places around the world,
including Aotearoa New Zealand, due to physical disturbance, deteriorating water quality and
the changing climate. There is concern that a continued decline of seagrass meadows will
lead to adverse effects on marine ecosystems. To protect seagrass meadows and promote
their recovery, there has been a global effort to facilitate their restoration. In Aotearoa New
Zealand, restoration of seagrass beds has been trialled through transplanting from wild donor
meadows, but this method can cause damage to the wild meadows. Seagrass restoration
methods with a lower environmental impact have been trialled successfully around the world,
although these methods have not been tested in Aotearoa New Zealand.
Our study examined the feasibility of seed-based restoration techniques for the species of
seagrass found in Aotearoa New Zealand (Zostera muelleri) using Nelson / Whakatū
(hereafter Nelson) as a trial region. Methods for surveying and collecting flowers and
extracting, storing and germinating seeds were developed based on approaches used by
collaborators in Australia, who are researching seed-based restoration techniques for the
same species.
We evaluated the success of the methods to determine whether it is possible to develop a
seed-based technique for seagrass restoration in Aotearoa New Zealand. The findings from
this study showed there would be sufficient flowers within the seagrass meadows for seed-
based restoration if collected at peak flowering (November to December during our study
season). The seeds were successfully extracted from the flowers using a specially designed
tank that took advantage of the negatively buoyant seeds. This system was effective within a
controlled laboratory environment as well as outdoors where the tank was exposed to natural
fluctuations in environmental conditions. The seeds that were collected using this method
were sterilised and stored in pottles of autoclaved seawater until the germination trial was
started (in our study the first germination trials were conducted within 2 months of collection).
All the methods used in our study were successful and showed that seed-based restoration
is feasible in Aotearoa New Zealand. However, further research is required to optimise these
processes so that they can be easily upscaled to carry out larger restoration projects
nationwide.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
iii
TABLE OF CONTENTS
1. INTRODUCTION ........................................................................................................... 1
2. FLOWER SURVEYS ...................................................................................................... 5
2.1. Study aim........................................................................................................................................................ 5
2.2. Methods .......................................................................................................................................................... 5
2.2.1. Site selection ............................................................................................................................................. 5
2.2.2. Temperature loggers ................................................................................................................................. 8
2.2.3. Surveys ..................................................................................................................................................... 8
2.2.4. Data analysis ............................................................................................................................................. 9
2.3. Results ......................................................................................................................................................... 11
2.3.1. Flowering shoots through time ................................................................................................................ 12
2.3.2. Flowering shoots across tidal heights...................................................................................................... 14
2.3.3. Flower stages .......................................................................................................................................... 15
2.3.4. Relationship between flowering shoots and spathes with the percent cover of seagrass ....................... 16
2.3.5. Temperature data .................................................................................................................................... 19
2.4. Discussion .................................................................................................................................................... 21
3. FLOWER COLLECTION ...............................................................................................24
3.1. Study aim...................................................................................................................................................... 24
3.2. Methods ........................................................................................................................................................ 24
3.3. Results and discussion ................................................................................................................................. 24
4. SEED EXTRACTION ....................................................................................................27
4.1. Study aim...................................................................................................................................................... 27
4.2. Methods ........................................................................................................................................................ 27
4.3. Results and discussion ................................................................................................................................. 29
4.3.1. Flowers and seeds .................................................................................................................................. 29
4.3.2. Tank conditions ....................................................................................................................................... 31
5. SEED STERILISATION AND STORAGE ......................................................................33
5.1. Study aim...................................................................................................................................................... 33
5.2. Methods ........................................................................................................................................................ 33
5.3. Results and discussion ................................................................................................................................. 34
6. SEED GERMINATION ..................................................................................................36
6.1. Study aim...................................................................................................................................................... 36
6.2. Methods ........................................................................................................................................................ 36
6.3. Results and discussion ................................................................................................................................. 37
7. OVERALL CONCLUSIONS ..........................................................................................39
8. RECOMMENDATIONS FOR FUTURE STUDIES .........................................................40
9. ACKNOWLEDGEMENTS .............................................................................................41
10. APPENDICES ...............................................................................................................43
11. REFERENCES .............................................................................................................46

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
iv
LIST OF FIGURES
Figure 1. Examples of fauna using the seagrass habitat. .................................................................. 1
Figure 2. Structure of the Zostera muelleri inflorescence (cluster of male and female flowers). ....... 3
Figure 3. Diagram demonstrating how the quadrats were arranged at each survey site. ................. 6
Figure 4. Location of fine-scale flower survey sites and the temperature logger in Nelson Haven. .. 6
Figure 5. Location of fine-scale flower survey sites and the temperature logger in Waimea Inlet..... 7
Figure 6. Location of fine-scale flower survey sites and the temperature logger in Delaware
Inlet. .................................................................................................................................... 7
Figure 7. Nelson Haven temperature logger once deployed. ............................................................ 8
Figure 8. Examples of the flower stages, from the left: pre-seed (with stigmas out), developing
seed (containing immature seeds), post-seed (spathe that has dropped its seeds). ......... 9
Figure 9. Example of the random point count approach. ................................................................. 10
Figure 10. Number of individual spathes detected over the 5 months of surveys (November
2022–March 2023) across the tidal heights (including the Waimea Inlet ‘patch’) for
each estuary. ..................................................................................................................... 11
Figure 11. Number of flowering shoots found in each quadrat over the 5 months of surveys
(November 2022–March 2023) at each estuary ............................................................... 13
Figure 12. Number of spathes found in each quadrat over the 5 months of surveys (November
2022–March 2023) at each estuary .................................................................................. 13
Figure 13. Number of flowering shoots found in each quadrat over the three tidal heights (High,
Mid, Low) at each estuary ................................................................................................. 14
Figure 14. Number of spathes found in each quadrat over the three tidal heights (High, Mid, Low)
separated into the three different estuaries ...................................................................... 15
Figure 15. Stacked bar graphs showing the total number of spathes at each stage over the 5
months of surveys ............................................................................................................. 16
Figure 16. Number of flowering shoots found in December at Delaware Inlet, Waimea Inlet and
Nelson Haven plotted against the percent cover of seagrass at three different tidal
heights ............................................................................................................................... 17
Figure 17. Number of spathes found in December at Delaware Inlet, Waimea Inlet and Nelson
Haven plotted against the percent cover of seagrass at three different tidal heights ....... 18
Figure 18. Delaware meadow maximum, mean and minimum temperatures (°C) from 14
September 2022 to 3 March 2023. ................................................................................... 20
Figure 19. Nelson Haven meadow maximum, mean and minimum temperatures (°C) from 25
August 2022 to 26 March 2023. ........................................................................................ 20
Figure 20. Waimea Inlet meadow maximum, mean and minimum temperatures (°C) from
6 September 2022 to 11 January 2023. ........................................................................... 21
Figure 21. (A) Number of flowering spathes collected by stage at each collection event in Nelson
Haven between December 2022 and January 2023. (B) Number of spathes collected
within each flower stage category. .................................................................................... 25
Figure 22. Representation of effort for the flower collection. Number of spathes picked per person
per hour for each collection event in Nelson Haven. ........................................................ 26
Figure 23. Tank set-up (‘seagrass spas’). .......................................................................................... 28
Figure 24. Proportions of seagrass spathes at each stage in each tank (inside and outside). ......... 29
Figure 25. Number of seagrass spathes added to the tanks (inside and outside) and number of
seeds collected from each tank. ....................................................................................... 30
Figure 26. Development of a seagrass spathe in the inside tank. (1) 23 January. (2) 24 January.
(3) 27 January. (4) 31 January. ......................................................................................... 30
Figure 27. Temperature (°C, plot above) and light (lux, plot below) conditions of the inside and
outside tanks. .................................................................................................................... 32
Figure 28. (A) Seagrass seeds extracted from the tank before being sterilised. (B) Seeds being
soaked in ethanol solution. (C) Seeds being mixed in bleach solution............................. 33
Figure 29. Proportions of colours observed in seagrass seeds extracted from the inside and
outside tanks. .................................................................................................................... 35
Figure 30. (A) Seagrass seeds being put through an overnight freshwater flush. (B) Seeds in
Petri dish, monitored for germination. (C) Germinated seed with emerging cotyledon. ... 37
Figure 31. Germinated seagrass seed grown into a seedling in the laboratory. ................................ 38

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
v
LIST OF TABLES
Table 1. Coordinates and deployment dates for temperature loggers in the three estuaries. .......... 8
Table 2. Minimum, maximum, and mean temperature (°C) and light (lux) recorded at the
surface and bottom of the inside and outside tanks over the course of one week (19–
25 January 2023). ............................................................................................................. 31
Table 3. One-off salinity measurements from the inside and outside tanks. .................................. 32
LIST OF APPENDICES
Appendix 1. Survey site coordinates. .................................................................................................. 43
Appendix 2. Dates of flower surveys and flower presence. ................................................................. 44
Appendix 3. Summary of estuary temperature data: minimum, mean and maximum air / water
temperatures (°C) recorded in Nelson seagrass meadows over the 2022/23 summer. ............... 45

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
1
1. INTRODUCTION
Seagrass is a flowering plant (an angiosperm) that grows in the ocean. It can form
vast meadows in both intertidal and subtidal areas on every continent around the
world except Antarctica (Turner and Schwarz 2006). Zostera muelleri is the only
species of seagrass that is found in Aotearoa New Zealand and most commonly
grows in estuaries on sand and mudflats. The plant has adapted well to living in
aquatic environments by forming an extensive system of rhizomes that grow beneath
the sediment, allowing the plant to withstand tidal currents. Above the sediment, it has
blade structures and flowers that are fertilised by the movement of string-like pollen
through the water column, a phenomenon known as hydrophilous pollination (Turner
and Schwarz 2006).
Seagrass is considered an ecosystem engineer, as it significantly modifies the
surrounding environment, providing many benefits to the coastal ecosystem. Due to
the complex structure that the plant creates above and below the sediment, the
meadows can promote areas of increased biodiversity within a seafloor of
homogeneous soft sediment. This modified habitat is used by a variety of species for
different lifecycle stages (Figure 1); for example, juvenile fish seek refuge amongst
seagrass blades (Spalding et al. 2003). This nursery function makes the meadows
critical for fisheries production around the world.
Figure 1. Examples of fauna using the seagrass habitat. Photo credits: Dana Clark, Bruce Green,
Breanna Hindmarsh, Rachel Hooks, Cawthron Institute.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
2
Seagrass meadows can also improve water quality. In meadows with a well-
developed canopy, the rate of water flow is slowed, promoting deposition of sediment
and organic material (Turner and Schwarz 2006). This process leads to a decrease in
water turbidity, which, in turn, benefits the seagrass by improving light penetration
through the water. Dense rhizome structures beneath the sediment surface help to
stabilise the deposited sediment and reduce erosion and resuspension of estuarine
sediment during storms.
Seagrass blades and rhizomes take up excess nutrients from the water column and
sediments, contributing to water quality improvement. These nutrients are converted
into biomass and are stored in the plants’ extensive rhizome system, which acts as a
below-sediment carbon store. Carbon originating from sources outside the meadow,
can also be stored when deposited and retained within the meadows (e.g. in the form
of carbon-rich sediments). As such, seagrass habitats are important contributors to
climate change mitigation through carbon sequestration (Macreadie 2014; Nordlund et
al. 2016).
Seagrass is known to reproduce asexually through the process of rhizome growth or
fragmentation. It was previously thought this was the most common form of
reproduction for seagrass in Aotearoa New Zealand. However, as seagrass is an
angiosperm, it can also reproduce sexually through pollination of female flowers. The
seagrass flower structure is monoecious,
1
where multiple separate male and female
flowers are temporally distinct within the same spathe
2
(Figure 2). A flowering shoot
may contain multiple spathes, and each female flower, when successfully pollinated,
will produce one seed (Conacher et al. 1994). The first study of sexual reproduction of
Z. muelleri in Aotearoa New Zealand was conducted in 1997 (Ramage and Schiel
1998). Since then, a further three studies have confirmed that Z. muelleri flowering is
more common than originally thought (Dos Santos et al. 2017; Zabarte-Maeztu et al.
2021; Hindmarsh and Hooks 2022). The flowers are very cryptic and can be difficult to
detect, which possibly explains why they were previously overlooked. In the Nelson
Region, the presence of flowers was confirmed by Cawthron Institute in the summer
of 2021/22 (Hindmarsh and Hooks 2022).
1
It has both male and female reproductive organs on the same individual.
2
A spathe is a modified leaf that forms a protective sheath surrounding the male and female flowers.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
3
Figure 2. Structure of the Zostera muelleri inflorescence (cluster of male and female flowers).
(A) Spathe structure. (B) Male flower (will release a string like pollen). (C) Female flower
with one stigma (hook like structure that catches pollen) and one ovary. Photo credits:
Phil Garnock-Jones.
Seagrass relies on light penetration through the water column and an adequate
nutrient supply, making the plant sensitive to changes in the surrounding environment.
The current global decline in seagrass meadows is linked to anthropogenic activities,
such as modification of natural coastlines and intensive land use within catchments.
Land run-off from developed areas often contains increased sediment, excessive
nutrients and toxic compounds, which lead to poor water quality, a known key driver of
seagrass loss (Short and Wyllie-Echeverria 1996; Bester 2000). Physical damage to
meadows can be caused by vehicles, boat anchors and moorings, and dredging
(Walker et al. 1989; Erftemeijer and Robin Lewis 2006; Ceccherelli et al. 2007; Šunde
et al. 2017). Globally, it is estimated that at least 29% of seagrass meadows have
already been lost (Waycott et al. 2009). In Aotearoa New Zealand, Z. muelleri is listed
as ‘At Risk – Declining’ under the Zealand Threat Classification System (de Lange et
al. 2018).
To date, most seagrass restoration efforts in Aotearoa New Zealand have been
transplant-based (e.g. Reed et al. 2004; Gibson and Marsden 2016; Matheson 2016;
Matheson et al. 2017; Morrison 2021; Zabarte-Maeztu 2021). Transplant-based
restoration involves taking seagrass from a donor meadow and planting it at a
restoration site. This technique, which is used globally, has varying success rates and
can potentially damage the donor population (Bull et al. 2004; Ferretto et al. 2021).

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
4
Furthermore, the method of transplantation can limit the genetic diversity of the
restored population if all transplants are sourced from the same donor meadow
(Williams 2001). An alternative to transplantation is seed-based restoration. This
method can reduce negative impacts on existing meadows and will likely result in
greater genetic diversity in restored meadows. Although seed-based restoration
methods are yet to be trialled using Z. muelleri in Aotearoa New Zealand, they have
been successfully used for Z. muelleri in Australia (e.g. Tan et al. 2023), as well as for
other species around the world (e.g. Zostera marina in Virginia, USA; Orth et al.
2020).
Clark and Berthelsen (2021) recently reviewed seagrass restoration techniques in the
context of Aotearoa New Zealand, and several components of seed-based restoration
were trialled in Nelson (Hindmarsh and Hooks 2022). Our research builds on this work
by describing seagrass flowering patterns in the Nelson Region and developing
methods for seed-based restoration of seagrass in Aotearoa New Zealand. In
consultation with research associates from Central Queensland University (Associate
Professor Emma Jackson) and Deakin University (Associate Professor Craig
Sherman) in Australia, we developed methods to survey and collect flowers, and
extract and store seeds. We also conducted a series of preliminary seed germination
trials. This report outlines the findings of our work.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
5
2. FLOWER SURVEYS
2.1. Study aim
This study aimed to describe spatial and temporal / seasonal patterns in seagrass
flowering within three meadows in the Nelson Region to inform future flower collection
methods for seed-based restoration.
2.2. Methods
2.2.1.
Site selection
Seagrass meadows were selected from three estuaries in the Nelson / Whakatū
Region: Nelson Haven / Paruparuroa, Waimea Inlet / Waimeha and Delaware Inlet /
Wakapuaka. Tidal cycles in Aotearoa New Zealand are semi-diurnal; that is, on most
dates, two high and two low tides will occur (MfE Coastal Factsheet 4
3
). Fine-scale
survey sites were selected at three tidal heights (low, mid and high) in each meadow
based on LiDAR
4
data. From August 2022 (Nelson Haven) or September 2022
(Waimea and Delaware Inlets), meadows were visited monthly to determine whether
flowering had begun. Flowering presence was determined by searching for flowers in
five randomly placed 0.5 m × 0.5 m quadrats at each of the predetermined low, mid
and high tide locations. Less intensive searches were also conducted outside the
quadrats to identify whether flowering was occurring in other parts of the meadow.
If a flower was found, a permanent 5 m transect was set up (refer to Section 2.2.3 for
details) and marked at each end with labelled garden pegs. The GPS location of each
transect was recorded as a reference for subsequent monthly surveys. This process
was repeated for each tidal height across the three survey sites (Appendix 1,
Figures 3–6). Permanent survey transects were set up after flowering commenced,
enabling the selection of survey sites with conditions that were suitable for seagrass
growth and flowering.
At Waimea Inlet, an additional transect was set up across an isolated patch of dense
seagrass located very high on the shore (hereafter referred to as the ‘patch’ site). This
site was selected because large numbers of flowers has been observed in this
location during the 2021/22 summer (unpublished data).
3
https://environment.govt.nz/assets/Publications/Files/MFE_Coastal_Fact-Sheet-4.pdf
4
LiDAR (light detection and ranging) is a remote sensing method that generates three-dimensional information
about the Earth’s topography.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
6
Figure 3. Diagram demonstrating how the quadrats were arranged at each survey site.
Figure 4. Location of fine-scale flower survey sites and the temperature logger in Nelson Haven.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
7
Figure 5. Location of fine-scale flower survey sites and the temperature logger in Waimea Inlet.
Figure 6. Location of fine-scale flower survey sites and the temperature logger in Delaware Inlet.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
8
2.2.2.
Temperature loggers
At each estuary, two temperature loggers were installed within the seagrass meadow
(Table 1, Figure 7). Both loggers collected temperature information every 20 minutes:
one logger was placed 2 cm below the sediment surface to measure sediment
temperature and the other was positioned 2 cm above the sediment surface to
measure water and air temperature. Temperature data were usually downloaded each
month when the flower surveys were carried out. Only the water / air temperature data
results are presented here, as sediment logger temperature data was unavailable at
the time of writing this report.
Table 1. Coordinates and deployment dates for temperature loggers in the three estuaries.
Location
Deployment date
Latitude
Longitude
Nelson Haven
25/08/2022
-41.23409
173.31157
Waimea Inlet
06/09/2022
-41.29143
173.22113
Delaware Inlet
15/09/2022
-41.16817
173.44116
Figure 7. Nelson Haven temperature logger once deployed. Photo credit: Dana Clark.
2.2.3.
Surveys
The first flowers were observed in October at the ‘patch’ survey site in Waimea Inlet.
At all other survey sites, the first flowers were observed in November 2022 (refer to
Appendix 2 for dates). Once flowering commenced and permanent transects were
established, surveys were conducted monthly at all sites. This report presents data

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
9
collected between November 2022 and March 2023, although surveys were continued
after this period.
At the survey sites, five quadrats (0.5 m × 0.5 m) were evenly spaced along each
marked transect, approximately 50 cm apart. To ensure consistency across
monitoring periods, the leftmost quadrat, when orientated towards the shore, was
named ‘Quadrat 1’. The number of flowering shoots, number of spathes per shoot,
and stage of each spathe (Figure 8) within the quadrat were recorded. Initially, spathe
maturity was defined by four stages: 1) early-stage spathe with stigmas protruding; 2)
pre-seed (identified by the presence of stigmas and absence of seeds) or immature
seeds; 3) mature seeds; and 4) a spathe that has dropped its seeds. However, in the
field, there was an inconsistency between the assignment of Stages 1 and 2, as the
form that was originally thought to be immature seeds was later discovered to be
immature male flowers. Therefore, the stages were recategorised into the following
three groups (used in this report):
• pre-seed (previously Stages 1 and 2)
• developing seed (previously Stage 3)
• post-seed (previously Stage 4).
Figure 8. Examples of the flower stages, from the left: pre-seed (with stigmas out), developing seed
(containing immature seeds), post-seed (spathe that has dropped its seeds). Photo
credits: Cawthron Institute.
2.2.4.
Data analysis
Percent cover of seagrass at each survey site was quantified using a random point
count approach (Meese and Tomich 1992). For each replicate, percent cover of each
taxon and substrate was calculated by dividing the total number of assignments for
the category by the total number of dots (50) and then multiplying by 100 using
ArcMap 10.8.2 software and a customised tool (Figure 9). The options for taxa
included:
• seagrass – brown / black
• seagrass – covered with light brown film

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
10
• seagrass – green
• Ulva
• red algae.
The options for substrate type were:
• gravel
• mud / sand
• terrestrial sediment (identified by distinctive orange colouration)
• shell.
Data were recorded in a database and then converted to percent cover by dividing the
total number of points for a given taxon / substrate in a replicate by 50 and multiplying
it by 100.
Figure 9. Example of the random point count approach. Each of the green dots is manually
assigned a taxon or substate from the lists in the grey box. The software uses the name
of the photo to record the results in a database.
Differences in the number of flowering shoots and spathes between estuaries, tidal
heights and over time were assessed using separate Kruskal-Wallis rank sum tests.
Kruskal-Wallis was used because the data were not normally distributed. If significant
results were observed, Wilcox pairwise tests were undertaken. A nonparametric
quantile regression approach was used to test the relationship between the number of
flowering shoots and percent cover of seagrass, as well as between the number of
spathes and percent cover of seagrass. Quantile regression allows assumptions to be

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
11
discarded where it does not assume any distribution for the response and data are not
normal (Muggeo et al. 2013). This analysis was conducted using R programming and
the quantreg package (Koenker 2023).
2.3. Results
In October 2022, the first 17 flowering shoots (with a total of 19 spathes) were
recorded at the ‘patch’ site in the Waimea Inlet meadow. In November 2022, more
flowers were observed at this location than at any other site or time. On average, 56.2
flowering shoots and 84.4 individual spathes were detected per 0.25 m2 at the ‘patch’
site, compared to only between one and14 flowering shoots and between one and 15
spathes per 0.25 m2 at other sites (Figure 10). Unfortunately, the seagrass patch did
not persist throughout the summer, and by December 2022, there was only an
average of 5.4 flowering shoots and 7.2 spathes at the ‘patch’ site. By January 2023,
no flowers and very little seagrass remained. Seagrass at this survey site was not
present for much of the season, and the November data were not a good
representation of the wider population in the Waimea Inlet meadow; therefore, the
‘patch’ data were excluded from further analyses.
Figure 10. Number of individual spathes detected over the 5 months of surveys (November 2022–
March 2023) across the tidal heights (including the Waimea Inlet ‘patch’) for each
estuary. Note the difference in y-axis scales across plots.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
12
2.3.1.
Flowering shoots through time
Analysis of the number of flowering shoots over time illustrates a clear pattern of
flowering progression during the season and across the Nelson estuaries (Figure 11).
At all three meadows, the number of flowering shoots differed significantly between
months (Delaware Inlet: χ2 = 26.785, df = 4, p < 0.001; Nelson Haven: χ2 = 30.587, df
= 4, p < 0.001; Waimea Inlet: χ2 = 16.949, df = 4, p = 0.002). At the Delaware Inlet
meadow, the presence of flowers was recorded from November 2022 to March 2023,
with a total of 745 flowering shoots counted across all survey sites. For this meadow,
peak flowering occurred in December (total of 295 flowering shoots) and January
(total of 290 flowering shoots). Flowers were present in the Nelson Haven and
Waimea Inlet meadows (excluding the ‘patch’) from November 2022 to January 2023,
with peak flowering in December 2022 (totals of 19 and 45 flowering shoots in the
Nelson Haven and Waimea Inlet meadows, respectively).
The number of spathes present over time shows a similar pattern to that of flowering
shoots (Figure 12). For all meadows, there was a significant difference in the number
of spathes detected across months (Delaware Inlet: χ2 = 27.608, df = 4, p < 0.001;
Nelson Haven: χ2 = 30.581, df = 4, p < 0.001; Waimea Inlet: χ2 = 16.955, df = 4, p =
0.002). During peak flowering (December 2022 and January 2023) at the Delaware
Inlet meadow, a total of 968 spathes were counted across all sites; this is an average
of 1.7 spathes produced per flowering shoot. At the Nelson Haven (total of 27
spathes) and Waimea Inlet (total of 65 spathes) meadows, there was an average of
1.4 spathes per flowering shoot during peak flowering (December 2022).

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
13
Figure 11. Number of flowering shoots found in each quadrat over the 5 months of surveys
(November 2022–March 2023) at each estuary (Delaware Inlet, Nelson Haven and
Waimea Inlet [excluding ‘patch’ data]). Note the difference in y-axis scales across plots.
Figure 12. Number of spathes found in each quadrat over the 5 months of surveys (November
2022–March 2023) at each estuary (Delaware Inlet, Nelson Haven and Waimea Inlet
[excluding ‘patch’ data]). Note the difference in y-axis scales across plots.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
14
2.3.2.
Flowering shoots across tidal heights
Analysis of the number of flowering shoots and spathes present across different tidal
heights shows how variable flowering can be within an estuary. There was no obvious
trend in the number of flowering shoots or spathes across tidal heights at the Nelson
Haven and Waimea Inlet sites (Figures 13 and 14; Flowering shoots – Nelson Haven:
χ2 = 0.2619, df = 2, p = 0.8773; Waimea Inlet: χ2 = 0.16089, df = 2, p = 0.9227.
Spathes – Nelson Haven: χ2 = 0.19223, df = 2, p = 0.9084; Waimea Inlet: χ2 =
0.22187, df = 2, p = 0.895).
At the Delaware Inlet site, there was a difference in the number of flowering shoots
and spathes found across the tidal heights (Flowering shoots – χ2 = 12.693, df = 2,
p = 0.002. Spathes – χ2 = 13.438, df = 2, p = 0.001). A pairwise Wilcox test showed
that the high tidal zone tended to have a greater number of flowers and spathes
compared to the mid and low tidal heights in Delaware Inlet (p = 0.004).
Figure 13. Number of flowering shoots found in each quadrat over the three tidal heights (High, Mid,
Low) at each estuary (Delaware Inlet, Nelson Haven and Waimea Inlet [excluding ‘patch’
data]). Note the difference in y-axis scales across plots.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
15
Figure 14. Number of spathes found in each quadrat over the three tidal heights (High, Mid, Low)
separated into the three different estuaries (Delaware Inlet, Nelson Haven, and Waimea
Inlet [excluding ‘patch’ data]). Note the difference in y-axis scales across plots.
2.3.3.
Flower stages
The number of spathes at each of the three development stages fluctuated over the
season (Figure 15). Across all sites, peak flowering (i.e. highest number of spathes)
was observed in December 2022, along with a peak in pre-seed flowers. In January
2023, spathe numbers remained relatively high but proportionally more mature flowers
(developing seed and post-seed stages) were observed. A limited number of post-
seed spathes were recorded in March. This may have been due to the spathes
breaking off from the flowering shoot or beginning to break down, making it difficult to
identify them as post-seed spathes.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
16
Figure 15. Stacked bar graphs showing the total number of spathes at each stage over the 5 months
of surveys (November 2022–March 2023; excluding Waimea Inlet ‘patch’ data).
2.3.4.
Relationship between flowering shoots and spathes with the percent cover of seagrass
Peak flowering in December was further investigated to see whether there was a
relationship between the number of flowering shoots or spathes and the percent cover
of seagrass for each estuary (considering all tidal heights). The following figures
illustrate the number of flowering shoots (Figure 16) and the number of spathes
(Figure 17) occurring in December in relation to the percent cover of seagrass across
the three different tidal heights at Delaware Inlet, Waimea Inlet and Nelson Haven.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
17
Figure 16. Number of flowering shoots found in December at Delaware Inlet, Waimea Inlet and Nelson Haven plotted against the percent cover of seagrass at three
different tidal heights: high zone (orange dots), mid zone (green dots), and low zone (black dots). Each coloured point represents a 0.25 m2 quadrat. A
0.5 quantile regression line was applied. Note the difference in y-axis scales across plots.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
18
Figure 17. Number of spathes found in December at Delaware Inlet, Waimea Inlet and Nelson Haven plotted against the percent cover of seagrass at three different
tidal heights: high zone (orange dots), mid zone (green dots), and low zone (black dots). Each coloured point represents a 0.25 m2 quadrat. A 0.5
quantile regression line was applied. Note the difference in y-axis scales across plots.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
19
At Delaware Inlet, the quantile regression between number of flower shoots and
percent cover of seagrass had a negative slope (flowering shoots = 31.8 + -0.35
seagrass cover
5
). The number of flowering shoots was higher in the high tidal zone in
lower percent cover of seagrass, with an average of 47.2 ± 22.3 flowering shoots
(±SE, n = 5). However, the regression was not significant (p = 0.694). Similarly, the
number of spathes was higher in the high tidal zone in lower percent cover of
seagrass (spathes = 31.8 + -0.35 seagrass cover), with an average of 65.6 ± 34.0
spathes (±SE, n = 5), but the regression was not significant (p = 0.780).
At both Waimea Inlet and Nelson Haven, there were negligible relationships between
number of flowering shoots and percent cover of seagrass (intercept = 0; p = 1), and
similarly, no relationship between the number of spathes and percent cover of
seagrass across tidal heights (intercept = 0; p = 1).
2.3.5.
Temperature data
Data captured by the temperature loggers demonstrate the degree of fluctuation
throughout the summer period (Figures 18–20 and Appendix 3). The maximum
water / air temperature for 2022/23 summer was 37.9 °C recorded in the Nelson
Haven meadow on 20 January 2023. The lowest temperature was 1.5 °C, recorded in
the Waimea Inlet meadow on 6 October 2022. This was an extreme low seen at all
three sites. All three graphs show a slow increase in overall temperature from
October to February; the temperature then began to decrease again from March for
Delaware Inlet and Nelson Haven (Figures 18–20). Maximum temperatures at
Delaware Inlet tended to be lower than in Nelson Haven and Waimea Inlet, but
average and minimum temperatures were similar across all three meadows
(Appendix 3). Between September (when no flowers were recorded) and November
(when flowers were observed at every site) the average monthly temperature
increased by 5.2 °C in Nelson Haven and Waimea Inlet, and 4 °C in the Delaware
Inlet meadow.
5
For both flowering shoots and number of spathes, the slope and intercept values are reported as the line
equation y = β0 + β1x where β0 is the intercept, β1 is the slope, and x is the seagrass cover.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
20
Figure 18. Delaware meadow maximum, mean and minimum temperatures (°C) from 14 September
2022 to 3 March 2023.
Figure 19. Nelson Haven meadow maximum, mean and minimum temperatures (°C) from 25 August
2022 to 26 March 2023.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
21
Figure 20. Waimea Inlet meadow maximum, mean and minimum temperatures (°C) from
6 September 2022 to 11 January 2023.
2.4. Discussion
The findings from our fine-scale study of seagrass flowering in Nelson estuaries
improves the understanding of when and where flowering is occurring. Flowering
began in November 2022 in all three meadows and continued until January 2023 in
Nelson Haven, February in Waimea Inlet and at least March in Delaware Inlet.
Flowers were also observed at the ‘patch’ site in the Waimea Inlet in October 2022,
but no flowers were found in the rest of the meadow at that time.
Information on flowering timing generally aligns with observations reported for flower
surveys in Nelson Haven and Delaware Inlet over the 2021/22 summer (Hindmarsh
and Hooks 2022). In both years, a decline in flowering shoots was observed between
November and January at Nelson Haven, but no significant difference in flowering
was detected between November and January at Delaware Inlet. Peak flowering was
at a similar time across the two seasons (e.g. November to January). Finding similar
flowering patterns, or trends such as these, allows us to be more confident at
predicting when sexual reproduction will take place. Our results also generally align
with the study by Dos Santos and Matheson (2017) in Tauranga Harbour (North
Island, Aotearoa New Zealand), which reported peak flowering between November
and February.
Applying these results to the seed-based restoration context, future collection efforts
can now target the period when flowering peaks and before seeds begin to mature
and drop out of the flowers (late November and December). This is the period when

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
22
most flowers are present and before the seeds begin to mature and drop out of the
flowers.
As well as identifying the optimal time of year to collect flowers, it is also important to
understand the best places to find flowers. Seagrass flowering can be very patchy
across a meadow (Hindmarsh and Hooks 2022), so information on what drives this
patchiness is important to guide flower collection efforts. Our study showed that
flowering did not differ between tidal heights in Nelson Haven and Waimea Inlet;
however, at Delaware Inlet, significantly more flowers were found higher on the shore
than lower down. The 2021/22 season study also documented consistent flowering
across tidal heights in Nelson Haven, and variation across tidal heights at Delaware
Inlet. However, the pattern of variation reported by the 2021/22 study was the inverse
of our results – more flowers were found at the low tide site than the high tide site. In
more stressful environmental conditions, e.g. high temperatures, it is common for
plants to allocate more resources to reproduce sexually, resulting in greater
production of flowers (McDonald et al. 2016). Elevated stress, arising from drivers
such as desiccation and / or photoinhibition
6
in the higher tidal zones, may have
contributed to the increased number of flowering shoots found in the high tide site at
Delaware Inlet in 2022/23. However, the variable nature of flowering may mean that a
sampling site with high numbers of flowers was selected by chance for the recent
study.
High numbers of flowers were also observed at the ‘patch’ site in the Waimea Inlet,
which was located very high on the shore and separated from the main meadow. This
site also had very high numbers of flowers during the summer of 2021/22
(unpublished data). Unfortunately, there was a rapid decline in seagrass cover at this
site in December 2022. One theory is that increased sediment deposits in the area
(possibly a legacy effect of the August 2022 floods) smothered the blades and
flowers, and thus they were no longer visible on the surface. The seagrass remained
‘buried’ until at least March 2023 (the end of the observation period), with less
seagrass visible at each survey.
Since the water / air temperature increased 4–5 °C over the flowering season, it is
reasonable to question whether this may be one of the cues that trigger flowering,
although daylight length is a confounding factor. However, a laboratory study by
Ramage and Schiel (1998) found that in sediment cores containing seagrass kept at
5 °C, more spathes were detected than in samples kept at 15 and 25 °C. It is unlikely
that a single factor is attributable to flowering, as temperature can be influenced by
both the time of day and tidal height. To determine flowering cues, further
investigation into other environmental factors, such as light levels, day length and
salinity, is required.
6
This is when too much light reduces the photosynthetic capacity of the plant.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
23
Our study demonstrated that at all three estuary sites, the number of flowering shoots
or spathes was not generally related to percent cover of seagrass. However, on the
high shore in Delaware Inlet, more flowering shoots and spathes were found in areas
of low seagrass percent cover. In 2021/22, Hindmarsh and Hooks (2022) also tested
for a relationship between flowering shoot abundance and percent cover of seagrass.
Their study did not identify a relationship in Delaware Inlet but did observed more
flowers in areas of lower seagrass percent cover in Nelson Haven. Conversely, the
study by Dos Santos and Matheson (2017) in Tauranga Harbour reported significantly
higher seagrass cover in sites containing seagrass flowers than in sites with no
flowers. It should be noted that both the current study and the work by Hindmarsh and
Hooks (2022) are based on very few data points; therefore care should be taken when
drawing conclusions from these results. Furthermore, the three particularly high
values at the Delaware and Waimea Inlets may be biasing the results of the
regression. We recommend analysing the data from both the Nelson studies
alongside any future work to draw more robust conclusions.
It would be beneficial to further investigate the factors that influence the spatial and
temporal variability of seagrass flowering to improve the efficiency of the flower
collection component of seed-based restoration. For example, analysis of the
relatedness of plants could reveal whether a patch of flowers is produced by the same
rhizome. This knowledge, along with additional information about flowering cues for
Z. muelleri, will forward the development of targeted flower collection protocols.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
24
3. FLOWER COLLECTION
3.1. Study aim
This phase of the study aimed to develop methods for seagrass flower collection,
which is the first stage of seed-based restoration. In the future, these processes will
be informed by the results of flower surveys (Section 2) to ensure that flowers are
collected at the height of the season.
3.2. Methods
Our flower collections were focused on the Nelson Haven seagrass meadow, as this
is the most extensive meadow and access to the site is straightforward. Meadows
were visited at low tide and individual flowering shoots were identified and collected.
To minimise impacts to the plants and beds, flowering shoots were picked below the
peduncle, retaining associated leaves, without removing any of the rhizomes. We also
took care not to pick too many flowers from any single area of the meadow, and the
amount collected was within permitted limits. No attempt was made to differentiate
between flowering stages while picking. Once collected, the flowering shoots were
placed in a mesh bag. The bag was placed in a 3-litre bucket containing seawater to
avoid desiccation during transport.
At the laboratory, the collected material was initially weighed (wet weight), then each
flowering shoot was counted, and the development stage was identified and recorded
(see Section 2.2.3, Figure 8). Any material that did not contain a flowering shoot was
weighed again and then discarded. The weight of discarded material was subtracted
from the initial weight to calculate the net weight of the flowers. Once the flowering
shoots were counted, assessed for stage, and weighed, they were placed in one of
two pre-prepared seed extraction tanks (see Section 4.2).
3.3. Results and discussion
Flower collections were carried out over nine visits between November 2022 and
January 2023, and a total of 3,456 flowers were picked (Figure 21A). The number of
flowers collected per hour per outing was influenced by three factors: 1) number of
people collecting the flowers (i.e. collectors); 2) collectors’ experience; and 3) number
of available flowers. Across all outings, an average of 77.8 flowers were picked per
person per hour. Two flower collection events had particularly high success rates.
This was likely due to a higher number of pickers and their increased experience,
along with sufficient flower availability. The most flowers picked per person per hour
occurred on 15 December 2022 (Figure 22) by one experienced picker at a location
with abundant flowers. Fewer spathes were collected during the three outings that

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
25
occurred in January, likely due to the decrease in the number of flowers (see Section
2.3). Most of the flowers collected on each outing were recorded as being at the pre-
seed stage (Figure 21B). Collection of pre-seed flowers may have been favoured, as
they are often easier to visually detect.
This phase of the project offered a good opportunity to involve the local community.
For several flower collection outings, interested community members were invited to
assist with collection. For future studies, it will be beneficial to include more
community members to help increase the number of flowers that can be collected at
each event (taking into account actions to reduce any impact of flower collection on
the individual plants and meadow – refer to Section 3.2). It will be important to ensure
that community flower collection days are held during peak flowering times, as it can
be difficult to teach the collection methods when flower numbers are limited.
Figure 21. (A) Number of flowering spathes collected by stage at each collection event in Nelson
Haven between December 2022 and January 2023. (B) Number of spathes collected
within each flower stage category.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
26
Figure 22. Representation of effort for the flower collection. Number of spathes picked per person
per hour for each collection event in Nelson Haven.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
27
4. SEED EXTRACTION
4.1. Study aim
This phase of the study aimed to develop an efficient method for extracting seagrass
seeds from collected flowers.
4.2. Methods
Seagrass flowers were held in custom designed tanks until maturation, when seeds
dropped from the flowers and could be collected. Two tapered tanks
(0.8 m × 0.55 m × 1 m) on stilts were set up (Figure 23) and a 1 cm valve installed to
drain the seawater from the tank through a double-layer sieve. The top and bottom
layers of mesh were 1100 nm and 250 nm, respectively. Seawater draining from the
tanks was collected in a bucket and pumped back up into the main tank (i.e.
recirculated). The bucket was fitted with an overflow system to prevent flooding and to
ensure that the seeds could potentially be refloated from the sieve in the event of
pump failure. A loop of garden weeper hose was fixed inside the tank and attached to
an air pump to create bubbles in the water. The bubbles ensured aeration of the water
and agitation of the seagrass material to promote seed drop. The flowers were placed
on 3 mm mesh cloth that sat over the top of the tank, allowing the flowers to be
submerged in the aerated seawater. Each tank had a clear plastic lid to limit
evaporation and rainwater ingress.
7
When released from the flower, the negatively buoyant seagrass seeds fell through
the mesh to the bottom of the tank, were flushed out and collected in the sieve. There
was always water draining through the sieve to ensure that the seeds did not dry out.
7
Lower salinity conditions can trigger the germination of seagrass seeds.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
28
Figure 23. Tank set-up (‘seagrass spas’). One tank was inside a laboratory (left) in controlled
environmental conditions. The outdoor tank (right) was exposed to more extreme
fluctuations in environmental conditions, representing conditions in the natural
environment.
One tank was set up inside the condition-controlled laboratory and the other was
outside (see Section 4.3.2). This method allowed us to test whether seed
development progressed differently in stable light and temperature conditions
(indoors), compared to those developing in natural, fluctuating conditions (outdoors).
An additional light was suspended above the inside tank and was automated to turn
on and off to mimic day and night conditions (8 hours of darkness and 16 hours of
light); although even when the light was off, there was a low level of light due to
another experiment that was operating nearby. The temperature was set to 19+/-1 °C.
Light and temperature conditions for the outdoor tank were not controlled. To measure
temperature, environmental dataloggers (of the brand HOBO) were placed in both
tanks, at the water's surface and on the bottom of the mesh, and measurements were
recorded over a week in January. Salinity was measured when ‘fresh’ seawater was
added to the tank and 1 week later, before tank cleaning.
The recirculating water system was cleaned weekly to refresh the water and remove
any algal growth. Before the tanks were cleaned, the top mesh containing the

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
29
seagrass flowers was removed and placed in a bucket of seawater. Tanks were
drained through the sieve set-ups to ensure collection of any seeds that were still
suspended in the water column. The sides of the tanks were washed and then the
bottom valve was closed and the tanks refilled with fresh seawater. The top mesh and
seagrass were returned to the tanks, the valves opened, and pumps turned back on.
Tank lids were also wiped down to ensure maximum light penetration.
In the inside tank, four spathes were selected at an early pre-seed stage, separated
into a small mesh bag within the tank, and observed over time. This method was used
to check whether development progressed and seeds were produced in the tank
environment.
4.3. Results and discussion
4.3.1.
Flowers and seeds
Flowers collected over the course of the study were distributed between the two
tanks, with similar proportions of development stages added to each tank (Figure 24).
The inside tank was set up first and therefore received more spathes (2270) than the
outside tank (1186) (Figure 25).
Figure 24. Proportions of seagrass spathes at each stage in each tank (inside and outside).

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
30
Figure 25. Number of seagrass spathes added to the tanks (inside and outside) and number of
seeds collected from each tank.
One of the spathes we tracked over time in the inside tank matured and possibly
produce a seed (Figure 26). However, there was no evidence that seeds in the other
three spathes matured; these spathes were observed to die and began to break
down.
Figure 26. Development of a seagrass spathe in the inside tank. (1) 23 January. (2) 24 January. (3)
27 January. (4) 31 January.
The average number of seeds produced per spathe was 0.2 in both the inside and
outside tank (Figure 24). The number of seeds produced from only the developing
seed stage spathes was 0.8 seeds per spathe for the inside tank and 0.6 seeds per

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
31
spathe for the outside tank. We observed that each spathe contained, on average,
four female flowers, and therefore, had the potential to produce up to four seeds.
Thus, our study shows that not every spathe observed at the ‘developing seed stage’
produced a seed. This result, along with our observations of development for the four
individual spathes (see above), suggests that seed development within the tank
environment was compromised. Seed extraction success was lower than expected,
and it is possible that the seeds may have been trapped in the seagrass material at
the top of the tank, preventing the process of dropping out into the sieve. Additional
agitation (e.g. via more vigorous air bubbling) may have been required to separate the
seeds from the seagrass material. More intensive trials are required to better
understand seed development in tank environments and to determine the factors that
limit progression.
Seed development may have been influenced by a pump failure that occurred midway
through the seed extraction trials in the outside tank. The water drained out of the tank
and the seagrass dried out, preventing further (potential) seed extraction. This issue
also impacted the inside tank bubbling system, resulting in water stagnation and no
plant material agitation, potentially influencing seed drop. For future trials, the tank
design needs to be altered to avoid desiccation in the event of a pump failure.
4.3.2.
Tank conditions
Temperature and light levels fluctuated very little in the inside tank. Temperature
remained between 19.5 °C and 20.2 °C. Light levels never reached zero, as there
were experiments taking place on the other side of the lab that required constant, low
levels of light.
In comparison, light exposure and water temperature of the outside tank fluctuated
between day and night (Figure 27). Water temperature at the surface ranged from
18 °C to 47 °C over the course of the week of monitoring (Figure 27, Table 2). The
high temperatures resulted in elevated water evaporation, which was reflected by an
increase in salinity of 13.6 ppt (34.4 ppt to 48 ppt). Water salinity in the inside tank
increased by 3.68 ppt (Table 3).
Table 2. Minimum, maximum, and mean temperature (°C) and light (lux) recorded at the surface
and bottom of the inside and outside tanks over the course of one week (19–25 January
2023).
Temperature (°C)
Light (Lux)
Tank
Min
Max
Mean
Min
Max
Mean
Inside bottom
19.5
20.2
19.9
64.6
430.6
196.4
Inside surface
19.5
20.2
19.8
613.5
3788.6
1941.3
Outside surface
18.2
47.3
25.2
0
275557.0
47055.0
Outside bottom
18.2
33.0
24.3
0
34444.7
4028.9

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
32
Figure 27. Temperature (°C, plot above) and light (lux, plot below) conditions of the inside and
outside tanks. A HOBO datalogger was placed on the bottom of the mesh holding the
seagrass (i.e. bottom) and at the surface of the water (i.e. top) in both the inside and
outside tanks. These loggers measured the changes in temperature and light for a week
from 19–24 January 2023.
Table 3. One-off salinity measurements from the inside and outside tanks. The first measurement
was recorded when new seawater was added to the tanks (19 January 2023). The
second measurement was taken the following week, before the tanks were cleaned (24
January 2023). ppt = parts per thousand.
Tank
Salinity (ppt)
at time of water change
Salinity (ppt)
after 1 week
Inside tank
34.0
37.6
Outside tank
34.4
48.0

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
33
5. SEED STERILISATION AND STORAGE
5.1. Study aim
This phase of the study aimed to develop effective methods to sterilise and store
seagrass seeds.
5.2. Methods
Sterilisation methods were adapted from unpublished methods used by collaborators
at Central Queensland University and Deakin University in Australia.
Seeds were collected weekly from the tank sieves (see Section 4) and sterilised via
the following process: the contents of both sieves (from the inside and outside tanks)
were placed on separate white 30 cm × 15 cm trays for sorting. Seeds were manually
separated from plant material using sterilised tweezers and pipettes. The seeds were
then placed into a 100 ml beaker with autoclaved seawater. Any flower shoots that
were found in the sieves were returned to the tank, and the remaining organic matter
was discarded. Once separated, seeds were counted and then graded against a seed
colour chart provided by Manuja Lekammudiyanse from Central Queensland
University.
Once all the seeds were separated and counted, the seawater was drained from the
beaker (Figure 28A) and a 70:30 solution of ethanol and autoclaved seawater was
added. Seeds were soaked for 2 minutes, then drained and a 10:90 solution of bleach
and autoclaved seawater was added (Figure 28B). During preparation, both solutions
precipitated to form a cloudy liquid, but they were deemed suitable for use. The
beaker was then agitated for 5 minutes using a magnetic stirring plate and sterilised
stirring bar (Figure 28C). Finally, the seeds were rinsed in autoclaved seawater. The
above ethanol and bleach soaking and stirring process, as well as the seawater rinse,
was repeated three times to ensure adequate sterilisation.
Figure 28. (A) Seagrass seeds extracted from the tank before being sterilised. (B) Seeds being
soaked in ethanol solution. (C) Seeds being mixed in bleach solution.
A. B. C.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
34
Once sterilised, the seeds were divided into sterilised 50 ml screw top pottles,
ensuring that the seeds were not overlapping on the bottom. The pottles were then
half-filled with autoclaved seawater and placed in a container covered in aluminium
foil to block light. Half were stored in a fridge at 4 °C, and half in an ambient
temperature-controlled room at 17 °C. Each pottle was labelled with the date of
sterilisation, the method of storage (‘Fridge’ or ‘Ambient’), and which tank the seeds
had been collected from (‘Inside’ or ‘Outside’). Once a week, the autoclaved seawater
in each container was replaced and the seeds were checked. In the first 2 months of
storage, seeds showing any signs of fungal growth were removed (e.g. during the
weekly checks), and the remaining seeds were put through one round of sterilisation
before returning to storage. On 15 February 2023, all seeds in storage were put
through a second compete sterilisation in an attempt to stop mould growth. From this
point, if mould growth was observed, the seeds were only rinsed in autoclaved
seawater, as the long-term effect of multiple bleach and ethanol treatments is not
known.
5.3. Results and discussion
In total, 516 and 89 seeds were collected from the inside and outside tanks,
respectively. The difference between these totals can partially be attributed to
stocking differences (fewer seeds in the outside tank, refer to Figure 24) and to
seagrass desiccation when the pump failed on the outdoor tank. Although flowers
added to each tank comprised similar proportions of development stages, the colours
of the seeds collected differed between the tanks (Figure 29). The outside tank
produced a greater proportion of black and brown seeds compared to the inside tank.
Darker seed colour may be indicative of more advanced maturity and, in this trial,
could be associated with higher light levels and / or water temperature in the outdoor
tank. Timing of flower collection may be another factor contributing to differences in
seed colour / maturity stage, as the outside tank contained proportionally more flowers
collected slightly later in the season.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
35
Figure 29. Proportions of colours observed in seagrass seeds extracted from the inside and outside
tanks.
Mould development was observed throughout the seed storage period, particularly for
seeds kept at ambient temperature. On one occasion, white filamentous growth was
detected within pottles, leading to the seeds and pottles being sterilised. The
presence of mould may be an ongoing occurrence with the seeds collected for the
current study. The causes of mould development are not known and the impact on the
seeds is also unclear. We recommend investigation into the effects of sterilisation and
fungal growth on seed viability. Our Australian collaborators indicated that fungal
growth and unusual smells are common during seed storage, necessitating weekly
water exchange.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
36
6. SEED GERMINATION
6.1. Study aim
This phase of the study aimed to test whether the seagrass seeds we collected were
viable and whether collection and storage conditions influence seed germination.
6.2. Methods
A pilot-scale germination trial was conducted to test the effect of four treatments on
seed viability. Ten seagrass seeds of a brown colour were selected from each of the
different treatment pottles: 1) inside tank stored at ambient temperature; 2) inside tank
stored in the fridge; 3) outside tank stored at ambient temperature; and 4) outside tank
stored in the fridge. The germination process was adapted from Stafford-Bell et al.
(2016) and began by exposing seeds to sterile fresh water for 24 hours (Figure 30A).
This freshwater pulse is understood to break the dormancy of the seeds. After 24
hours in fresh water, seeds were returned to autoclaved seawater and kept in Petri
dishes in dark, ambient temperature conditions (17 °C). Over the next 14 days, the
seeds were monitored to track the progress of germination (Figure 30B). Germination
was defined as emergence of the first cotyledon from the seed coat (Figure 30C).
After 14 days, some seeds did not show signs of germination. These seeds were put
through a second freshwater flush for 24 hours. At this time, eight seeds that had not
been sterilised or stored were also put through their first freshwater flush. This
process was used to determine the germination rates of seeds that had not been
subjected to any storage treatments.
Observations of the seeds continued for another 7 days before mould was detected in
all five trials. At this point, the trial was terminated, and all seeds that did not show any
signs of germination or rot were put through a round of sterilisation and stored in a
separate container.
Germinated seeds were transferred into small pottles containing estuarine sediment
and seawater, kept in controlled laboratory conditions (refer to Section 4.2 for details
on the controlled conditions), and monitored for further development.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
37
Figure 30. (A) Seagrass seeds being put through an overnight freshwater flush. (B) Seeds in Petri
dish, monitored for germination. (C) Germinated seed with emerging cotyledon.
6.3. Results and discussion
Of 48 seeds flushed with fresh water and monitored for germination, only four seeds
germinated (approximately 10% germination rate). We considered this a positive
result for initial trials. The seeds that germinated had all been stored at ambient
temperature (two from the inside tank and two from the outside tank). This
demonstrates that both the inside and the outside tank can produce viable seeds,
despite the differences in conditions. However, note that outside of this freshwater
flush trial, two seeds were found to spontaneously germinate in the fridge storage.
A potential explanation for lack of germination in refrigerated seeds is thermal shock,
as seeds stored at 4 °C were immediately immersed in 17 °C fresh water. This
change in temperature may have been too extreme for the seeds We recommend
removing seeds from refrigeration early and allowing slow adjustment to room
temperature (possibly overnight) before freshwater treatment.
One germinated seed successfully developed into a seedling (Figure 31). To our
knowledge, this is the first time that seagrass has been grown from seed in laboratory
conditions in Aotearoa New Zealand, and the second time that laboratory-based seed
germination has been achieved (one seed was germinated by Hindmarsh and Hooks
in 2022). These results provide support for the potential of seed-based restoration of
seagrass in Aotearoa New Zealand.
A. B. C.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
38
Figure 31. Germinated seagrass seed grown into a seedling in the laboratory.
Seed germination trials for Z. muelleri on the southeast coast of Victoria, Australia
reported a similar initial success rate in the first season (10%; Tan 2022). An
improved rate was reported for the second season, but these results cannot be
compared with our study, as the trials were field-based and seedling emergence was
considered a successful outcome. Our laboratory-based trials instead defined success
as cotyledon development; however, cotyledon formation does not always lead to
establishment of a seedling (Cumming et al. 2017).
We recommend that future laboratory trials consider testing germination rates in
sediment medium and define success as shoot emergence from the substrate. This
approach would provide a better indication of seed survival and be more consistent
with future field-based trails. Our study’s low germination rate may be attributed to
missing environmental cues (e.g. temperature, light, or salinity threshold or pattern) or
conditions (e.g. sediment microbial community), which individually or cumulatively
contribute to the process of seed germination. Many studies have documented that
temperature and salinity are key factors in higher plant seed germination (Cumming et
al. 2017). We recommend that future trials test the influence of potential cues,
particularly temperature, on germination. As trials progress into the field, this
additional research will add a new set of variables that will need to be monitored.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
39
7. OVERALL CONCLUSIONS
Seagrass meadows in Aotearoa New Zealand are declining and research on
appropriate restoration techniques is limited. Our study tested the feasibility of key
components of seed-based restoration, using Nelson estuaries as a case study
region.
Flower surveys carried out over the 2021/22 and 2022/23 summers indicated that
peak flowering of seagrass in the Nelson Region occurs in late November and
December. We recommend, therefore, that flowers for seed-based restoration are
collected during this period.
Flower collection and subsequent seed extraction was successful, and the findings
support the development of seed-based restoration methods. We found that flowers
drop mature seeds whether seed extraction tanks are located in a controlled
laboratory environment or outside where tanks are exposed to the elements.
However, it is yet to be determined whether the quality of seeds differs between these
environmental conditions. Further refinement of tank set-up is required to ensure
redundancy and avoid desiccation in the event of pump failure.
All seeds collected throughout this study were successfully sterilised and stored in
either refrigerated or ambient conditions. Sterilisation of seeds prior to storage did not
always prevent mould, which tended to be more prevalent in ambient conditions.
While this leads to the assumption that refrigerated storage is optimal, our trials only
achieved successful germination of seeds that were stored at ambient temperature.
However, this outcome may have been due to storage methods, or the methods used
to test germination.
Our study and the results represent a positive step forward in developing seed-based
methods for seagrass restoration in Aotearoa New Zealand. Our recommendations for
developing these methods further are summarised below.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
40
8. RECOMMENDATIONS FOR FUTURE STUDIES
• Further investigate the factors that influence the spatial and temporal variability of
seagrass flowering to improve the efficiency of the flower collection component of
seed-based restoration. For example, analysis of the relatedness of plants could
reveal whether a patch of flowers is produced by the same rhizome.
• Further investigate the cues (e.g. light levels, day length, salinity and stress) for Z.
muelleri flowering.
• Standardise the size and quality of quadrat photos collected in fine-scale studies
to improve efficiency and consistency of analyses (i.e. percent cover of seagrass
and other taxa as well as of substrates).
• Refine the tank system to develop a design that prevents desiccation or
stagnation in the event of any equipment failure.
• Improve characterisation of the environmental conditions in each of the tank set-
ups.
• Conduct a focused study of seed development within tanks. Questions to consider
include:
o Can seed fertilisation occur inside the tank?
o Do tank conditions provide adequate resources for pre-seed flowers to
produce mature seeds when detached from rhizomes?
o Can flowering shoots continue to photosynthesise in the tank environment?
• Conduct seed quality assessments, comparing the success (germination and
emergence) of seeds collected from tanks located inside and outside.
• Test the effects of sterilisation processes and storage conditions (including
presence of mould) on germination and subsequent growth of seeds.

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
41
9. ACKNOWLEDGEMENTS
Cawthron supervisors
Firstly, Demi and Irisa would like to thank their supervisors Dana Clark, Anna
Berthelsen and Daniel Crossett from the Cawthron Institute. Their support and
guidance over the course of the project (10 weeks for Irisa and 20 weeks for Demi)
was invaluable.
Acknowledgements for other Cawthron staff
We would like to thank Veronica Beuzenberg for helping us source the equipment
required throughout the summer, as well as for allowing us to use her laboratory
space and for preparing our sterilisation solutions. We also thank Fiona Gower for
providing equipment from the Taxonomy Laboratory. We would also like to thank:
Nathan Carmody – the engineer that helped us gather parts and assemble the seed
extraction tanks, Andy Selwood who helped with sourcing lab space for the tanks,
Lisa Floerl who helped with mapping, and Paul Barter who helped with the PAR
sensors and the dots-on-rocks analysis.
Fieldwork contributions
We would like to thank Lucie Bizzozero and Akash Pai for their help with fieldwork.
We also had several community volunteers who helped us pick flowers, including Don
Morrisey, Phil-Garnock Jones, Kindra Douglas, Vikki Ambrose, Ashleigh Bunning and
Anna Bradley.
Collaborators in Australia
We would like to give a huge thanks to Emma Jackson (Central Queensland
University) and Craig Sherman (Deakin University) for being so generous with their
time and expertise. Emma and Craig hosted Anna and Dana for 10 days and were
instrumental in the success of this research. We would also like to thank their students
Lucy Coals, Abbie Wookey, Laney Callahan, Elizabeth Keehner, Elizabeth Andrews,
Manuja Lekammudiyanse and Anna Hegarty, who have also generously shared their
knowledge with us.
Funding and in-kind support
Our project was made possible thanks to the support of local iwi and the generous
contributions from the Westpac New Zealand Government Innovation Fund, Port
Nelson, OneFortyOne and Friends of Nelson Haven and Tasman Bay. We also thank
mana whenua for their support of our project. Our collaboration with Emma Jackson
and Craig Sherman was supported through a Royal Society Catalyst Grant. We thank
Vikki Ambrose for the original inspiration for this project. Previous work that laid the
foundation for this project was funded by the Cawthron Institute, Nelson City Council
and Ngā Pae o te Māramatanga.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
42

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
43
10. APPENDICES
Appendix 1. Survey site coordinates.
Location
Latitude
Longitude
Nelson Haven
Low
-41.23502
173.31184
Mid
-41.23447
173.31201
High
-41.23378
173.31201
Waimea Inlet
Low
-41.17466
173.13245
Mid
-41.17484
173.13266
High
-41.17502
173.13290
Patch
-41.17496
173.13369
Delaware Inlet
Low
-41.16763
173.44128
Mid
-41.16796
173.44159
High
-41.16808
173.44215

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
44
Appendix 2. Dates of flower surveys and flower presence.
Location / Month
Survey Dates
Flowers
Nelson Haven
September
07/09/2022
No Flowers
October
10/10/2022
No Flowers
November
17/11/2022
Present
December
06/12/2022
Present
January
11/01/2023
Present
February
03/02/2023
No Flowers
March
02/03/2023
No Flowers
Waimea Inlet
September
06/09/2022
No Flowers
October
09/10/2022
Present – Patch only
November
18-23/11/2022
Present
December
05/12/2022
Present
January
12/01/2023
No Flowers
February
07/02/2023
No Flowers
March
06/03/2023
No Flowers
Delaware Inlet
September
15/09/2022
No Flowers
October
17/10/2022
No Flowers
November
09/11/2022
Present
December
01/12/2022
Present
January
10/01/2023
Present
February
02/02/2023
Present
March
03/03/2023
Present

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
45
Appendix 3. Summary of estuary temperature data: minimum, mean and maximum air /
water temperatures (°C) recorded in Nelson seagrass meadows over the
2022/23 summer.
Location
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Nelson Haven
Min
2.4
1.6
9.2
10.6
12.6
11.3
8.9
Mean
13.1
14.8
18.3
20.3
21.5
20.7
19.8
Max
27.0
31.0
29.6
35.9
37.9
37.2
35.6
Waimea Inlet
Min
2.5
1.5
8.3
10.4
11.5
Mean
13.1
14.8
18.3
20.4
21.3
Max
23.3
27.4
30.0
33.6
36.3
Delaware Inlet
Min
4.5
1.6
8.6
9.8
10.7
12.5
14.6
Mean
13.4
14.1
17.4
19.3
20.2
20.1
20.6
Max
21.9
24.4
27.0
30.8
29.8
28.4
29.4

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
46
11. REFERENCES
Bester K. 2000. Effects of pesticides on seagrass beds. Helgoland Marine Research.
54(2–3):95–98.
Bull JS, Reed DC, Holbrook SJ. 2004. An experimental evaluation of different
methods of restoring Phyllospadix torreyi (surfgrass). Restoration Ecology.
12(1):70–79.
Ceccherelli G, Campo D, Milazzo M. 2007. Short-term response of the slow growing
seagrass Posidonia oceanica to simulated anchor impact. Marine
Environmental Research. 63(4):341–349.
Clark D, Berthelsen A. 2021. Review of the potential for low impact seagrass
restoration in Aotearoa New Zealand. Nelson: Cawthron Institute. Cawthron
Report No. 3697. Prepared for Nelson City Council.
Conacher C, Poiner I, O'donohue M. 1994. Morphology, flowering and seed
production of Zostera capricorni Aschers. in subtropical Australia. Aquatic
Botany. 49(1):33–46.
Cumming E, Jarvis JC, Sherman CDH, York PH, Smith TM. 2017. Seed germination
in a southern Australian temperate seagrass. PeerJ. 5:e3114.
de Lange PJ, Rolfe JR, Barkla JW, Courtney SP, Champion PD, Perrie LR,
Beadel SM, Ford KA, Breitwieser I, Schonberger I, et al. 2018. Conservation
status of New Zealand indigenous vascular plants, 2017. Wellington:
Department of Conservation. New Zealand Threat Classification Series 22.
Dos Santos VM, Matheson FE. 2017. Higher seagrass cover and biomass increases
sexual reproductive effort: a rare case study of Zostera muelleri in New
Zealand. Aquatic Botany. 138:29–36.
Erftemeijer PLA, Robin Lewis RR. 2006. Environmental impacts of dredging on
seagrasses: a review. Marine Pollution Bulletin. 52(12):1553–1572.
Ferretto G, Glasby TM, Poore AGB, Callaghan CT, Housefield GP, Langley M,
Sinclair EA, Statton J, Kendrick GA, Vergés A. 2021. Naturally-detached
fragments of the endangered seagrass Posidonia australis collected by citizen
scientists can be used to successfully restore fragmented meadows. Biological
Conservation. 262:109308.
Gibson K, Marsden ID. 2016. Seagrass Zostera muelleri in the Avon Heathcote
Estuary / Ihutai, summer 2015–2016. Christchurch: University of Canterbury.
Prepared for Ihutai Trust. Estuarine Research Report 44.
Hindmarsh B, Hooks R. 2022. Research to inform seagrass restoration in Aotearoa.
Nelson: Cawthron Institute. Cawthron Report No. 3775.
Koenker K. 2023. quantreg: quantile regression. R package version 5.95. [accessed
12 October 2023]. https://CRAN.R-project.org/package=quantreg

CAWTHRON INSTITUTE | REPORT NO. 3994 DECEMBER 2024
47
Macreadie P, Baird M, Trevathan-Tackett S, Larkum A, Ralph P. 2014. Quantifying
and modelling the carbon sequestration capacity of seagrass meadows – a
critical assessment. Marine Pollution Bulletin. 83(2):430–439.
Matheson F. 2016. Seagrass transplants contribute to regenerating seagrass in
Whangarei Harbour, New Zealand. [accessed 3 September 2021].
https://seagrassrestorationnetwork.com/zostera-restoration-in-Aotearoa New
Zealand
Matheson FE, Reed J, Dos Santos VM, Mackay G, Cummings VJ. 2017. Seagrass
rehabilitation: successful transplants and evaluation of methods at different
spatial scales. New Zealand Journal of Marine and Freshwater Research.
51(1):96–109.
McDonald AM, Prado P, Heck Jr KL, Fourqurean JW, Frankovich TA, Dunton KH,
Cebrian J. 2016. Seagrass growth, reproductive, and morphological plasticity
across environmental gradients over a large spatial scale. Aquatic
Botany.134:87–96.
Meese RJ, Tomich PA. 1992. Dots on rocks: a comparison of percentage cover
methods. Journal of Experimental Marine Biology and Ecology. 165(1):59–73.
Morrison MA. 2021. Hauraki Gulf Marine Park habitat restoration potential. Wellington:
Ministry for Primary Industries. New Zealand Aquatic Environment and
Biodiversity Report No. 265.
Muggeo VM, Sciandra M, Tomasello A, Calvo S. 2013. Estimating growth charts via
nonparametric quantile regression: a practical framework with application in
ecology. Environmental and ecological statistics. 20:519–531.
Nordlund LM, Koch EW, Barbier EB, Creed JC. 2016. Seagrass ecosystem services
and their variability across genera and geographical regions. PLOS ONE.
11(10):e0163091.
Orth RJ, Lefcheck JS, McGlathery KS, Aoki L, Luckenbach MW, Moore KA,
Oreska MPJ, Snyder R, Wilcox DJ, Lusk B. 2020. Restoration of seagrass
habitat leads to rapid recovery of coastal ecosystem services. Science
Advances. 6(41):eabc6434.
Ramage DL, Schiel DR. 1998. Reproduction in the seagrass Zostera novazelandica
on intertidal platforms in southern New Zealand. Marine Biology. 130(3):479–
489.
Reed J, Schwarz A, Gosai A, Morrison M. 2004. Feasibility study to investigate the
replenishment / reinstatement of seagrass meadows in Whangarei Harbour –
Phase 1. Auckland: National Institute of Water & Atmospheric Research. NIWA
Client Report: AKL 2004-33. Prepared for Northland Regional Council.
Short FT, Wyllie-Echeverria S. 1996. Natural and human-induced disturbance of
seagrasses. Environmental Conservation. 23(1):17–27.

DECEMBER 2024 REPORT NO. 3994 | CAWTHRON INSTITUTE
48
Spalding M, Taylor M, Ravilious C, Short FT, Green E. 2003. The distribution and
status of seagrasses. In: Green EP, Short FT, editors. World atlas of
seagrasses. Berkley (CA): University of California Press; p. 5–26.
Stafford-Bell RE, Chariton AA, Robinson RW. 2016. Germination and early-stage
development in the seagrass, Zostera muelleri Irmisch ex Asch. in response to
multiple stressors. Aquatic Botany. 128:18–25.
Šunde C, Berthelsen A, Sinner J, Gillepie PA, Stringer K, Floer L. 2017. Impacts of
vehicle access at Delaware (Wakapuaka) Inlet. Nelson: Cawthron Institute.
Cawthron Report No. 3015. Prepared for Nelson City Council.
Tan YM. 2022. Tools and techniques for restoring Zostera muelleri in temperate
Australia [thesis]. Melbourne (VIC): Deakin University.
Tan YM, Coleman RA, Biro PA, Dalby O, Jackson EL, Govers LL, Heusinkveld JHT,
Macreadie PI, Flindt MR, Dewhurst J, et al. 2023. Developing seed‐and shoot‐
based restoration approaches for the seagrass, Zostera muelleri. Restoration
Ecology. 31(5):e13902.
Turner SJ, Schwarz A-M. 2006. Management and conservation of seagrass in New
Zealand: an introduction. Wellington: Department of Conservation. Science for
Conservation 264.
Walker DI, Lukatelich RJ, Bastyan G, McComb AJ. 1989. Effect of boat moorings on
seagrass beds near Perth, Western Australia. Aquatic Botany. 36(1):69–77.
Waycott M, Duarte CM, Carruthers TJB, Orth RJ, Dennison WC, Olyarnik S,
Calladine A, Fourqurean JW, Heck KL, Hughes AR, et al. 2009. Accelerating
loss of seagrasses across the globe threatens coastal ecosystems.
Proceedings of the National Academy of Sciences. 106(30):12377–12381.
Williams SL. 2001. Reduced genetic diversity in eelgrass transplantations affects both
population growth and individual fitness. Ecological Applications. 11(5):1472–
1488.
Zabarte-Maeztu I. 2021. Sediment-effects on seagrass Zostera muelleri in New
Zealand [PhD thesis]. Hamilton: University of Waikato.
Zabarte-Maeztu I, Matheson FE, Manley-Harris M, Hawes I. 2021. Sexual
reproduction of seagrass Zostera muelleri in Aotearoa New Zealand: are we
missing a restoration opportunity? New Zealand Journal of Marine and
Freshwater Research. 57(3):447–453.