Ongoing climate change is affecting marine ecosystems through warming ocean temperatures, rising sea levels, ocean acidification, increasing ocean stratification and decreasing sea ice cover (IPCC, 2022). These stressors are altering the structure and function of ecosystems, and are reducing global and local biodiversity (Vörösmarty et al., 2010; Halpern et al., 2019). Hence, interventions aimed to halt biodiversity loss and restore degraded ecosystems are essential (Duarte et al., 2020).

It is difficult for some degraded ecosystems to naturally return to historic states within a reasonable time scale (Lotze et al., 2011). Even after the causes of degradation are removed, recovery can still take decades or even centuries (Dobson et al., 1997; Lotze et al., 2011) and sometimes does not occur at all (Cox et al., 2017). Therefore, conservation that only focuses on removing, lessening, or ameliorating environmental stressors may not be effective in stopping or reversing trajectories of change (McCrackin et al., 2017; Lindegren et al., 2018). Combining traditional conservation with active restoration (i.e. anthropogenic interventions to accelerate the recovery of degraded ecosystems) is necessary to address the rapid changes in marine ecosystems (Basconi et al., 2020). Recognising this, the United Nations has declared 2021-2030 as the ‘UN Decade on Ecosystem Restoration’ and the ‘UN Decade of Ocean Science for Sustainable Development’ (Waltham et al., 2020). With both these UN calls to action, marine restoration ecology may thus become the leading field in ecological science over the next decade (Basconi et al., 2020).

Kelp forests are created by large habitat-forming brown macroalgae predominantly from the order Laminariales and extend over approximately 25% of the world’s temperate and polar coastline (Wernberg et al., 2019b). Kelps are foundation species that support local biodiversity (Norderhaug et al., 2005; Teagle et al., 2017) and provide critical ecosystem services to human societies (Smale et al., 2013; Bennett et al., 2016; Blamey and Bolton, 2018) that are worth billions of USD year^-1 (Eger et al., 2021). Long-term monitoring programs (> 20 years) show that 40% to 60% of the global kelp forests are declining (Krumhansl et al., 2016; Wernberg et al., 2019b). Many of these declines are due to anthropogenic climate change and ocean warming, either directly or indirectly (Smale, 2020, Wernberg et al., 2019b). Warming can shift kelp forests from diverse, three-dimensional habitats to structurally simple turf-dominated reefs (Filbee-Dexter and Wernberg, 2018; Pessarrodona et al., 2021). Many turf reefs have not recovered (Coleman et al., 2020; Filbee-Dexter et al., 2020; Fredriksen et al., 2020) demonstrating discontinuous phase shifts with hysteresis (Scheffer et al., 2001), and future projections suggest further losses (Assis et al., 2018; Martínez et al., 2018; Davis et al., 2022). Hence, active strategies such as restoration and rehabilitation are needed to regrow kelp forests in these areas (Coleman et al., 2020; Eger et al., 2022b).

Restoration of kelp forests provide an unparalleled means to enhance biodiversity and support healthy coastal environments (Feehan et al., 2021; Filbee-Dexter et al., 2022). Kelp forest restoration efforts have shown variable success rates, with projects often either focusing on transplanting healthy individuals to a degraded area, adding new reefs or structures, and/or the removal of the cause of decline (e.g. sea urchins; Eger et al., 2020; Layton et al., 2020). While some of these efforts resulted in short-term improvements, long-term success has been limited (Layton et al., 2020). Furthermore, many kelp restoration efforts have been small scale relative to the scales of loss (less than 100 m², Eger et al., 2022b). Moreover, for restoration to be successful, it is imperative to remove pressures on kelp forests from climate change, herbivory, sedimentation and pollution (Duarte et al., 2020).

Some kelp restoration efforts have focused on harvesting donor plants for transplanting into degraded reefs. These plants must survive environmental stressors, such as wave and storm exposure, until reproduction occurs (Campbell et al., 2014; Vergés et al., 2020). Other techniques involve seeding reefs with early life-stages of kelps using spore bags or adding kelp propagules directly. However, this comes with its own set of challenges. Although kelp propagules (i.e. zoospores) can easily be released in the laboratory (e.g. Alsuwaiyan et al., 2019), they are challenging to grow and upkeep due to their high sensitivity to laboratory conditions. Kelp zoospores are also difficult to plant or scatter onto deteriorated reefs (Vanderklift et al., 2020) especially in areas where turf algae have established (e.g. Moy and Christie, 2012; Filbee-Dexter et al., 2016; Feehan et al., 2019). “Green gravel” (Fredriksen et al., 2020) is a new restoration technique where kelp propagules are seeded onto small gravel in the laboratory ( Figure 1 ), then out-planted back onto degraded reefs. This approach proved to be a promising method to combat kelp forest declines (Fredriksen et al., 2020). Green gravel has the potential for up-scaling to treat large degraded areas due to its low cost and low maintenance (Fredriksen et al., 2020), however, this method may be sensitive to high wave exposure or could introduce foreign substances into the ocean. Indeed, important questions around the green gravel approach remain unanswered such as what methods are optimal for gametophyte survival and whether other kelp species can be seeded in this way.

Here, we tested the green gravel method on Australia’s dominant underwater forest forming species, Ecklonia radiata, which has undergone declines on Australia’s east and west coasts (Wernberg et al., 2019a). Notably, E. radiata experienced one of the most severe marine heatwaves in recorded history during the austral summer of 2011. This heatwave caused mortality and range contractions of E. radiata across several hundred kilometers of coastline (Wernberg et al., 2016), with some populations entirely collapsing (i.e. the low latitude range edge populations; Wernberg et al., 2018). Species distribution models under future climate change scenarios further predict that E. radiata could lose 49% to 71% of its current distribution range under RCP 2.6 and 6.0 emissions scenarios (Martínez et al., 2018). Along Western Australia’s coastline, E. radiata populations show patterns of genetic structuring and adaptation to local thermal environments (Vranken et al., 2021), thus making E. radiata a logical target species for assisted adaptation and restoration in areas where it has disappeared.

In this study, we focused on the first step of the green gravel approach and tested the feasibility of cultivating E. radiata gametophytes on different gravel substrata. In addition, we tested the potential for gametophytes that detach from the gravel to settle and develop ex situ on the surrounding substratum. This is important in understanding if green gravel can, in some instances, help overcome the problem of recruitment limitation in degraded reefs by providing a way to add reproductive material directly to the reef surface. This study provides new insights into optimising green gravel for restoration.

Gametophyte fragments from stock cultures were able to adhere to all four tested substrata successfully, however, gametophytes that adhered to the limestone gravel (the natural reef type off Western Australia) suffered extreme tissue bleaching and died. Gametophytes that adhered to the small and large basalt and laterite gravel were healthy, fertilised following seeding and microscopic sporophytes were visible within 14 days. Seawater pH in containers with limestone gravel ranged between 3.88 and 4.07, whereas in containers with the other tested substrata, pH ranged between 7.18 and 7.70.

A key challenge in the current Decade of Ecosystem Restoration is scaling up restoration tools to meet the rapid rates of loss (Saunders et al., 2020; Filbee-Dexter et al., 2022). Considering that kelp forests have fast growth rates and have declined over large areas of their global range, there is great opportunity to increase efforts to protect and restore kelp forests and secure the many fundamental ecosystem services that they provide to human society (Feehan et al., 2021). Green gravel is a simple, cost-effective and scalable restoration technique that involves two main processes: culturing kelp propagules in the laboratory and out-planting them back into the field (Fredriksen et al., 2020). Here, we tested the feasibility of cultivating Ecklonia radiata gametophytes on gravel and the ability of gametophytes that detach from gravel to develop ex situ. Our results showed that E. radiata gametophytes from stock cultures were able to adhere and develop successfully on gravel, demonstrating green gravel to be a promising option to restore E. radiata populations. Moreover, we also showed that gametophytes detached from gravel can develop on the surrounding substratum, making green gravel a versatile tool to seed degraded reefs either directly through juvenile sporophytes attaching onto the gravel or indirectly by providing a way to disseminate reproductive material onto the underlying reef to enrich restored or degraded sites.

In this study, we focused on optimising the first step of the green gravel process: culturing E. radiata gametophytes on small gravel. This is challenging because early life-stages of kelp are extremely sensitive to contamination (Redmond et al., 2014) and most culture studies on E. radiata do not continue beyond the microscopic sporophyte stage. Substrata type (i.e. gravel) had some effects on the survival of E. radiata gametophytes. The extreme tissue bleaching in gametophytes that adhered to the limestone gravel highlights the importance of testing the choice of gravel prior to starting experiments. Some gravel can potentially modify seawater chemistry (Brockmann and Janse, 2008) and adversely affect the survival of the very small early stages of kelp, thus resulting in ineffective interventions and wasting valuable resources. Gametophytes that adhered to the three other tested gravel types, however, developed into healthy juvenile sporophytes, and surprisingly, seeding density had no effect on sporophyte length. This result was unexpected because intraspecific competition for nutrient and space increases with density, and therefore growth in high density cultures is reduced due to competition, compared to low density cultures (Reed et al., 1991; Steen, 2003). However, it is possible that this competition could have induced density-dependent mortality (Reed, 1990), thus partially explaining the low number of sporophytes that were eventually growing on the gravel after 12 months. Further, both seeding densities used in this study are significantly lower than densities previously used for kelp aquaculture (e.g. Flavin et al., 2013; Redmond et al., 2014), yet seeding with considerably different densities yielded similar results with no differences in growth in Saccharina latissima (Fredriksen et al., 2020). This suggests that seeding density may not limit the successful development of sporophytes on the gravel. This is particularly important when green gravel is being seeded with gametophytes from stock cultures, where material could be limited.

Besides the successful adhesion and development of E. radiata gametophytes on the gravel itself, gametophytes that detached from the gravel were able to settle onto the surrounding substratum, then fertilise and develop into juvenile sporophytes. Given that eventually only few sporophytes will grow into large kelp on each gravel regardless of initial seeding density, as seen in our results and Fredriksen et al. (2020), the ability of microscopic stages to settle onto surrounding substrata after detaching from the gravel can help to increase the number of juveniles that may directly attach to the reef and reach the adult stage at the restoration site. Substrata type had a significant effect on the density of detached gametophytes, with larger and rougher textured gravel showing higher densities of detachment. This result was expected given that propagule settlement was found to be higher on rougher substratum than on smooth substratum (Falace et al., 2018), and thus more gametophytes will detach as they grow larger and competition for space increases. Notably, timing in which green gravel was moved did affect the density of detached gametophytes, with higher densities observed 30 days post-seeding compared to 100 days. This could be because most gametophytes had undergone gametogenesis during the longer incubation time in the laboratory, which reflected on the low density of detached gametophytes 100 days post-seeding. These results also suggest that with time, gametophytes may become solidly attached to the gravel and might not be capable of detaching, which highlights the importance of timing when out-planting the gravel into the field. It is suggested that the out-planting process should be carried out within 1-2 months of cultivating in the laboratory, as kelp cultured for long periods in laboratory settings may perform poorly in the field (Eger et al., 2022a). However, this might not be possible with realistic time frames of cultivation. Whilst the rates of gametophyte detachment in our experiments became significantly low 100 days post-seeding, questioning the effectiveness of green gravel as a vector of dispersal, reproductive success (i.e. sporophyte density on the surrounding substratum) was similar between the two post-seeding times. This suggests that sporophytes that developed on the gravel may also detach and adhere successfully onto surrounding substrata even after a few months of seeding on the gravel. This was also supported by observations in our aquarium experiments, where sporophytes were observed growing on other structures in the tank many months after transferring from the incubators (e.g. surrounding trays, wave makers). This was only observed in aquarium tanks with green gravel, which shows that it was not an artefact from inflowing seawater. The presence of sporophytes away from the gravel could also be due to the mobility of kelp gametes (Lüning, 1990). Whilst kelp eggs are immobile, they have the potential to break away from the fertile gametophyte after gametogenesis. This may result in an instance where a detached egg is fertilised by a mobile sperm and thus develop into a sporophyte on surrounding substrata. Indeed, kelp propagules will need suitable substrate to settle and develop on after detaching from the gravel, which is often not available in degraded areas (e.g. turf algae). In some instances, substrates can be added to overcome this limitation. In addition, the abundance of turf algae regularly displays seasonal patterns (Diaz-Pulido and Garzón-Ferreira, 2002), thus carrying out the out-planting process during periods when abundance is low may help increase the successful adhesion of kelp propagules onto surrounding substrata (Eger et al., 2022a). Factors that could challenge the successful settlement of kelp early-life stages in the field require more investigation including sedimentation, grazing activity and wave action.

In this study, full scale field trials of green gravel were not possible due to delays in receiving restoration permits, highlighting that the permitting and regulatory process could be a significant barrier to developing these techniques efficiently (Shumway et al., 2021). Although regulations, policies and legislation guide restoration programs, such governance can be a major hurdle to the success and sustainability of restoration programs (Sapkota et al., 2018). Indeed, current restoration governance needs improvement and strengthening in order to successfully implement restoration programs (Richardson and Lefroy, 2016).

Contamination is a major challenge in cultivating kelp and other seaweeds in an aquarium setting (Redmond et al., 2014). Microscopic epiphytic algae were found to likely be the primary cause of kelp degeneration and mortality in aquarium tanks (Tew et al., 2011). Aquarium settings reported in this study were developed after many challenges we faced in early trials, including diatom and red algal contaminations. Indeed, seawater carries many microorganisms that can graze on and compete with kelp in aquariums (Redmond et al., 2014) and using sand-filtered and UV-treated seawater was sufficient to remove biological contaminants. Reducing light levels and adding germanium dioxide (GeO₂) also helped in controlling contamination levels. Yet, poor quality seawater and contamination could be a limitation for green gravel success in some aquaria and for some kelp species. In this study we established some fundamental knowledge for optimum cultivation of E. radiata not only for green gravel application context, but also for the seaweed farming industry (Buschmann et al., 2017; Filbee-Dexter et al., 2022). It should also be noted that while we were successful in culturing E. radiata in aquarium tanks for over 12 months, growth rates of these kelp are low, and their size was small compared to wild juvenile kelp (Wernberg et al., 2019a). This underdevelopment is likely to be a consequence of being cultured in aquaria, thus further efforts to optimize such settings are required.

Collectively, our results show the potential in applying green gravel to restore lost or declining E. radiata populations. Indeed, determining the ability of E. radiata gametophytes to adhere and develop successfully on green gravel, as well as on surrounding substrata through lateral movement, is a critical first step in optimising this tool for restoration programs in Australia. Further research is required to improve and finesse green gravel for restoration and conservation, including field tests of this tool. Studies that take into consideration potential stressors and competitors that can inhibit kelp recruitment (e.g. sediments and turf algae; Watanabe et al., 2016; Layton et al., 2019) are necessary to optimise the out-planting process and increase restoration success.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

NAA, KF-D, MAC and TW conceived and designed the study. SV carried out culture preparation. NAA carried out the experiment. CB carried out aquarium monitoring. NAA analysed the data, produced all tables and figures, and took the lead in writing the manuscript. KF-D helped with data analysis. All authors contributed to the article and approved the submitted version

This work was funded through the Australian Research Council grants DP160100114 (TW, MAC), DP190100058 (TW, MAC, KF-D), LP1931001500 (TW, MAC, KF-D) and DE1901006192 (KFD). NAA was supported by a postgraduate scholarship from Qassim University, Kingdom of Saudi Arabia. CB was supported by a Forrest Research Foundation scholarship and Australian Government Research Training Program scholarship.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.