Skip to main content
NCBI home page
Evolutionary Applications logoLink to Evolutionary Applications
. 2026 May 20;19(5):e70257. doi: 10.1111/eva.70257

Minimising Risks of Reduced Genetic Diversity in Marine Restoration

G Wood 1,2,✉, K Filbee‐Dexter 1,3, T Wernberg 1,3, M A Coleman 1,4
PMCID: PMC13239296  PMID: 42256506

ABSTRACT

Marine habitat restoration is expanding globally and increasingly relies on aquaculture and hatchery‐based propagation, yet the genetic consequences of producing and outplanting large numbers of habitat‐forming organisms remain poorly evaluated. A lack of best‐practice guidelines and insufficient genetic monitoring creates risks including bottlenecks, inbreeding and domestication, with long‐term impacts possible for both restored and remnant populations. Here, we assessed the impacts of hatchery processes on genetic diversity and domestication in the kelp, Ecklonia radiata, by genotyping wild donor populations and their hatchery‐bred F 1 cohorts. Despite using ~100 donor adults per site, hatchery‐bred F 1 individuals showed clear reductions in genetic diversity. On average, neutral heterozygosity declined by ~20%–30%, and more than half of the percentage of polymorphic loci was lost, indicating strong genetic drift and erosion of potentially locally adapted variation. Moreover, a small number of outlier loci were strongly associated with hatchery conditions, indicating early selection and domestication responses within a single generation. F 1 cohorts also exhibited elevated inbreeding and sharp reductions in effective population size (N e falling from ~500 in the wild to as low as 21 in the hatchery). These findings demonstrate that unmonitored propagation can rapidly alter the genetic composition of foundation species under hatchery conditions, potentially undermining adaptive capacity and field performance. Incorporating genetic guidelines and monitoring into restoration programs will be essential to support long‐term persistence and evolutionary resilience.

Keywords: adaptation, aquaculture, captive breeding, domestication, genetic management, genetics, marine restoration, risk

1. Introduction

Marine ecosystems are experiencing unprecedented degradation due to human impacts and accelerating climate change (Halpern et al. 2008; Doney et al. 2012; IPCC 2023). Nearly half of the world's foundational coastal habitats, including coral reefs, kelp forests, seagrasses and saltmarshes, have been lost or severely degraded (Wernberg et al. 2024). In response, global initiatives such as the UN Decade on Ecosystem Restoration and the Kunming–Montreal Global Biodiversity Framework call for rapid scaling of restoration actions to 2030 (UN 2019; Duarte et al. 2020; CBD 2021).

To meet these restoration goals at scale, marine practitioners are increasingly turning to aquaculture propagation in hatcheries. This mirrors terrestrial nurseries that supply vast quantities of seed and builds on decades of experience in fish and shellfish stock enhancement (Naish et al. 2007; Kitada 2018). Aquaculture is now central to large‐scale oyster and mussel restoration (Carranza and zu Ermgassen 2020), salt marsh and mangrove restoration (Vanderklift et al. 2020), and coral initiatives that outplant hundreds of thousands of fragments and juveniles annually (Lirman and Schopmeyer 2016; Boström‐Einarsson et al. 2020). Comparable approaches are rapidly emerging for macroalgae, particularly kelps, with new programs developing across > 17 ecoregions globally (Wood et al. 2024).

Genetic diversity underpins resilience and adaptive capacity in natural populations, shaping responses to environmental change across ecological and evolutionary scales. As a result, it is a foundational pillar of ecological restoration (IUCN 2013; Gann et al. 2019). In practice, the benefits of genetic diversity depend on how it is distributed among donor source populations and whether locally adapted variation is retained, as the provenance of donor stock strongly influences immediate outplant performance. Poorly matched genotypes may show reduced survival, growth and recruitment when introduced into contrasting thermal, hydrodynamic, or biotic regimes (Leimu and Fischer 2008; Hereford 2009). This is of particular concern for habitat‐forming, sessile taxa where local adaptation is common (Hays et al. 2021; Veenhof et al. 2023) and associated pressures such as temperature and grazing can also cause high restoration failure rates (Eger et al. 2022; Wood et al. 2024; Kendrick et al. 2025). High levels of genetic variation supported by adequate effective population size can also reduce the risk of genetic drift and inbreeding, maintain fecundity and mating success, and broaden tolerance to environmental and biotic stress, thereby sustaining establishment, self‐recruitment and multi‐generation persistence (Sgrò et al. 2011; Weeks et al. 2011; Wernberg et al. 2018). Genetic diversity also often underpins functional traits such as growth, morphology, nutritional or carbon content that underpin key ecosystem functions (Crutsinger et al. 2006; Whitham et al. 2006; Starko et al. 2026) that are important to maintain in restoration programs.

The risks of inadequate genetic management causing reduced genetic diversity during hatchery processes have been demonstrated in decades of captive breeding and stock enhancement research. Across finfish, shellfish and terrestrial plants, hatchery rearing often reduces effective population size, increases inbreeding and selects for traits favoured under benign culture conditions rather than in the wild (Hamilton 1994; Araki et al. 2007; Fraser 2008). Genomic changes can arise in a single generation (Christie et al. 2012, 2014, 2016), while fitness declines often emerge after several generations as genetic load accumulates. Poorly designed releases can also compromise genetic diversity in the wild, for example by reducing system‐wide N e via the Ryman–Laikre effect (Ryman and Laikre 1991; Laikre et al. 2010).

In fisheries, these risks are generally mitigated through broodstock guidelines, rare‐allele retention modelling and genetic monitoring of N e and diversity (Camara and Vadopalas 2009; FAO 2017). Equivalent conservation genetics frameworks, however, are almost entirely absent for marine habitat‐forming taxa such as kelps and corals, despite rapid declines in these ecosystems and growing investment in restoration (Eger et al. 2023; Wernberg et al. 2024; Wood et al. 2024).

Kelp forests dominate temperate reefs worldwide but have suffered widespread decline due to warming, marine heatwaves, eutrophication and overgrazing (Krumhansl et al. 2016; Eger et al. 2024). Hatchery‐based kelp restoration programs are now emerging globally (e.g., Wood et al. 2024), yet very few monitor genetic outcomes (Wood et al. 2020; Dykman et al. 2025), hindering the development of best‐practice guidelines to avoid unintended genetic risks. These same challenges apply to the design of macroalgal biobanks, where stored material must typically pass through nursery or hatchery phases prior to outplanting. Critical knowledge gaps include how to select optimal donor numbers, optimal breeding designs and the risks of domestication during hatchery rearing.

Here, we use kelp forests as a case study to empirically assess the genetic risks associated with hatchery reared restoration activities. We genotyped wild donors from two environmentally distinct populations and their hatchery‐bred F 1 cohorts to test if propagation processes impacted genetic diversity in one generation. Specifically, we asked whether our breeding protocols impacted genetic diversity at neutral and putatively adaptive outlier loci, increased inbreeding, reduced effective population size, or produced signals of selection to hatchery conditions (domestication). We then coupled these data with forward genetic simulations to quantify how donor number influences diversity retention across outplanted generations, providing quantitative targets for hatchery processes and a practical genetic monitoring framework for kelp restoration programs and policy.

2. Methods

2.1. Study Species

Ecklonia radiata dominates temperate reefs across southern Australasia and southeast Africa (Wernberg et al. 2019). Like other Laminariales, it has a haplodiplontic life cycle, with microscopic gametophytes and macroscopic sporophytes, enabling both vegetative propagation and zoospore‐based seeding. The species has experienced widespread climate‐ and water‐quality‐driven declines across Australia (Connell et al. 2008; Vergés et al. 2016; Wernberg et al. 2016; Young et al. 2023).

2.2. Restoration Hatchery Experiment

We sampled two genetically and environmentally distinct Western Australian kelp populations: Port Gregory (“warm”) and Perth (“cool”) (Vranken et al. 2021; Minne 2024). During peak reproduction (autumn 2022), ~100 reproductively active adults per site were collected by SCUBA. Individuals were collected at ~12 m depth from exposed rocky reef habitats at each site. To minimise sampling of closely related individuals, thalli spaced approximately 1–2 m apart were haphazardly selected across an area of ~100 m2. Samples were transported on ice in insulated containers to the UWA Watermans Bay Marine Science Centre and processed within 8 h.

For each population, we pooled sorus tissue from 100 of the donors from each site to produce two mass‐spawned hatchery cohorts. Two 3 cm sorus discs per donor were excised, hydrated in filtered seawater, and gently agitated to release zoospores. Wild donor tissue was sampled for genetic analysis at this time following the protocol described below (Figure 1). Zoospore suspensions were standardised (~5000 mL−1) and used within 24 h to seed basalt substrates (12 rocks per tank; 8 tanks per provenance) in 50 × 50 cm culture tanks containing UV‐treated seawater. Seasonal temperature and light regimes followed ambient conditions, and flow‐through seawater commenced after 2 weeks.

FIGURE 1.

FIGURE 1

Open in a new tab

Experimental design showing collection of warm and cool donors (n = 100 each), mass spawning, hatchery culture, and temporal sampling (5, 7, and 9 months post‐spore release) of F 1 cohorts.

F 1 sporophytes were destructively sampled at 5, 7 and 9 months post‐seeding by selecting one sporophyte from three different rocks per tank. Tissue was rinsed, snap‐frozen, freeze‐dried and sent to Diversity Arrays Technology (Canberra, Australia) for DNA extraction and DArTseq genotyping (Kilian et al. 2012). At the 7‐month time point, the warm treatment was sampled twice (six days apart), resulting in an increased sample size at this time point. F 1 samples were pooled for downstream analyses of genetic selection and structure; however, genetic diversity measures were calculated for each time point to determine if there was an impact of time. Genetic diversity metrics comparing the two sampling events in month 7 are presented separately in the Supporting Information (Table S1).

2.3. Genetic Analyses

Genetic analyses were conducted in R 4.3.1 (R Core Team 2023). Of 49,220 initial SNPs (mean depth = 6.79; 18.3% missing data), filtering was conducted sequentially in dartR (Gruber et al. 2018; Mijangos et al. 2022) by retaining loci with read depth between 5 and 12, reproducibility ≥ 0.96 and removing secondary loci at random. Loci with minor allele frequency ≤ 1% were excluded, followed by filtering for missingness (retaining loci with < 90% missing data and individuals with < 80% missing data), and removal of monomorphic loci. This resulted in a final dataset of 1601 loci across 301 individuals.

We identified outlier loci using redundancy analysis (RDA), testing associations with (i) provenance (“warm” vs. “cool”) and (ii) hatchery environment (wild vs. hatchery) (Rellstab et al. 2015). RDA is a multivariate ordination method that identifies associations between loci and explanatory variables, capturing putative multilocus responses to selection (Forester et al. 2018; Capblancq and Forester 2021). Predictor variables (population and hatchery versus wild cohorts) were coded as binary dummy variables. Redundancy Analysis (RDA) detected significant associations between 42 unique loci and the fitted variables (see Results). Candidate loci were annotated via BLASTx (e‐value < 10−1) against the closest gene model Ectocarpus siliculosus (v2) from the ORCAE database (https://bioinformatics.psb.ugent.be/gdb/ectocarpusV2/). Subsequent analyses were conducted using neutral, provenance‐associated, and hatchery‐associated subsets.

Population structure was visualised by PCoA (gl.pcoa, dartR) using bitwise distances. Pairwise genetic differentiation (Weir & Cockerham F ST) was estimated with 999 bootstrap replicates and FDR‐corrected p‐values.

We assessed donor representation in the F 1 cohorts using the neutral dataset. Effective population size (N e) was estimated using the molecular coancestry method (Nomura 2008) implemented in NeEstimator v2.1 (Do et al. 2014) via gl.LDNe (dartR). Genetic diversity metrics included observed and expected heterozygosity (H o, H e), percentage of polymorphic loci (PL), rarefied allelic richness (AR), private alleles (PA) and inbreeding (F IS). F IS estimates were assessed for significance using 1000 bootstrap replicates across loci to generate 95% confidence intervals in hierfstat (Goudet and Jombart 2024). These metrics are generally robust to moderate differences in sample size; allelic richness was rarefied and effective population size estimates incorporate sample size corrections, although we note that the proportion of polymorphic loci may be more sensitive to sample size and should be interpreted with caution when comparing donor and F 1 cohorts. Metrics were calculated separately for neutral and outlier loci using dartR, hierfstat and poppr (Kamvar et al. 2014).

We also explored how the number of restoration donors used might influence genetic diversity in future restored populations. Neutral genetic patterns were simulated for the F 1–F 100 offspring cohorts using gl.sim.offspring in dartR, utilising a 1:1 sex ratio and random mating (Praeger et al. 2022) and assuming no mixing with any remaining extant kelp plants in restored sites. Initial donor numbers ranged from two to 100 (increments of two up to 10, then by 10 hereafter), each bootstrapped 100 times, with 500 randomly mating offspring for each of 100 generations. Diversity (H e, PL) was calculated for F 1, F 3, F 10 and F 100 using gl.report.heterozygosity.

2.4. Survival and Growth

Kelp density and size on rocks in the hatchery were quantified at 5 and 7 months postseeding. Density was measured from three 1 cm‐radius quadrats per rock (3 rocks per tank). Linear mixed models (LMMs) were fitted using lme4 (Bates et al. 2015), with provenance (“warm” vs. “cool”) and time as fixed effects and tank nested within provenance as a random effect. Significance was assessed using Type III ANOVA (Kenward–Roger df). Estimated marginal means were obtained using emmeans (Lenth and Piaskowski 2025). Model assumptions were checked via residual diagnostics.

3. Results

3.1. Genetic Analyses

Redundancy Analysis (RDA) detected significant associations between 42 unique loci and the fitted variables. Of these, 27 loci (1.7% of all SNPs) were associated with provenance and 15 loci (0.9%) were associated with hatchery cultivation (Figure S1). BLAST analysis revealed that none of the candidate loci under hatchery selection had matches to annotated genes in the Ectocarpus genome. However, one locus (46357600‐33), polymorphic only in the warm donor population, was linked to growth, immunity and membrane transport. Two loci with polymorphic alleles exclusive to the cool donor population (100215211‐30 and 54904122‐13) were associated with protein transport and RNA processing.

Linkage disequilibrium analysis indicated that three‐quarters of Cool donors contributed to the F 1 hatchery gene pool (estimated N e = 74, Table 1), while warm hatchery kelp had an N e of just 21, about a fifth of the number of donors used (Table 2). In both cases, effective population sizes were substantially lower (7–25 X) than in the wild populations (cool N e = 512 and warm N e = 495).

TABLE 1.

Estimates of effective population size (N e) for wild and hatchery‐cultivated cohorts of Ecklonia radiata of warm and cool provenance calculated with N e Estimator.

N e a Overall R 2 Expected R 2 N donors (%) n
Cool donor 512 (473, 557) 0.011 0.011 100
Warm donor 495 (441, 563) 0.012 0.011 100
Cool hatchery 74 (68, 80) 0.034 0.033 100 (74%) 36
Warm Hatchery 21 (21, 21) 0.032 0.018 100 (21%) 62
Open in a new tab

Note: Calculated with 1559 neutral loci using the linkage disequilibrium method. 95% confidence intervals shown in brackets. N donors = number of donor kelp plants and the % that contributed to the F 1 generation. n = number samples used in analysis.

a

Lower and upper CI shown in brackets.

TABLE 2.

Genetic diversity metrics across time.

Population Cohort Age (months) n Neutral loci Provenance‐associated outliers Hatchery‐associated outliers
PL (%) AR PA (%) H o H e F IS PL (%) AR PA (%) H o H e F IS PL (%) AR PA (%) H o H e F IS
Warm Wild > 12 98 36.626 1.038 20.975 0.005 0.006 0.200 62.963 1.369 3.704 0.080 0.095 0.146 60.000 1.074 0.000 0.011 0.011 −0.005
Hatchery 62 17.960 3.784 0.004 0.006 0.183 66.667 0.000 0.051 0.640 0.150 93.333 20.000 0.127 0.171 0.323
5 12 29.787 1.026 0.449 0.003 0.005 0.250 18.519 1.153 0.000 0.045 0.044 −0.003 73.333 1.670 0.000 0.130 0.221 0.360
7 33 35.106 1.029 1.860 0.004 0.005 0.174 66.667 1.321 0.000 0.061 0.080 0.153 86.667 1.645 0.000 0.129 0.144 0.121
9 17 38.830 1.037 1.155 0.004 0.006 0.366 29.630 1.224 0.000 0.037 0.059 0.245 80.000 1.658 0.000 0.112 0.156 0.248
Cool Wild > 12 101 54.586 1.090 26.491 0.014 0.017 0.175 70.370 1.408 0.000 0.089 0.103 0.121 20.000 1.027 0.000 0.004 0.004 −0.005
Hatchery 36 27.646 3.784 0.011 0.016 0.206 66.667 0.000 0.064 0.111 0.415 33.333 0.000 0.008 0.028 0.526
5 9 12.957 1.099 1.347 0.015 0.019 0.226 48.148 1.459 0.000 0.097 0.122 0.200 13.333 1.131 0.000 0.008 0.033 0.807
7 18 16.164 1.078 1.411 0.009 0.015 0.361 55.556 1.428 0.000 0.054 0.093 0.348 6.667 1.063 0.000 0.004 0.017 0.773
9 9 7.633 1.090 1.090 0.008 0.012 0.315 51.852 1.466 0.000 0.051 0.112 0.440 26.667 1.232 0.000 0.015 0.040 0.500
Open in a new tab

Note: Based on putatively neutral (n = 1559) and outlier loci associated with Provenance (n = 27) and Hatchery (n = 15), sequenced across Donor (n = 199) and Hatchery (n = 102) Ecklonia radiata cohorts. All F IS estimates were significantly different from zero across groups, indicating consistent departures from random mating. The colour shading and bold formatting indicates these are the overall genetic diversity metrics calculated on each hatchery/wild cohort. In contrast, the rows below that are not coloured or in bold correspond to the hatchery cohorts broken down by age (5,7,9 months as indicated). Red corresponds to the warm population, blue corresponds to the cool population.

Abbreviations: AR = rarefied allelic richness; F IS = inbreeding coefficient; H e = expected heterozygosity; H o = observed heterozygosity; PA = private alleles; PL = polymorphic loci.

Genetic differentiation between warm and cool wild donor populations was relatively strong (pairwise F ST across all SNPs = 0.083, pairwise F ST for provenance‐associated SNPs = 0.356; Figure 2) and this differentiation was maintained and slightly elevated when comparing hatchery cohorts of contrasting provenance (F ST across all SNPs = 0.092, F ST provenance‐associated SNPs = 0.374). There was only a small amount of overall genetic differentiation between hatchery cohorts and their respective wild donors; this was mainly due to kelp of warm provenance for the overall dataset, but was accentuated in the hatchery‐associated loci dataset for both Warm and Cool hatchery cohorts (cool: F ST across all SNPs = 0.001, F st provenance‐associated SNPs = 0.01, F ST hatchery‐associated SNPs = 0.064; warm: F ST all SNPs = 0.015, F st provenance‐associated SNPs = 0.024, F ST hatchery‐associated SNPs = 0.116; Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

Genetic structure of donor kelp and hatchery‐cultivated cohorts. (a) Constrained ordination of axis 1 and 2 from Principle Component Analysis (PCoA) of four Ecklonia radiata cohorts (N = 297), shown for five subsets of the SNP data; all loci (n = 1601), putatively neutral loci (n = 1559), all outlier loci (n = 42), provenance‐associated outlier loci (n = 27) and hatchery‐associated outlier loci (n = 15). (b) Pairwise F ST estimates between corresponding cohorts, Values significant following the FDR correction shown in bold.

Cool donors exhibited roughly double the genetic diversity of warm donors (e.g., H e neutral: 0.017 vs. 0.006). This pattern remained in hatchery cohorts but diversity declined modestly relative to donors, with neutral H0 decreasing by ~21% on average (Table 2, Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

Genetic impacts of hatchery propagation on seaweed populations, showing changes in observed heterozygosity (H o), inbreeding coefficient (F is), and percent polymorphic loci (PL). Based on putatively neutral (n = 1559), provenance‐associated outliers (n = 27) and hatchery‐associated outliers (n = 15) loci sequenced across Donor (n = 199) and Hatchery (n = 102) Ecklonia radiata cohorts (pooled population diversity metrics across both warm and cool populations sampled at 5, 7 and 9 months old).

Inbreeding coefficients (F IS) were positive and significantly different to zero across wild populations (> 0.17) and were modestly higher in hatchery kelp, particularly within the cool cohort (> 0.2; Table 2).

Neutral genetic diversity (H o and AR) of hatchery cohorts was broadly similar to that of donors (Table 2; Figure 3). Metrics that captured rare alleles (PL and PA) indicated greater than 50% declines in neutral and provenance‐associated diversity in hatchery compared to wild cohorts (Table 2, Figure 3). In contrast, hatchery‐associated loci showed increased diversity, particularly in the warm cohort, which contained three fixed heterozygous alleles absent from both wild populations—strong evidence of hatchery‐driven selection or very early stochastic processes (Table 2, Figure 3). No consistent temporal trend was detected across 5‐, 7‐ and 9‐month timepoints (Table 2, Figure 3).

Our simulations revealed that H E plateaued by ~20 donors in F 1, but maintaining this across F 100 required ~100 donors. PL increased continuously with donor number, showing no saturation at 100 donors (Figure 4). Empirical warm F 1 diversity corresponded to simulations with only ~4 donors under a random mating model, consistent with a genetic bottleneck. Observed F 1 heterozygosity was consistently lower than predicted by the simulations (Figure 4), likely reflecting departures from model assumptions such as random mating and equal parental contribution, and indicating reduced realised genetic diversity in hatchery cohorts.

FIGURE 4.

FIGURE 4

Open in a new tab

Impact of initial number of donor kelp on genetic diversity estimates for cultivated and restored populations, under best case scenarios. Mean ± standard error expected heterozygosity (left) and percentage of polymorphic loci (right) for simulated F 1, F 3, F 10 and F 100 offspring, calculated over 100 bootstrap replicates of 2–100 initial donors (x‐axis). Estimates were calculated across 1559 neutral SNPs from wild donor Ecklonia radiata from two populations in Western Australia. Mean values for expected heterozygosity and percentage of polymorphic loci for true (non‐simulated) F 1 generations sampled in the hatchery indicated by red dotted lines.

3.2. Survival and Growth

Both warm‐ and cool‐provenance kelps grew successfully in the hatchery (Figure 5, Table S2). Density increased slightly (4–6 cm−2) but not significantly (F 1,56 = 0.51, p > 0.4). Size of sporophytes increased significantly with time (4–6 cm; F 1,56 = 4.62, p = 0.04). No provenance effect or interaction was detected on density or size (F 1,56 = 0.001, p > 0.9; F 1,56 = 0.004, p > 0.9, respectively).

FIGURE 5.

FIGURE 5

Open in a new tab

Density and growth of kelp in the hatchery. Effect of provenance (cool vs. warm) and month in hatchery (5 vs. 7) on density and size (mean ± s.e shown).

4. Discussion

Aquaculture‐based restoration offers important opportunities to rebuild degraded marine habitats, yet our results show that even the earliest stages of propagation can substantially reshape the genetic composition of restored populations. A single hatchery generation of kelp was sufficient to generate strong genetic drift, elevated inbreeding and shifts in allele frequencies at loci associated with both provenance and hatchery conditions. Notably, these changes occurred despite high apparent performance in the hatchery, indicating that early cultivation success does not necessarily translate into the genetic characteristics required for long‐term restoration, and that evolutionary risks can arise during the initial stages of propagation. Here, we explore the mechanisms driving these genetic changes during propagation, consider their potential consequences for outplanting and persistence, and discuss their implications for restoration practice and policy (Figure 6).

FIGURE 6.

FIGURE 6

Open in a new tab

Conceptual framework illustrating genetic risks across the restoration pipeline and points for mitigation. Adapted from Baums (2008). Each step of the restoration workflow can alter the genetic composition of restored cohorts. Donor selection determines the initial breadth of diversity entering the pipeline, while variation in sorus size, spore release, and fertilisation success can create strong reproductive skew and elevate inbreeding. Early gametophyte development and nursery cultivation expose cohorts to controlled conditions that may favour particular genotypes, generating early domestication signals. Following outplanting, environmental filters and differential survival further modify allele frequencies. Key mitigation options include increasing donor numbers, standardising spore release, maintaining separate gametophyte lines or sexes to equalise parental contribution, aligning hatchery conditions more closely with field environments, and incorporating routine genetic monitoring to detect early drift or selection.

4.1. Hatcheries Impose Powerful Evolutionary Filters

The reduction in effective population size (N e) from ~500 in the wild to 21–74 in the hatchery demonstrates that reproductive bottlenecks can emerge rapidly under mass‐spawning conditions. Only 21%–74% of donor individuals contributed genetically to the F 1 generation, indicating that a substantial proportion of available genetic diversity—particularly rare alleles—was lost at the outset of propagation. This pattern aligns with observations from finfish, shellfish and coral hatcheries, where unequal parental contributions drive drift, reductions in heterozygosity and increased inbreeding (Araki et al. 2007; Christie et al. 2012). Such reproductive skew is therefore also likely to be a key mechanism shaping genetic outcomes in kelp restoration systems, unless measures are developed and taken to mitigate this risk (Baums 2008; Baums et al. 2022).

Elevated inbreeding coefficients, particularly within warm‐provenance cohorts, further suggest that kin mating and self‐fertilisation contributed to reduced genetic diversity. These processes are plausible in laminarian kelps, where gametophytes develop in close proximity and reproductive synchrony varies among individuals. Selfed or inbred sporophytes typically show reduced growth and competitive performance relative to outcrossed individuals (Raimondi et al. 2004; Johansson et al. 2013), and recent field trials indicate that selfed individuals are less likely to survive early outplanting stages (Dykman et al. 2025). As climatic variability intensifies, reduced genetic variation and elevated inbreeding may constrain the capacity of restored populations to persist under changing environmental conditions (Wernberg et al. 2018). Hence, mitigating this risk through ensuring more even parental contributions to outplanted generations is essential.

4.2. Early Domestication Responses Occur Alongside Drift and Inbreeding

Genomic analyses revealed two forms of genetic divergence: (i) persistent differences associated with donor provenance, and (ii) shifts in allele frequencies arising during the hatchery phase alone. The former likely reflect long‐term adaptation to regional environmental regimes, while the latter are consistent with changes occurring within a single generation under controlled conditions. Although most hatchery‐associated loci lacked functional annotation to the closest available reference genome, this may reflect phylogenetic divergence rather than a true absence of functional relevance. The consistent association of these loci with cultivated cohorts across tanks and rocks suggests that environmental conditions in the hatchery (e.g., temperature, light, flow, nutrient availability or microbial conditions) may have influenced allele frequencies. However, because tanks and substrates were inoculated from a common mixed spore slurry, we cannot exclude the possibility that early stochastic processes (e.g., variance in reproductive success or genetic drift during initial settlement) contributed to these patterns and were subsequently propagated during grow‐out. Together, these results are consistent with a combination of early‐stage stochasticity and potential selection under hatchery conditions.

4.3. Loss of Allelic Diversity Signals Vulnerability Despite Stable Heterozygosity

Genetic metrics also showed differing sensitivities to hatchery propagation. Expected heterozygosity changed little between donor and hatchery cohorts, but allelic richness, private alleles and the proportion of polymorphic loci declined substantially. These metrics are sensitive to the loss of rare variants that contribute disproportionately to adaptive capacity (Allendorf and Luikart 2006; Hoban et al. 2020). Recent whole‐genome data from the bull kelp (Nereocystis luetkeana) show that reductions in effective population size can rapidly elevate drift load and erode adaptive potential, even when heterozygosity appears relatively stable (Bemmels et al. 2025). Our forward simulations reinforce this pattern, as modest donor numbers were sufficient to maintain heterozygosity, but retaining rare alleles required substantially larger broodstock numbers. Monitoring heterozygosity alone may therefore mask early losses of evolutionary potential during hatchery propagation. As most kelp restoration programs use fewer than 30 donors (Wood et al. 2024; Dykman et al. 2025), erosion of adaptive variation is likely unless donor numbers are increased and breeding designs are optimised for the sector.

4.4. Genomic Composition May Help Explain Low Early Outplant Survival

The 9‐month samples analysed here were collected from individuals in the hatchery only; however, outplanting trials were conducted separately on a subset of individuals. Preliminary field observations indicated equally low persistence of outplanted sporophytes from both provenance groups (Figure S2). Although in situ environmental stressors undoubtedly contribute to early mortality in the field, the genetic changes observed here (reduced allelic diversity, elevated inbreeding and early domestication) provide plausible additional mechanisms for their loss. Similar mismatches between hatchery and field performance have been documented in finfish, shellfish and corals, where genotypes favoured in culture or hatchery environments perform poorly in natural environments (Naish and Hard 2008; Christie et al. 2014; Bosch et al. 2023). Outplant success depends not only on retaining overall genetic diversity, but also on minimising maladaptive selection during propagation. Hatchery environments that better reflect field conditions, including realistic temperature and light regimes, natural microbial communities, variable flow or staged stress conditioning, may help reduce unintended domestication and favour genotypes capable of persisting under natural selection.

4.5. Implications for Marine Restoration Design and Policy

Our results have direct implications for kelps and the marine restoration sector more broadly, as the sector enters a stage of heightened investment in large‐scale activities (Figure 6). First, donor pools need to be sufficiently large to retain rare alleles. Our empirical and simulation results suggest that > 100 donors may be required to achieve this in many cases. Second, modifying protocols for spore release and fertilisation to reduce reproductive skew can increase realised N e. This can be achieved by maintaining separate gametophyte family lines or male–female cultures and using controlled crosses to equalise contributions among donors. For example, our kelp restoration work in this area now separates hatchery tanks into multiple groups of 10 distinct individuals, to maximise genetic diversity retained. Third, aligning hatchery conditions with environmental regimes at restoration sites, through temperature matching, natural microbial communities, realistic flow regimes or shortened culture periods, may reduce domestication effects and ensure that restored populations are best adapted to field conditions. Finally, integrating routine genetic monitoring into restoration pipelines would allow practitioners to detect the risks of early drift, inbreeding and selection and adapt protocols accordingly.

4.6. Future Steps and Conclusion

Our results demonstrate that hatchery propagation can reshape the genetic composition of restoration cohorts within a single generation, even when donor numbers are high and short‐term performance appears strong. The reduction in rare alleles, elevated inbreeding and signals of early domestication observed here indicate that evolutionary risks can arise well before outplanting. As marine restoration scales globally, integrating genetic safeguards into propagation pipelines will be essential to ensure that restored populations retain the adaptive capacity required to persist under accelerating environmental change.

Genetic management must become a formal and enforceable component of marine restoration practice. Quantitative targets for donor number, realised effective population size and diversity retention should be embedded within project design, permitting frameworks and funding criteria. Without explicit standards and monitoring, restoration efforts risk unintentionally eroding the evolutionary resilience they seek to rebuild.

Funding

G.W. was supported by an Australian Research Council (ARC) Linkage Grant (LP190100346) and an ARC Industry Fellowship (IE230100464). T.W., M.A.C. and K.F.‐D. were supported by ARC grants DP190100058 and LP190100346, KFD by ARC Fellowship FT230100214, and T.W. and M.A.C. by ARC grant DP240100230. T.W., K.F.‐D., G.W. and M.A.C. also received funding from the Norwegian Research Council (GecoKelp Project no. 335371), and T.W. and K.F.‐D. were also supported by the Schmidt Marine Technology Partners.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Genetic diversity metrics across time, including the breakdown of both sampling points at 7 months for the warm hatchery cohort.

Figure S1: RDA analysis plot showing loci in grey and cohort in coloured circles which match Figure 2 in the main text.

Table S2: ANOVA output on LMM testing the effects of provenance and age on kelp density and size in the hatchery.

EVA-19-e70257-s001.docx (3.8MB, docx)

Figure S2: Depicts low survival and growth of kelp in the hatchery versus field.

EVA-19-e70257-s002.docx (3.1MB, docx)

Acknowledgements

The authors wish to thank the members of the Wernberg lab, particularly Mason Sullivan, Dani Bandt, Grace Edwards and Nahlah Alsuwaiyan, and the many volunteers who made this work possible. We thank the two anonymous reviewers for their thoughtful and constructive feedback, which substantially improved the clarity and overall quality of this manuscript. G.W. was supported by an Australian Research Council (ARC) Linkage Grant (LP190100346) and an ARC Industry Fellowship (IE230100464). T.W., M.A.C. and K.F.‐D. were supported by ARC grants DP190100058 and LP190100346, K.F.‐D. by ARC Fellowship FT230100214, and T.W. and M.A.C. by ARC grant DP240100230. T.W., K.F.‐D., G.W. and M.A.C. also received funding from the Norwegian Research Council (GecoKelp Project no. 335371). T.W. and K.F.‐D. were also supported by the Schmidt Marine Technology Partners. Open access publishing facilitated by The University of Western Australia, as part of the Wiley ‐ The University of Western Australia agreement via the Council of Australasian University Librarians.

Data Availability Statement

Data for this study are available at the Dryad Digital Repository: https://doi.org/10.5061/dryad.2v6wwq049.

References

  1. Allendorf, F. , and Luikart G.. 2006. Conservation and Genetics of Populations. Blackwell Publishing. [Google Scholar]
  2. Araki, H. , Cooper B., and Blouin M. S.. 2007. “Genetic Effects of Captive Breeding Cause a Rapid, Cumulative Fitness Decline in the Wild.” Science 318: 100–103. [DOI] [PubMed] [Google Scholar]
  3. Bates, D. , Mächler M., Bolker B., and Walker S.. 2015. “Fitting Linear Mixed‐Effects Models Using lme4.” Journal of Statistical Software 67: 1–48. [Google Scholar]
  4. Baums, I. B. 2008. “A Restoration Genetics Guide for Coral Reef Conservation.” Molecular Ecology 17: 2796–2811. [DOI] [PubMed] [Google Scholar]
  5. Baums, I. B. , Chamberland V. F., Locatelli N. S., and Conn T.. 2022. “Maximizing Genetic Diversity in Coral Restoration Projects.” In Coral Reef Conservation and Restoration in the Omics Age, edited by van Oppen M. J. H. and Aranda Lastra M., 35–53. Springer International Publishing. [Google Scholar]
  6. Bemmels, J. B. , Starko S., Weigel B. L., et al. 2025. “Population Genomics Reveals Strong Impacts of Genetic Drift Without Purging and Guides Conservation of Bull and Giant Kelp.” Current Biology 35: 688–698.e688. [DOI] [PubMed] [Google Scholar]
  7. Bosch, W. , Pandit S., Sandford B., Temple G., Johnston M., and Larsen D.. 2023. “Effects of Volitional Emigration Timing and Smolt Size on Survival and Age‐At‐Return in a Pacific Salmon Hatchery Population.” Environmental Biology of Fishes 106: 1–23. [Google Scholar]
  8. Boström‐Einarsson, L. , Babcock R. C., Bayraktarov E., et al. 2020. “Coral Restoration—A Systematic Review of Current Methods, Successes, Failures and Future Directions.” PLoS One 15: e0226631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Camara, M. D. , and Vadopalas B.. 2009. “Genetic Aspects of Restoring Olympia Oysters and Other Native Bivalves: Balancing the Need for Action, Good Intentions, and the Risks of Making Things Worse.” Journal of Shellfish Research 28: 121–145. [Google Scholar]
  10. Capblancq, T. , and Forester B. R.. 2021. “Redundancy Analysis: A Swiss Army Knife for Landscape Genomics.” Methods in Ecology and Evolution 12: 2298–2309. [Google Scholar]
  11. Carranza, A. , and zu Ermgassen P. S. E.. 2020. “A Global Overview of Restorative Shellfish Mariculture.” Frontiers in Marine Science 7: 2020. [Google Scholar]
  12. CBD . 2021. “Convention on Biological Diversity. First Draft of the Post 2020 Global Biodiversity Framework.”
  13. Christie, M. R. , Ford M. J., and Blouin M. S.. 2014. “On the Reproductive Success of Early‐Generation Hatchery Fish in the Wild.” Evolutionary Applications 7: 883–896. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Christie, M. R. , Marine M., French R., Waples R. S., and Blouin M.. 2012. “Effective Size of a Wild Salmonid Population Is Greatly Reduced by Hatchery Supplementation.” Heredity 109: 254–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Christie, M. R. , Marine M. L., Fox S. E., French R. A., and Blouin M. S.. 2016. “A Single Generation of Domestication Heritably Alters the Expression of Hundreds of Genes.” Nature Communications 7: 10676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Connell, S. D. , Russell B. D., Turner D. J., et al. 2008. “Recovering a Lost Baseline: Missing Kelp Forests From a Metropolitan Coast.” Marine Ecology Progress Series 360: 63–72. [Google Scholar]
  17. Crutsinger, G. M. , Collins M. D., Fordyce J. A., Gompert Z., Nice C. C., and Sanders N. J.. 2006. “Plant Genotypic Diversity Predicts Community Structure and Governs an Ecosystem Process.” Science 313: 966–968. [DOI] [PubMed] [Google Scholar]
  18. Do, C. , Waples R. S., Peel D., Macbeth G. M., Tillett B. J., and Ovenden J. R.. 2014. “NeEstimator v2: Re‐Implementation of Software for the Estimation of Contemporary Effective Population Size (Ne) From Genetic Data.” Molecular Ecology Resources 14: 209–214. [DOI] [PubMed] [Google Scholar]
  19. Doney, S. C. , Ruckelshaus M., Emmett Duffy J., et al. 2012. “Climate Change Impacts on Marine Ecosystems.” Annual Review of Marine Science 4: 11–37. [DOI] [PubMed] [Google Scholar]
  20. Duarte, C. M. , Agusti S., Barbier E., et al. 2020. “Rebuilding Marine Life.” Nature 580: 39–51. [DOI] [PubMed] [Google Scholar]
  21. Dykman, L. N. , Steell S. C., Bemmels J. B., et al. 2025. “Testing the Roles of Local Adaptation and Genetic Diversity to Improve Giant Kelp ( Macrocystis pyrifera ) Restoration.” Restoration Ecology 33: e70114. [Google Scholar]
  22. Eger, A. M. , Altamirano M., Giraldo‐Ospina A., et al. 2023. “A Roadmap for Protecting and Restoring 4 Million Hectares of Kelp Forests by 2040.”
  23. Eger, A. M. , Eddy N., McHugh T. A., et al. 2024. “State of the World's Kelp Forests.” One Earth 7: 1927–1931. [Google Scholar]
  24. Eger, A. M. , Marzinelli E. M., Christie H., et al. 2022. “Global Kelp Forest Restoration: Past Lessons, Present Status, and Future Directions.” Biological Reviews 97: 1449–1475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. FAO . 2017. Incorporating Genetic Diversity and Indicators Into Statistics and Monitoring of Farmed Aquatic Species and Their Wild Relatives. Food and Agriculture Organisation. [Google Scholar]
  26. Forester, B. R. , Lasky J. R., Wagner H. H., and Urban D. L.. 2018. “Comparing Methods for Detecting Multilocus Adaptation With Multivariate Genotype–Environment Associations.” Molecular Ecology 27: 2215–2233. [DOI] [PubMed] [Google Scholar]
  27. Fraser, D. J. 2008. “How Well Can Captive Breeding Programs Conserve Biodiversity? A Review of Salmonids.” Evolutionary Applications 1: 535–586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Gann, G. D. , McDonald T., Walder B., et al. 2019. “International Principles and Standards for the Practice of Ecological Restoration. Second Edition.” Restoration Ecology 27: S1–S46. [Google Scholar]
  29. Goudet, J. , and Jombart T.. 2024. “Hierfstat: Estimation and Tests of Hierarchical F‐Statistics,” R Package Version 0.5‐11.
  30. Gruber, B. , Unmack P. J., Berry O. F., and Georges A.. 2018. “Dartr: An r Package to Facilitate Analysis of SNP Data Generated From Reduced Representation Genome Sequencing.” Molecular Ecology Resources 18: 691–699. [DOI] [PubMed] [Google Scholar]
  31. Halpern, B. S. , Walbridge S., Selkoe K. A., et al. 2008. “A Global Map of Human Impact on Marine Ecosystems.” Science 319: 948–952. [DOI] [PubMed] [Google Scholar]
  32. Hamilton, M. B. 1994. “Ex Situ Conservation of Wild Plant Species: Time to Reassess the Genetic Assumptions and Implications of Seed Banks.” Conservation Biology 8: 39–49. [Google Scholar]
  33. Hays, C. G. , Hanley T. C., Hughes A. R., Truskey S. B., Zerebecki R. A., and Sotka E. E.. 2021. “Local Adaptation in Marine Foundation Species at Microgeographic Scales.” Biological Bulletin 241: 16–29. [DOI] [PubMed] [Google Scholar]
  34. Hereford, J. 2009. “A Quantitative Survey of Local Adaptation and Fitness Trade‐Offs.” American Naturalist 173: 579–588. [DOI] [PubMed] [Google Scholar]
  35. Hoban, S. , Bruford M., D'Urban Jackson J., et al. 2020. “Genetic Diversity Targets and Indicators in the CBD Post‐2020 Global Biodiversity Framework Must Be Improved.” Biological Conservation 248: 108654. [Google Scholar]
  36. IPCC . 2023. “Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.”
  37. IUCN . 2013. International Union for Conservation of Nature. Guidelines for Reintroductions and Other Conservation Translocations. Version 1.0. IUCN Species Survival Commission. [Google Scholar]
  38. Johansson, M. L. , Raimondi P. T., Reed D. C., Coelho N. C., Serrão E. A., and Alberto F. A.. 2013. “Looking Into the Black Box: Simulating the Role of Self‐Fertilization and Mortality in the Genetic Structure of Macrocystis pyrifera .” Molecular Ecology 22: 4842–4854. [DOI] [PubMed] [Google Scholar]
  39. Kamvar, Z. N. , Tabima J. F., and Grünwald N. J.. 2014. “Poppr: An R Package for Genetic Analysis of Populations With Clonal, Partially Clonal, and/or Sexual Reproduction.” PeerJ 2: e281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kendrick, G. A. , Austin R., Ferretto G., van Keulen M., and Verduin J. J.. 2025. “Lessons Learnt From Revisiting Decades of Seagrass Restoration Projects in Cockburn Sound, Southwestern Australia.” Restoration Ecology 33: e70040. [Google Scholar]
  41. Kilian, A. , Wenzl P., Huttner E., et al. 2012. “Diversity Arrays Technology: A Generic Genome Profiling Technology on Open Platforms.” In Data Production and Analysis in Population Genomics: Methods and Protocols, edited by Pompanon F. and Bonin A., 67–89. Humana Press. [DOI] [PubMed] [Google Scholar]
  42. Kitada, S. 2018. “Economic, Ecological and Genetic Impacts of Marine Stock Enhancement and Sea Ranching: A Systematic Review.” Fish and Fisheries 19: 511–532. [Google Scholar]
  43. Krumhansl, K. A. , Okamoto D. K., Rassweiler A., et al. 2016. “Global Patterns of Kelp Forest Change Over the Past Half‐Century.” Proceedings of the National Academy of Sciences 113: 13785–13790. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Laikre, L. , Schwartz M. K., Waples R. S., and Ryman N.. 2010. “Compromising Genetic Diversity in the Wild: Unmonitored Large‐Scale Release of Plants and Animals.” Trends in Ecology & Evolution 25: 520–529. [DOI] [PubMed] [Google Scholar]
  45. Leimu, R. , and Fischer M.. 2008. “A Meta‐Analysis of Local Adaptation in Plants.” PLoS One 3: e4010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lenth, R. , and Piaskowski J.. 2025. “emmeans: Estimated Marginal Means, aka Least‐Squares Means.” R package version 2.0.1.
  47. Lirman, D. , and Schopmeyer S.. 2016. “Ecological Solutions to Reef Degradation: Optimizing Coral Reef Restoration in the Caribbean and Western Atlantic.” PeerJ 4: e2597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Mijangos, J. L. , Gruber B., Berry O., Pacioni C., and Georges A.. 2022. “dartR v2: An Accessible Genetic Analysis Platform for Conservation, Ecology and Agriculture.” Methods in Ecology and Evolution 13: 2150–2158. [Google Scholar]
  49. Minne, A. 2024. “Incorporating Genomic Data to Assess The Adaptive Potential of Australian Kelp Forests to Climate Change.” Doctoral dissertation, The University of Western Australia.
  50. Naish, K. A. , and Hard J. J.. 2008. “Bridging the Gap Between the Genotype and the Phenotype: Linking Genetic Variation, Selection and Adaptation in Fishes.” Fish and Fisheries 9: 396–422. [Google Scholar]
  51. Naish, K. A. , Taylor J. E. 3rd, Levin P. S., et al. 2007. “An Evaluation of the Effects of Conservation and Fishery Enhancement Hatcheries on Wild Populations of Salmon.” Advances in Marine Biology 53: 61–194. [DOI] [PubMed] [Google Scholar]
  52. Nomura, T. 2008. “Estimation of Effective Number of Breeders From Molecular Coancestry of Single Cohort Sample.” Evolutionary Applications 1: 462–474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Praeger, C. , Magnusson M., and Lawton R.. 2022. “Optimising the Zoospore Release, Germination, Development of Gametophytes and Formation of Sporophytes of Ecklonia Radiata.” Journal of Applied Phycology 34: 2535–2549. [Google Scholar]
  54. R Core Team . 2023. “A Language and Environment for Statistical Computing.” Vienna, Austria.
  55. Raimondi, P. T. , Reed D. C., Gaylord B., and Washburn L.. 2004. “Effects of Self‐Fertilization in the Giant Kelp, Macrocystis pyrifera .” Ecology 85: 3267–3276. [Google Scholar]
  56. Rellstab, C. , Gugerli F., Eckert A. J., Hancock A. M., and Holderegger R.. 2015. “A Practical Guide to Environmental Association Analysis in Landscape Genomics.” Molecular Ecology 24: 4348–4370. [DOI] [PubMed] [Google Scholar]
  57. Ryman, N. , and Laikre L.. 1991. “Effects of Supportive Breeding on the Genetically Effective Population Size.” Conservation Biology 5: 325–329. [Google Scholar]
  58. Sgrò, C. M. , Lowe A. J., and Hoffmann A. A.. 2011. “Building Evolutionary Resilience for Conserving Biodiversity Under Climate Change.” Evolutionary Applications 4: 326–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Starko, S. , Burkholz C., Edgeloe J. M., et al. 2026. “Evidence of a Genomic Basis for Growth Rate Variation in a Natural Kelp Population.” Scientific Reports 16: 6622. 10.1038/s41598-026-36286-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. UN . 2019. “United Nations Environment Agency. Resolution 73/284: United Nations Decade on Ecosystem Restoration 2020–2030).” Accessed 22 December 2023. https://undocs.org/A/RES/73/284.
  61. Vanderklift, M. A. , Doropoulos C., Gorman D., et al. 2020. “Using Propagules to Restore Coastal Marine Ecosystems.” Frontiers in Marine Science 7: 724. [Google Scholar]
  62. Veenhof, R. J. , Champion C., Dworjanyn S. A., Schwoerbel J., Visch W., and Coleman M. A.. 2023. “Projecting Kelp (Ecklonia Radiata) Gametophyte Thermal Adaptation and Persistence Under Climate Change.” Annals of Botany 133: 153–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Vergés, A. , Doropoulos C., Malcolm H. A., et al. 2016. “Long‐Term Empirical Evidence of Ocean Warming Leading to Tropicalization of Fish Communities, Increased Herbivory, and Loss of Kelp.” Proceedings of the National Academy of Sciences 113: 13791–13796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Vranken, S. , Wernberg T., Scheben A., et al. 2021. “Genotype–Environment Mismatch of Kelp Forests Under Climate Change.” Molecular Ecology 30: 3730–3746. [DOI] [PubMed] [Google Scholar]
  65. Weeks, A. R. , Sgro C. M., Young A. G., et al. 2011. “Assessing the Benefits and Risks of Translocations in Changing Environments: A Genetic Perspective.” Evolutionary Applications 4: 709–725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Wernberg, T. , Bennett S., Babcock R. C., et al. 2016. “Climate‐Driven Regime Shift of a Temperate Marine Ecosystem.” Science 353: 169–172. [DOI] [PubMed] [Google Scholar]
  67. Wernberg, T. , Coleman M. A., Babcock R. C., et al. 2019. “Biology and Ecology of the Globally Significant Kelp Ecklonia Radiata.” Oceanography and Marine Biology 57: 265–324. [Google Scholar]
  68. Wernberg, T. , Coleman M. A., Bennett S., Thomsen M. S., Tuya F., and Kelaher B. P.. 2018. “Genetic Diversity and Kelp Forest Vulnerability to Climatic Stress.” Scientific Reports 8: 1851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Wernberg, T. , Thomsen M. S., Baum J. K., et al. 2024. “Impacts of Climate Change on Marine Foundation Species.” Annual Review of Marine Science 16: 247–282. [DOI] [PubMed] [Google Scholar]
  70. Whitham, T. G. , Bailey J. K., Schweitzer J. A., et al. 2006. “A Framework for Community and Ecosystem Genetics: From Genes to Ecosystems.” Nature Reviews Genetics 7: 510–523. [DOI] [PubMed] [Google Scholar]
  71. Wood, G. , Filbee‐Dexter K., Coleman M. A., et al. 2024. “Upscaling Marine Forest Restoration: Challenges, Solutions and Recommendations From the Green Gravel Action Group.” Frontiers in Marine Science 11: 1364263. [Google Scholar]
  72. Wood, G. , Marzinelli E. M., Vergés A., Campbell A. H., Steinberg P. D., and Coleman M. A.. 2020. “Using Genomics to Design and Evaluate the Performance of Underwater Forest Restoration.” Journal of Applied Ecology 57: 1988–1998. [Google Scholar]
  73. Young, M. A. , Critchell K., Miller A. D., et al. 2023. “Mapping the Impacts of Multiple Stressors on the Decline in Kelps Along the Coast of Victoria, Australia.” Diversity and Distributions 29: 199–220. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1: Genetic diversity metrics across time, including the breakdown of both sampling points at 7 months for the warm hatchery cohort.

Figure S1: RDA analysis plot showing loci in grey and cohort in coloured circles which match Figure 2 in the main text.

Table S2: ANOVA output on LMM testing the effects of provenance and age on kelp density and size in the hatchery.

EVA-19-e70257-s001.docx (3.8MB, docx)

Figure S2: Depicts low survival and growth of kelp in the hatchery versus field.

EVA-19-e70257-s002.docx (3.1MB, docx)

Data Availability Statement

Data for this study are available at the Dryad Digital Repository: https://doi.org/10.5061/dryad.2v6wwq049.

Articles from Evolutionary Applications are provided here courtesy of Wiley

RESOURCES