2.1 Seed collection and storage

Seed-bearing shoots of Zostera marina were collected from a donor meadow at Hamburger Hallig (54.5996° N, 8.8184° E, Germany) during low tide in early September 2021, coinciding with peak seed maturity. A collection permit was released by Landesbetrieb für Küstenschutz, Nationalpark und Meeresschutz Schleswig-Holstein (Germany), on June 25th, 2021 (n. 3141–537.46). Seed-bearing shoots were transported within 28 h to the lab facility at Ghent University in refrigerated boxes (8 ± 1°C), where the experiment was performed.

Seed release from the reproductive shoots was facilitated by gentle seawater agitation provided through aeration with an air bubbler in buckets (30 L, 30 psu). Seed ripening was ensured by maintaining salinity and temperature conditions representative of in-situ conditions (30 psu, 10 ± 0.5°C) for a period of 42 days until natural temperature drop occurred.

Following seed release, seeds were hand-sorted, rinsed with natural seawater, and subjected to cold stratification in complete darkness at 3 ± 0.5°C for 120 days to mimic winter dormancy.

Sterilised natural seawater (autoclaved natural seawater, 121°C, 15 PSI, 20 min) with adjusted salinity (28 psu) was used as storage media (referred to as SSW). To prevent oomycete infections, 0.2 mg L⁻¹ copper sulphate (CuSO₄) was added to the storage solution, following the protocol of Govers et al. (2017). To maintain a consistent CuSO₄ concentration and ensure its biocidal efficacy, the storage solution was refreshed weekly. SSW salinity was specifically adapted to 28 psu to correspond with natural field salinity during the period of seed germination.

Field-specific temperature, salinity, and irradiance data used during storage and experiment were retrieved directly from the Federal Maritime and Hydrographic Agency of Germany (BSH, https://www.bsh.de/EN/DATA/data_node.html) and the National Oceanic and Atmospheric Administration (Mishonov et al., 2024) for the site (54.5996° N, 8.8184° E). These data were used to match seasonal averages during storage, cold stratification, and the germination period, ensuring that experimental conditions closely reflected those experienced by the seeds in their natural habitat.

Seeds were subjected to a viability test and visually inspected for maturity based on their colouration. Seeds ranging from cyan to black were selected, as also described by Xu et al. (2016). To ensure viability and, due to the lack of a rapid, non-destructive method for assessing seed viability, seed quality was evaluated manually. Seeds were first visually inspected using a microscope (Leica, MZ 16), and only seeds with an intact (i.e. no damages) and firm coat (after gentle pressure with tweezers) were shortlisted. A sinking test in natural seawater (28 psu) was adopted to further consolidate our selection. Seeds sinking quickly in natural seawater were selected and considered viable for the experiments. This non-destructive method was selected based on the experience of Marion & Orth (2010). Selected seeds were surface-sterilised with ethanol (EtOH 70% v/v) for 2 min and subsequently with a solution of 1% sodium hypochlorite (NaOCl) for 5 min. Seeds were then rinsed five times in SSW.

2.2 Experimental design, culture media, and culture condition

A fully crossed experimental design was adopted to assess the effect of light spectra, photoperiod, and hormone priming on the germination of Z. marina seeds (Figure 1). To ensure experimental feasibility and facilitate daily monitoring operation in full axenic conditions of the in vitro replicate and subsequent follow-up of each seedling, a reduced number of seeds was employed for this study.

Three light conditions were selected for incubation: white light (400–720 nm, 5000 K, 158 μmol m⁻² s⁻¹, BlueView), darkness (no light), and red light (620–750 nm, intensity 144 μmol m⁻² s⁻¹ BlueView). The white and red light were tested with two photoperiods of 12:12 h and 24:0 h (light:dark regime), in addition to a darkness incubation. Photosynthetically active radiation (PAR) was measured by means of a handheld full-spectrum quantum meter (MQ-200, Apogee).

Five hormone solutions (soaking media) were prepared by dissolving different concentrations of gibberellic acid (GA₃, Sigma-Aldrich), a major bioactive form of gibberellic acid, in SSW: 10, 50, 500, and 1000 mg L⁻¹, alongside a negative control with 0 mg L⁻¹. These concentrations were chosen based on studies indicating an effective GA₃ range for seed germination in various plant species, including marine plants (e.g. Corns, 1960; Thompson, 1969; Conacher et al., 1994; Ameen & Al-Imam, 2007; Mello et al., 2009; Martínez et al., 2016; Abdullah & Abdulrahman, 2017). Hence, a broader range of GA₃ concentrations was chosen due to the lack of prior experiments on Z. marina seeds. GA₃ solutions were mechanically filtered under laminar flow using sterile 0.2 μm syringe filters (Pall, Acrodisc). Excess solutions were stored at −20 ± 0.5°C and used for the weekly refreshment of the experimental units.

Each treatment, defined by the combination of light spectrum, photoperiod, and phytohormone, was replicated five times, resulting in a total of 15 seeds per treatment. Independent replication was achieved by applying each treatment across five separate Petri dishes, ensuring variability between units was accounted for. Within each Petri dish, three seeds were included to account for within-unit variability, resulting in a total of 375 seeds across all treatments (Figure 1).

Experimental units (Petri dishes, 35 mm, 5 mL, Avantor), each representing a unique combination of treatments as described in the previous section, were incubated at 10 ± 0.5°C for the entire duration of the experiment, a total of 42 days unless seeds germinated earlier.

The soaking media (SSW and GA₃ + SSW) were refreshed weekly to maintain stable hormone concentrations and, overcome hormone degradation and to prevent nutrient depletion, metabolic byproduct accumulation, and microbial growth. For each refreshment, in full aseptic condition under the laminar flow, the old media were carefully removed with a sterile pipette to avoid disturbing the seeds, and freshly prepared GA₃ + SSW (or SSW alone) solution was added to each Petri dish. This process ensured consistent, controlled conditions for evaluating treatment effects on germination.

2.3 Germination assessment and seedling development measurement

Germination was assessed three times a week over a period of 42 days. Germination was defined as the emergence of the cotyledon from the seed coat (Liu et al., 2023). Germinated seeds (hereafter referred to as seedlings) were transferred to a new experimental unit (Petri dish, 50 mm, 10 mL, Avantor) and supplemented with SSW (GA₃-free media). Seedlings were incubated at 10 ± 0.5°C under white light at a mean intensity of 158 μmol m⁻² s⁻¹ with a 12:12 h photoperiod in a phytohormone-free medium containing SSW. SSW was refreshed weekly to maintain the same nutrient composition. A photograph of each seedling was taken twice a week using a microscope (Leica MZ16) coupled with a Canon EOS 600D camera. Pictures were analysed using the ImageJ version 1.52d software for morphometric measurement of the seedlings (Schneider et al., 2012). A certified micrometre graticule (Pyser-SGI, serial n. SC1539) was used to calibrate the software for pixel-μm conversion. Morphometric measurements were recorded for each seedling, cotyledon, number of roots, and true leaves, with each of their lengths measured two times a week. True leaves were defined as the leaves capable of performing photosynthesis, as described by Sugiura et al. (2009) and Taylor (2011). The length of the cotyledon was defined as the distance from the basal hypocotyl to the tip of the cotyledonary blade, and the length of true leaves was defined as the distance from the base of the cotyledonary sheath to the tip of the leaf. Maximum growth was calculated as the sum of the linear measurements of cotyledon and leaf length.

2.4 Statistical analysis

All analyses were performed using R statistical software (v.4.0.2; R Core Team, 2020). Seed germination results are presented as sum and binomial confidence intervals. Mean germination time, cotyledon, and leaf length results are presented as arithmetic mean ± standard error, unless stated otherwise.

The effects of the light spectrum, photoperiod, and hormone concentration on germination were analysed using a penalised logistic regression model implemented through the R package logistf (Firth, 1993; Heinze & Puhr, 2010; Puhr et al., 2017). Firth's penalisation approach was selected to account for low event rates and mitigate bias in coefficient estimates. Backward stepwise selection between models was carried out using the Akaike Information Criterion (AIC). ANOVAs with post-hoc Tukey HSD tests were performed via the multcomp R package (Herberich et al., 2010) to compare nested models and determine the exclusion of variables. The outcome variable, germination probability (GB), represents the likelihood of seed germination under varying conditions of light, photoperiod, and phytohormone concentration. Significance was consistently interpreted at p < 0.05.

The effects of light spectrum, photoperiod, and hormone concentration on the time to germination were analysed using a time-to-event (survival) approach with the R packages survival and survminer (Therneau & Grambsch, 2000). Survival analysis considers both the occurrence of an event, such as germination (binary response), and the time it takes for the event to occur (continuous response). As a survival model, we selected a Cox proportional hazards method to account for both categorical and continuous predictors (McNair et al., 2012), with predictors being light spectrum, photoperiod, and hormone and the response variable being time to germination (days). To ensure an accurate comparison between the statistical analyses (penalised logistic regression and Cox proportional hazard model), the selection of variables was based on the outcomes of the logistic model's ANOVA. To visualise the effect of predictors on the response variable, survival curves were computed via the Kaplan–Meier method through the R package survival.

Linear models were used to assess the potential backdrop effects of light spectrum, photoperiod, and hormone concentration on cotyledon and leaf growth after 60 days. The normality of residuals was evaluated using the Shapiro–Wilk test (stats R package), while the homogeneity of variances was tested using Levene's test (car package). To meet the assumptions of the analysis, leaf length measurements were square root-transformed prior to model fitting. Models were fitted using the lm function (Zuur et al., 2009) and estimated marginal means (EMMs) were calculated using the emmeans R package. Pairwise comparisons of EMMs were conducted with Tukey-adjusted p-values to account for multiple testing. All statistical analyses were performed in R, and results are presented as means ± standard error unless otherwise stated.

3.1 Influence of light and hormone treatment on seed germination

Overall, total seed germination reached 7.57% (95% CI 4.1–13.6), with an average daily germination rate of 0.80 ± 0.40% seeds. Seeds incubated in darkness showed the highest germination (10.67%, 95% CI 5.5–19.7%) compared to those in red light (8.33%, 95% CI 5.1–13.3%), and white light (3.70%, 95% CI 1.7–7.8%), although these differences were not statistically supported (p = 0.1). Control groups (0 mg L⁻¹ GA₃) had consistently low germination, peaking at 11.12% (95% CI 3.1–32.8%) under red light with a 12:12 h photoperiod.

Following variable selection, the final logistic model included the factors of light spectra, hormone, and the interactions of light and hormone and hormone and day, while photoperiod was not included. Germination probability (GB) was significantly enhanced by GA₃ treatment alone (χ²(4) 42.55, p = 1.28 e⁻⁰⁸) and by the combination of light and GA₃ (χ²(8) 114.66, p < 0.001). Darkness positively influenced seed germination, though a negative dose-effect trend emerged as GA₃ concentrations increased (Figure 2b), except in the control group (0 mg L⁻¹ GA₃). Under this light condition, the highest mean germination was observed in seeds exposed to 10 mg L⁻¹ GA₃, 26.7%, 95% CI 10.8–51.9%, and this result was supported statistically (z = 2.82, p = 2.21 e⁻⁰⁴). Seeds incubated in darkness and in the absence of GA₃ (control) showed no germination, consistent with natural dormancy behaviours in many plant species. Although germination generally declined with higher GA₃ levels, a notable exception was observed with seeds exposed to 50 mg L⁻¹ GA₃ in darkness.

Similarly, there was no germination for seeds exposed to red light at 500 mg L⁻¹ GA₃ and under white light at both 10 or 500 mg L⁻¹. These findings suggest that specific light-GA₃ combinations can inhibit germination, possibly through synergistic interactions that counteract GA₃'s germination-promoting effects.

GA₃ treatment alone increased germination probability, with significant effects observed at 10 mg L⁻¹ GA₃ (z = 3.33, p = 1.29e⁻⁰⁶), 50 mg L⁻¹ GA₃ (z = 1.70, p = 0.063), and 1000 mg L⁻¹ GA₃ (z = 2.15, p = 0.011). 500 mg L⁻¹ GA₃ resulted in a modest increase in germination probability; however, the effect was not statistically significant (z = 0.95, p = 0.376). These results generally indicate a dose-dependent response, with higher GA₃ concentrations associated with enhanced germination. These findings suggest that optimised GA₃ levels could be beneficial for seed germination, offering potential applications in restoration efforts.

3.2 Temporal patterns of seed germination: a time-to-event approach

To better understand the temporal dynamics of seed germination, we analysed how different GA₃ concentrations and light conditions influenced the rate and likelihood of germination over time. This approach allowed us to assess not only which treatments enhance germination rates but also how quickly seeds respond under varying light and hormone conditions.

For seeds exposed to white light, the likelihood of germination increased by 2.72 ± 0.61 times at 1000 mg L⁻¹ GA₃, 1.96 ± 0.64 times at 50 mg L⁻¹ GA₃, and 1.74 ± 0.63 times at 10 mg L⁻¹ GA₃. Seeds exposed to white light and 50 mg L⁻¹ GA₃ exhibited the earliest germination response, with an average time to germination of 6.67 ± 1.26 days (Figure 3a). Notably, only the combination of the red light and 500 mg L⁻¹ GA₃ significantly reduced the time to germination (p = 0.033, Figure 3c). According to the Cox proportional hazard model, seeds exposed to red light germinated 1.83 ± 0.48 times faster than those exposed to white light; however, this difference was not statistically significant (p = 0.09).

3.3 Seedling development

The a posteriori effect of seed treatments on seedling development was monitored to assess potential side effects on seedling growth. After 60 days, cotyledons reached a mean maximum length of 18.96 ± 5.06 mm and grew at a mean rate of 0.75 ± 0.21 mm day⁻¹, while the mean maximum length was 16.78 ± 9.21 mm with a growth rate of 0.84 ± 0.39 mm day⁻¹. The first true leaf appeared on average 10.9 ± 2.80 days after germination.

The post-germination development of Zostera marina seedlings was assessed to evaluate the potential effects of light, photoperiod, and hormone treatments on cotyledon and leaf length responses. After a 60-day seedling follow-up, the maximum cotyledon length (28.23 mm) was observed in seeds exposed to red light and 1000 mg L⁻¹ GA₃ under a 24-hour photoperiod (Figure 4c, b, o) with the second-highest growth recorded in darkness with 10 mg L⁻¹ GA₃ (21.65 mm). Leaf development followed a similar trend, with the longest leaf length recorded under red light, a 24-hour photoperiod, and 10 mg L⁻¹ GA₃ (34.06 mm).

However, the linear model analysis indicated that the interaction between light, photoperiod, and hormone had non-significant effects on both cotyledon (F(12, 10) = 1.12, p = 0.43) and leaf lengths (F(8, 6) = 0.99, p = 0.51). Nonetheless, certain individual treatments exhibited notable growth. For instance, under darkness (24-hour photoperiod) and 500 mg L⁻¹ GA₃, substantial cotyledon (20.79 mm) and leaf (21.31 mm) development was observed, suggesting a potential influence of darkness-incubated treatments (with 500 mg L⁻¹ GA₃) on early seedling growth.

Conversely, the shortest cotyledon lengths were observed in some high-concentration treatments, such as darkness with 1000 mg L⁻¹ GA₃, where cotyledon length was as low as 2.50 mm. Our results highlight variability in post-germination responses. While certain light and hormone combinations were associated with significant growth, the overall interactions were not statistically significant, suggesting that light and hormone treatments may have a limited or negligible impact on early seedling development.

4.1 Effects of light on germination

The germination response observed in Z. marina seeds upon light exposure underscores an active involvement of plant photoreceptors in germination physiology. This seagrass species harbours six characterised photoreceptors, including phytochromes and phototropins. These photoreceptors likely drive light-dependent signalling pathways that activate enzymatic processes, facilitating seed wall weakening and endosperm degradation, which are crucial steps in promoting germination (Olsen et al., 2016).

Our findings demonstrate consistent germination of Z. marina seeds in the absence of GA₃ when exposed to white or red light but not in darkness, confirming the photoblastic nature of Z. marina seeds.

Seeds exposed to white light, encompassing both red and blue wavelengths, exhibited reduced germination compared to those exposed to red light alone. This response aligns with adaptations of Z. marina seeds in intertidal meadows, where seeds are often buried under thin sediment layers yet exposed to filtered daylight, with shifted red-to-far-red ratios (Probert & Brenchley, 1999; Strydom et al., 2017). Such conditions, combined with attenuated light intensity, likely mimic the germination-inhibiting effects of white light observed in our study.

During low tide, seeds are exposed to light enriched with blue and green wavelengths (Morel & Maritorena, 2001), potentially mimicking the germination-inhibiting effects of white light observed in this and other studies (Strydom et al., 2017; Waite et al., 2021). Similar photoinhibition mechanisms have been observed in terrestrial angiosperms, where it functions as an adaptive strategy to delay germination under unfavourable environmental conditions, such as high light intensity combined with water scarcity, which are commonly encountered in desert ecosystems or specific microhabitats in coastal regions (e.g. coastal dunes) (Mérai et al., 2023). A comparable strategy may exist in dynamic marine environments, where light, temperature, salinity, and irradiance fluctuate rapidly.

While our data did not indicate a significant effect of photoperiod under white light, the inhibition observed under a 24 h-photoperiod suggests the involvement of complex regulatory mechanisms that optimise germination timing and embryo development. The retention of key photoreceptor genes, such as CRY1, PHOT1, PHOT2, PHYA, and PHYB, in Z. marina, alongside the evolutionary loss of others like CRY2, CRY5, CRYDASH, and PHYC, highlights specialised adaptation for light perception in marine environments (Mawphlang and Kharshiing, 2017; Nautiyal et al., 2023).

Red light, known to promote germination in many terrestrial and aquatic species (Mathews, 2006; Ziegler et al., 2023), acts through the conversion of the inactive Pr form of phytochrome to the active Pfr state, triggering downstream germination-promoting pathways (Furuya and Schäfer, 1996). Notably, our study observed enhanced germination under red light, contrasting findings from Lidao Bay (China), where red light had no impact on Z. marina germination (Wang et al., 2017). This discrepancy may reflect population-specific adaptations or ecological responses to seasonal turbidity shifts, which increase red wavelengths in water (Strydom et al., 2017, Statton et al., 2017b). The positive effect of red light on germination has been similarly documented in other seagrasses, including Halophila ovalis (Waite et al., 2021), Thalassia hemprichii (Ehrenb.) Aschers. (Soong et al., 2013), and Zostera japonica Asch. & Graebn. (Kaldy et al., 2014), suggesting a shared evolutionary strategy among seagrass species.

The absence of germination in darkness (in the absence of GA₃), consistent with patterns observed in many terrestrial plant species (Penfield and King, 2009), suggests a shared light-dependent germination mechanism that may also operate in seagrasses. In terrestrial plants, this dependency is often linked to phytochrome-mediated activation, which prevents premature germination when seeds are buried too deeply in the soil. Although Z. marina germinated to varying degrees under all light treatments, the complete lack of germination in darkness highlights the essential role of light in breaking dormancy and triggering germination. The underlying mechanisms for this light dependency in seagrasses remain an intriguing subject for further research, particularly given their unique aquatic environment and evolutionary context.

4.2 Effects of GA₃ treatment on germination

Treating seeds with GA₃ is known to initiate a pre-germinative metabolic state that enhances germination and seed vigour in many terrestrial plants of both agricultural (Kaur et al., 2006; Azab, 2018; Budzeń et al., 2018; Lee et al., 2018) and conservation interest (Yang et al., 2005; Barden et al., 2017). Furthermore, GA₃ is widely employed in horticulture to stimulate germination and improve, for instance, the size and quality of fruits (Rademacher, 2015). Our findings provide the first insights into the use of GA₃ as an exogenous treatment to stimulate germination in Zostera marina, addressing a notable gap in current research.

We observed that seeds exposed to GA₃ showed increased germination compared to non-GA₃-treated controls, suggesting that GA₃ plays a role in overcoming dormancy. These results align with findings in terrestrial plants, where GA₃ has been widely employed to stimulate germination (Rademacher, 2015). However, prior studies on other seagrass species have yielded mixed results; for instance, Glasby et al. (2015) reported no effect of GA₃ (on concentrations ranging from 0.1 to 20 mg L⁻¹) on germination in Posidonia australis Hook.f., while Loques et al. (1990) found that 1 mg L⁻¹ GA₇ did not enhance germination in Zostera noltii Hornem. These inconsistencies may stem from differences in media composition or experimental conditions that could suppress germination responses (Hootsmans et al., 1987). In contrast, our study found that the earliest germination occurred at 10 mg L⁻¹ GA₃ under white light, underscoring the critical interplay between GA₃ concentration and light conditions in driving germination.

The interaction between light and GA₃ appears to be nuanced and complex. Notably, Conacher et al. (1994) reported that 50 mg L⁻¹ GA₃ promoted germination in Zostera capricorni Asch. while higher concentrations (500 mg L⁻¹) did not. Similarly, our observation in Z. marina reveals a comparable pattern, particularly under white and red light conditions. The absence of germination at 0 mg L⁻¹ GA₃ under complete darkness, contrasted with successful germination at higher GA₃ concentrations under identical light conditions, reflects findings in terrestrial plants where exogenous GA₃ treatments mimic red light-induced signalling cascades to promote germination (Toyomasu et al., 1998). In Lactuca sativa L., for instance, germination is regulated by phytochrome, and the application of gibberellin (GA) can bypass the red-light requirement by inducing GA biosynthesis pathways (Toyomasu et al., 1998).

The varied responses to GA₃ treatments in Z. marina, Z. capricorni, and Z. noltii suggest potential species-specific mechanisms, likely influenced by differences in the genetic and molecular regulation of germination (Xu et al., 2021; Zhang et al., 2022). Adaptations to the marine environment, including changes in hormone signalling and biosynthesis, have been identified at the genomic level (Chen et al., 2021). For instance, modifications in GA signalling pathways in Zostera muelleri Irmisch ex Asch. (Lee et al., 2016) and the loss of ethylene biosynthesis genes in Z. marina (Golicz et al., 2015) reflect evolutionary changes that could impact germination responses. Nevertheless, the specific roles of individual genes in GA pathways in seagrasses remain largely unexplored (Chen & Qiu, 2022).

While our study underscores the potential of light and GA₃ treatments to modulate germination in Z. marina, it also uncovers complexities, such as the non-linear dose response at specific GA₃ levels. The inhibitory effects observed at certain concentrations, such as 500 mg L⁻¹, challenge conventional dose–response assumptions and suggest a non-monotonic relationship. Possible explanations include complex hormonal interactions (Vanstraelen and Benková, 2012), physiological thresholds (Bradford & Trewavas, 1994), seed sensitivity variations (Jalal et al., 2021), and potential antagonistic interactions between light and hormone treatments (Seo et al., 2009).

These findings highlight the need for further investigation into the optimal concentrations and conditions for GA₃ to support seed germination and the release of dormancy. Future research should incorporate intermediate GA₃ concentrations, gene expression analyses, and hormonal profiling to better understand the underlying mechanisms. Such an effort would enhance our understanding of hormonal control in seagrasses and support the development of targeted seed treatment strategies for ecological restoration and conservation initiatives.

4.3 Impact of light and GA₃ treatments on early-stage seedling development

Although this study did not evaluate the effects of the light and hormone treatments on Zostera marina seedling development, the potential carryover effects of seed treatments should not be overlooked. In terrestrial plants, GA₃ and light treatments are widely employed to promote and support seedling development (Rademacher, 2015; Waqas et al., 2019). For example, GA₃ has been shown to enhance primary and secondary growth in adult plants such as sugar beet and korarima (Eyob, 2009; Azab, 2018). A similar approach has been explored in some marine plant species, with GA₃ concentrations ranging from 0.01 to 35 mg L⁻¹ promoting leaf and rhizome growth in Cymodocea nodosa (Ucria) Asch (Zarranz et al., 2010). However, no such effects were reported for Posidonia australis (Glasby et al., 2015). In contrast, Zostera noltii seedlings exposed to 1 mg L⁻¹ GA₇ exhibited reduced growth (Loques et al., 1990), and in Halophila ovalis, short-wavelength were found to significantly reduce above- and below-ground biomass (Strydom et al., 2017).

Overall, our study found that none of the seed treatments negatively impacted the growth, morphology, or survival of cotyledons and leaves in Z. marina seedlings. Subsequent seedling growth rates were comparable to previously reported values from studies without treatments (Taylor, 1957; van Lent & Verschnure, 1995; Niu et al., 2012). While these results suggest that seed treatments do not adversely affect Z. marina seedling development, it is important to note that our findings may provide only a partial understanding of the broader effects of seed treatments, particularly given that some GA₃ concentrations did not yield germinative outcomes.

To further elucidate the impact of seed treatments on seedling development, we recommend that future studies incorporate morphometric analyses, assessments of photosynthetic performance, leaf pigment composition, and the expression of genes encoding key enzymes involved in developmental and metabolic pathways. This comprehensive approach would offer deeper insights into the physiological and developmental effects of seed treatments on Zostera marina.

4.4 Implications for restoration

Seagrass conservation and restoration have been identified as global priorities by the United Nations (Environment, U.N., 2020) and the European Union (Council of the European Union & European Parliament, 2024). In response to the growing impacts of climate change on marine ecosystems, seed-based restoration of Zostera marina habitats is increasingly employed to rehabilitate this widely distributed yet declining seagrass species. Efforts within the seagrass research and restoration community are intensifying to advance scientific methods and logistic standards, aiming to scale up restoration initiatives. Enhancing germination success and maximising the utilisation of seed stocks are critical for bolstering restoration outcomes and strengthening ecosystem resilience (Nordlund et al., 2024).

To ensure the affordability and scalability of restoration projects, it is important to develop cost-effective methods that enhance the efficiency of restoration efforts. Over the past decades, researchers have described and improved seed-based restoration techniques (Unsworth et al., 2019a; Tan et al., 2020; Gräfnings et al., 2023), methods to support germination (Waite et al., 2021; Xu et al., 2021), and seedling establishment (Verduin et al., 2013). However, existing techniques, such as seed scarification and freshwater pulse, have yielded inconsistent and unpredictable results. Addressing current limitations in seed availability by improving germination rates and optimising nursery practices is essential to meeting the increasing demand for large-scale seagrass restoration.

Our study demonstrates that alleviating potential stress-induced dormancy in Z. marina seeds through targeted manipulation, such as treatments with light and phytohormones, offers a promising strategy to enhance germination potential prior to sowing. While our findings provide valuable insights into the potential of light and hormone seed pre-treatments to promote germination and overcome dormancy, further studies are needed to evaluate the consistency and predictability of these effects across a broader range of conditions, including different hormone concentrations. This research provides an important step toward evaluating the feasibility of these techniques for applications in nurseries and restoration projects. Future investigations should focus on determining the optimal concentrations and incubation durations needed to induce early germination processes (pre-germinative stages) without triggering sprouting before sowing. This approach has significant potential to enhance germination rates, reduce germination timelines, and achieve synchronised germination, thereby supporting large-scale restoration initiatives.

Moreover, as this study was conducted in a controlled laboratory setting, we acknowledge potential limitations that may arise from overlooking environmental variables or interactions present under natural conditions. These factors could influence the practical applicability of our findings in real-world scenarios. We, therefore emphasise the importance of translating this laboratory-based knowledge to field settings to gain a more comprehensive understanding of the efficacy of seed pre-treatment methods under natural environmental conditions. Such efforts are crucial for advancing the success of seagrass restoration projects and ensuring the practical utility of these techniques in large-scale applications.