Short-term effects of community-based marine reserves on green abalone, as revealed by population studies

Marine reserves (MRs) are implemented worldwide to protect, restore, and manage marine ecosystems and species. However, it is important to document the positive effects those marine reserves have on slow-growth, temperate invertebrates such as abalone. Abalone, Haliotis spp., are marine gastropods of high economic value extracted worldwide for decades, which has led to fisheries-driven population decreases. In this work, we focused on a case study and assessed the short-term (1–2 years) effects of marine reserves established and managed by a local fishing cooperative at Guadalupe Island, Mexico. We evaluated the population status of green abalone, H. fulgens, by conducting (1) an assessment of the green abalone population around Guadalupe Island through subtidal monitoring and (2) an evaluation of the effect of two recently established marine reserves on population parameters such as the increase in density (individuals·m2), biomass, number of aggregated abalone, egg production, and proportion of individuals bigger than 150 mm (minimum harvest size) compared to fished areas. To assess the population around Guadalupe Island, we surveyed 11,160 m2 during 2020 and 2021. We recorded 2327 green abalones with a mean ± SE shell length of 135.978 ± 0.83 mm and a mean density of 0.21 ± 0.02 individuals·m2. All variables were statistically higher at the MRs except for shell length in 2021. In this work, we report for the first time the green abalone population status at Guadalupe Island and a positive short-term biological response to community-based marine reserves. This study suggests that a network of MRs combined with good management could help abalone populations in the short term in Guadalupe Island, potentially leading to more sustainable fishing practices and social-ecological resilience.


Study site
Guadalupe is an oceanic island of volcanic origin located within the Mexican Pacific, 241 km off the western coast of the Baja California Peninsula (Fig. 1).Guadalupe Island is located in a transition zone in the Warm Temperate Northeast Pacific marine province, between the Southern California Bight and the Magdalena Transition ecoregions 31 .This particular location results in high ecosystem complexity and endemism.Guadalupe Island has the highest conservation status by the Mexican Government of Biosphere Reserve since 2005 31 .

Marine reserve establishment and monitoring
On July 1st, 2019, the first MR named "Plancha", was established by the fishing cooperative.Plancha MR covers an area of 9707 m 2 and was designated based on the ecological knowledge of the local fishing community 32 .The selection of this site was primarily influenced by its natural protection as a bay and the abundant presence of macroalgae, the main food source for abalone.Following this, the second MR, "Gaviota", was established on July 1st, 2020, encompassing an area of 10,885 m 2 .Gaviota MR, located at the wave-protected southern tip of Guadalupe Island, is recognized by local fishers for its importance as a green abalone reproductive aggregation site, further emphasizing the significance of establishing this second MR.
We conducted two monitoring campaigns to evaluate Guadalupe Island's green abalone populations at 32 sites around the island, including 30 fishing sites and the 2 MRs (Gaviota is considered a fishing site in 2020 and a MR in 2021).The first monitoring campaign took place from June 15th to July 15th, 2020, 1 year after the establishment of Plancha MR and just before the closure of Gaviota MR.The second monitoring campaign occurred from June 15th to July 15th, 2021, marking 2 years since the establishment of Plancha MR and 1 year since the establishment of Gaviota MR.Both monitoring campaigns coincided with the end of the fishing season, following the Mexican fishery zone 1 (2021: Februrary 1st-June 30th) (Fig. 2).
In the 2020 monitoring campaign, we deployed 94 transects at 31 sites, which included 2 transects inside the Plancha MR and 92 in fishing sites.For the 2021 campaign, we deployed 92 transects at 22 sites, with 7 transects inside and 85 outside the MRs.To facilitate the interpretation of our results, we grouped the 32 sites into 9 site locations: fishing sites numbered from 1 to 8, and MRs represented by location 9.The grouping was based on the official fishing polygons defined by the cooperative for the rotational fishery (Fig. 1).The rationale behind this grouping was to align the study with the cooperative's specific interests in assessing their fishery management within the rotational polygons.Therefore, the grouping was not driven by any additional criteria but rather by the cooperative's desire to obtain data relevant to their rotational polygon-based fishery management.Our belt transects covered areas of 60 m 2 each and were placed haphazardly on the rocky seafloor at depths ranging from 1 to 20 m, corresponding to the locations of the local abalone fishery.During these surveys, the best way that the visual sample and benthic habitat allowed, we recorded every green abalone with the total shell length in cm, measured by hand using the scale marked on the data table.We converted the data to mm.We also logged their location on the transect.With these data, we calculated the population parameters: proportion of individuals bigger than 150 mm (minimum harvest size), estimated body weight, density, total biomass, aggregations, and potential egg production.
Within each transect, we calculated the proportion (%) of abalone exceeding the minimum harvest size of 150 mm by tallying the number and dividing it by the total number of observed abalone.
We recorded aggregations at each transect as the number of groups of two or more abalone with no more than 1.5 m between two individuals, a critical distance for fertilization success for abalone 33 .We then obtained the number of individuals aggregated per transect.
We calculated the density (D, individuals•m 2 ) by dividing the number of abalone counted by the area.We converted abalone shell length to body weight (W) modifying the formula reported in 34 with a local, speciesspecific estimate of length-weight (R 2 = 0.953, P < 0.05): Then we obtained the total biomass (TB) per transect.
We determined the eggs produced per mature female (E) based on 35 : where L is the green abalone shell length in mm.We then calculated the potential egg production using adult abalones (> 49 mm as recommended for red abalone H. rufescens in 36 with the mean weights and densities per transect.Comparisons of egg production were then made based on protection (MRs vs. fished sites).
We also estimated the potential egg production per m 2 : where P s is the potential egg production per m 2 , E is the eggs produced per mature female green abalone (obtained in Eq. 4), O >49 mm is all counted adult organisms (> 49 mm), S is the sex ratio (0.5), and A is the monitoring area (m 2 ).We report egg production in millions of eggs and use eggs•m 2 for the discussion.(5) www.nature.com/scientificreports/

Statistical analyses
For the analyses, we used grouped site locations.Then, we analyzed the data at the transect level (N).In this approach, each transect was treated as an individual data point.We calculated the mean value for all the transects within each grouped site location.We tested for differences in abalone shell length, densities, total biomass, aggregations, egg production, and proportion of individuals > 150 mm among the categorical factor locations.Due to violations of homoscedasticity and variance assumptions in our data, we resorted to Kruskal-Wallis (K-W) tests for each year separately followed by Dunn tests.We analyzed the effect of the protection of MRs in the abalone biological variables against fished sites per year with Mann-Whitney U tests (M-W).Lastly, we analyzed changes inside each MR using M-W tests with both years as independent variables against the biological variables.We conducted the statistical analyses with R v.4.1.1 (R Core Team 2020) and JMP V.16 (Statistical Discovery LLC).

Results
The total monitored area during the 2 years was 11,160 m 2 from 186 transects of 60 m 2 each.In the first monitoring campaign in 2020, we monitored 5640 m 2 , and sampled 1220 green abalone equal to a mean density (± SE) of 0.21 ± 0.02 ind•m 2 with a mean size of 133.91 ± 1.46 mm and a mean estimated weight of 405.90 ± 11.32 g.For the second campaign in 2021, we monitored 5520 m 2 , and sampled 1107 green abalone equal to a mean density of 0.20 ± 0.02 ind•m 2 with a mean size of 138 ± 0.75 mm and a mean weight of 421.73 ± 5.99 g.

Monitoring campaign
We evaluated the effect of grouped site location around Guadalupe Island for the 2021 monitoring campaign.

Marine reserves in 2021
The second MR, Gaviota, was established on July 1st, 2020.For the 2021 monitoring campaign, Plancha MR had been closed to abalone fishing for 2 years and Gaviota MR for 1 year (Fig. 2).For these results, we used transects in both MRs (N = 7) versus the fished sites (N = 85).Again, in the MRs we found a higher mean (± SE) abalone proportion above the harvest size (U = 557, P < 0.0001; MRs = 14.2).

Discussion
We assessed the green abalone population and analyzed the short-term effects of two marine reserves (MRs) in the Northeastern Pacific.Our study is the first to report data on the green abalone population around Guadalupe Island.We focus on the density, biomass, aggregation, and egg production of green abalone as short-term positive effects of MRs.This work validates the relevance of protection and highlights the need for more studies to unravel complex population status, biophysical dynamics, and social-ecological attributes of this fishery.www.nature.com/scientificreports/Regarding the MRs, the cooperative established Plancha MR 1 year before our study (July 1st, 2019), followed by the establishment of Gaviota MR in the middle of our study (July 1st, 2020).Our results reveal an average density three times higher in both MRs than in fished sites.During the first monitoring campaign, Plancha MR exhibited the highest mean (± SE) density (0.7 ± 0.1 ind•m 2 ) and biomass (26.13 ± 3.46 kg) per transect in the survey, with a small decrease in 2021 (0.62 ± 0.06 ind•m 2 ; 19.01 ± 0.40 kg).The closure of the Gaviota MR caused a 50% increase in mean density (before = 0.44 ± 0.02; after = 0.66 ± 0.26 ind•m 2 ), which resulted in an 88.67% increase in mean biomass (before = 9.29 ± 0.39; after = 16.97 ± 6.14 kg).However, it is essential to exercise caution in attributing these observed effects solely to MR protection.The response may be influenced by the spatial heterogeneity of the benthic environment and other confounding factors 37 .Additionally, the limited number of transects within the MRs could result in wider standard errors.Nonetheless, this does not imply that MRs are ineffective, but that the positive effects observed in our study with green abalone, and other studies with fishes 11,38 , echinoderms 39,40 , and gastropods 5 , including abalone 4,36,41,42 should be interpreted with caution.
While our primary objective aimed to cover multiple locations across Guadalupe Island to better characterize the green abalone population, we acknowledge that the unequal distribution in sampling efforts could have implications.The uneven number of transects among different protection areas can lead to variations in the data.This variability could be a result of both natural differences in the population and the limited sample size.Also, the standard error may appear abnormally low in cases where the sample size is small.Lastly, the disparities in sampling effort might affect the representativeness of our observations.Some areas with fewer transects may not fully capture the diversity and dynamics of the green abalone population.
Our data suggests that abalone aggregations in MRs were 4-6 times higher than in fished sites (2020, Plancha MR = 41.00 ± 5.00; fished = 10.85 ± 1.20 ind; 2021, Plancha + Gaviota MR = 37.14 ± 10.86; fished = 6.29 ± 1.36 individuals per transect).Given abalone are broadcast spawners, bigger and more aggregated individuals can enhance fertilization rates potentially resulting in a positive effect on the population.This has been reported for green abalone by Parnell et al. 40 as long term effects of Californian MRs.Yet, to our knowledge, this is the first report in the region concerning short-term changes in green abalone aggregations after only 2 years of fishing closures.
Short-term effects of MRs on abalone size have been previously documented 30 .However, abalone are known for their slow growth, which means that observing a noticeable change in shell length within 1-2 years is generally considered a short timeframe.Hence, one potential effect of the MRs is the increase in the number of individuals that exceed the minimum harvest size, allowing them to "escape" the fishery (Fig. 7).This can lead to a higher proportion of large abalone inside the MR, as observed in this study (2020, Plancha MR = 67.36 ± 3.57; fished = 13.70 ± 1.90%; 2021, Plancha + Gaviota MR = 14.70 ± 3.71; fished = 2.94 ± 0.64%).The increases in abalone size and aggregations inside reserves combined with a higher biomass could enhance egg production.
On average, both MRs had 4.2 times more biomass per transect than the fished areas.Plancha MR in 2020 had 5.7 times (inside = 26.13± 4.90; outside = 4.55 ± 0.39 kg), and both MRs 4.4 times in 2021 (inside = 17.55 ± 4.80; outside = 3.96 ± 0.47 kg).Our results agree with the previously documented Californian MRs, implying that there are stronger responses to biomass metrics than density [43][44][45] .Also, as suggested by Lester et al. 5 , biomass responds faster than abundance to reserve protection, on average 4-5 times, while densities are only 2-3 times, as observed in this study.Thus, the combination of larger and more abundant animals should result in a higher egg production 46 .www.nature.com/scientificreports/ The establishment of MRs at Guadalupe Island suggests an average egg production 4.1 times higher than the fished sites (inside = 683,662 ± 43,583; outside = 166,522 ± 92,828 eggs•m 2 ).This result agrees with 36 , who reported higher potential egg production in Californian MRs than in fished sites for abalone.Also, abalone are more likely to be found in higher aggregations at protected areas relative to exploited sites 40 , as documented on this study.Additionally, the combined responses of abalone densities and size structure to protection can increase genetic diversity and reproductive output and sustain recruitment in areas surrounding the MRs through a spillover effect 4,35,47 .For example, at Natividad Island, Baja California Sur, green abalone has a larval dispersal estimation of ⁓ 300 m 4 .This means that sites on the southern tip of Guadalupe Island could receive spillover from the reserves and need to be considered in future research.Nonetheless, due to the short-term effects of this study and the cryptic behavior of juvenile abalone, the probability of observing this effect is low.It is also important to consider that fishing mortality directly impacts the reproductive potential, and a MR is one of the few management scenarios that enhances resilience 48 .
It has been documented that spatial depletion occurs in abalone fisheries, with the fishing grounds closest to the port being depleted first, removing virtually all the available stock in those areas 49,50 .In the first monitoring campaign in 2020, we observed similar effects as location 1, the closest to port, presented the lowest mean density (0.08 ± 0.02 ind•m 2 ) and biomass (2.16 ± 0.66 kg) per transect.The following effects were documented at location 4, at the island wave-protected area (Fig. 1).At these sites, fishing pressure is higher throughout the season than in the exposed part of the island.We acknowledge the importance of considering variations of exploited areas in our study, but the primary objective was to assess the population instead of the fishery.
Interestingly, we found higher abalone densities in Guadalupe Island than in most of the Northeastern Pacific, except for Van Damme State Park, California (Appendix Table S1).Such high population densities warrant studying Guadalupe Island to enhance our understanding of abalone population dynamics in remote regions.It is also imperative to consider that some sites at Guadalupe Island are at the density limit where recruitment failure is expected (i.e., 0.2-0.3ind•m 2 ) 33 .Thus, it is of utmost importance to continue the establishment of MRs to avoid the Allee effect on the local green abalone population 26 .Still, it is important to mention that both monitoring campaigns were at the end of the fishing season (Mexican fishery zone 1 in 2021: November 30th to June 30th), and local ecological knowledge suggests that the abalone densities are around 2-3 times higher at the start of the season.These hypotheses could be related to: (1) by not fishing below 20 m, the deeper abalone can potentially supply the shallower areas with larvae and adults; (2) cryptic abalone that escape the previous year's fishery; and (3) rotational fishing management that the cooperative has used for years.
The Guadalupe Island fishing cooperative is a good example of how community management might work to increase the resilience and sustainability of local marine resources.The values of the fishing cooperative extend beyond the mere administration of marine reserves.It embodies a collaborative approach where local stakeholders actively participate in decision-making processes.This knowledge, passed down through generations, informs resource management strategies that balance conservation goals with the needs of the fishing community.By emphasizing the cooperative's role and social-ecological contributions, it becomes evident that their involvement is instrumental in achieving sustainable fisheries management and preserving species like the green abalone.
Given the results from the present study, the life history characteristics of abalone, the social-ecological system, and the unique bio-physical, oceanographic, and geographical conditions at Guadalupe Island a set of no-take zones could create a persistent network for abalone conservation 51 .The size and spacing of that network depends on the settler-recruit relationship of a particular species, adult movement, and longshore currents 52 .Abalone larvae settlement can occur within a short (< 50 m), long (> 100 m), or short and long (dual mode) distance from their parents 53 .The green abalone population at Guadalupe Island belongs to a subspecies (H.fulgens guadalupensis) 54,55 , suggesting that local recruitment is predominant.Therefore, present, and future MRs will likely be critical to sustaining the local abalone populations and fishery yields.Particularly considering the broader context of abalone conservation throughout the rest of the country under uncertain future climate conditions as the frequency and length of marine heatwaves and other climate impacts increase [56][57][58] .
Several studies document the positive effects that MRs have on abalone fishery yields 15,59,60 .Nevertheless, the placement of reserves is a critical aspect.For example, placing reserve edges in a continuous habitat may enhance spillover and thus benefit fisheries 60 .Also, if the no-take zone is too big, sessile or sedentary species rarely move out of reserves; hence they are rarely captured and provide only the benefit of larval transport 61 .Thus, to increase fishery yields, the size of the reserves is important, as the largest yields are obtained with small reserves of around 100 m wide so that the export of larvae and spillover of adults are maximized 41,59 .Finally, the goals of MRs could be enhanced if located in source 48 or sink areas 62 .Nonetheless, payoffs of stand-alone marine reserves rarely compete with more traditional optimal management schemes 63 , such as the one applied by the fishing cooperative in this study.Still, they can be beneficial when stocks are heavily exploited 64 .Thus, the combination of MRs with good management could maintain a sustainable fishery at Guadalupe Island.
Continuing monitoring of the social-ecological system is critical to provide guidelines preventing abalone and other fished species from reaching a critical threshold beyond which recovery is virtually impossible 65 .However, we need more information to adequately evaluate the effects of establishing MRs, for example, utilizing Beforeafter-control-impact with a rigorous assessment of the benthic habitat 66 .Also, understanding the biological attributes of the focal species, such as ontogenetic movement with telemetry 67 , dispersal with genetic studies 53 , the population growth rate with long-term monitoring 68 , and the size of reserves relative to the home range 69 .Lastly, fisheries attributes such as the status of the fishery before implementation of reserves; fishers' behavior, and fleet dynamics before and after reserve implementations; and effective leadership and governance, including capacity for monitoring and enforcement 70 .
We acknowledge that our results cannot be fully attributed only to the MRs.For example, our observations spanned only 2 years, with the initial monitoring campaign in 2020 being essentially a single-point-in-time comparison between sites.Additionally, the selection of MR locations was not random; rather, it was based on www.nature.com/scientificreports/local ecological knowledge of the cooperative 32 .These chosen locations seem favorable to green abalone biological and demographic parameters, which could be linked to specific oceanographic and habitat conditions 37 .While this non-random selection process can introduce potential biases in our findings, they are relevant factors to be considered for the success of MRs worldwide 71,72 .Furthermore, we did not provide a detailed description of the heterogeneity of the benthic habitat.This factor can significantly contribute to the observed variations in abalone parameters across different locations 73 .Nonetheless, transects were deployed only on rocky reefs where abalone fishery occurs.As highlighted by Miller et al. 74 , if MRs and control (fished) sites are initially dissimilar in habitat, larval supply, or historical conditions, comparing them can be confounded by inherent site differences.Finally, MRs seem to be a great tool to improve abalone fishery management, but we also need to think about other technologies and adaptation strategies to cope with climate change.For example, we could boost the abalone densities and hence fertilization inside MRs with translocation efforts 75 .Another option is the development of conservation aquaculture, as sustainable mariculture systems 76 to complement the fishery and enhance wild populations through seed restocking [77][78][79] .To end, there is no single panacea but a combination of co-management approaches and the development of sustainable technologies to recover the affected wild abalone populations worldwide.

Figure 1 .
Figure 1.Geographic location of Guadalupe Island, Baja California, Mexico, and surveyed sites.Black dots represent monitoring sites.Small dots were monitored only in 2020, and wide dots were monitored in 2020 and 2021.Numbers indicate grouped site locations based on the cooperative official fishing polygons: Fished (locations 1-8) and marine reserves (location 9).The Inset map indicates marine reserve "Plancha" (A) and marine reserve "Gaviota" (B).P indicates port location.The map was created with ArcGis V. 10.7.1.

Figure 2 .
Figure 2. Timeline of the study development at Guadalupe Island.The abalone fishery is open from February 1st to June 30th, then has a complete closure from July 1st to January 31st (Mexican fishing zone 1 in 2021).

Figure 6 .
Figure 6.Mean green abalone density (individuals•m 2 ) in fished sites and marine reserves (MR) in 2020 (green) and 2021 (orange).Gaviota was open to the fishery in 2020 and a MR in 2021.Plancha was closed to fishery in 2020 and 2021.N = monitoring transects of 60 m 2 .Fished sites were different than the MR in 2020 (χ 2 = 10.54,df = 2, P = 0.0051), and 2021 (χ 2 = 17.26, df = 2, P = 0.0002).There were no differences inside fishing locations or MRs across the years.Vertical lines denote ± SE.

Figure 7 .
Figure 7. Overlapping size (mm) structure and density (individuals•m 2 ) of green abalone in marine reserve (orange) and fished (green) sites for the 2020 and 2021 monitoring campaigns.N = abalone observations.In the 2020 monitoring campaign, we deployed 94 transects in fished sites and 2 in the marine reserve Plancha.In 2021, we deployed 85 transects in fished sites and 7 in Gaviota and Plancha marine reserves.The black line indicates the minimum harvest size (150 mm). https://doi.org/10.1038/s41598-023-50316-9

Table 2 .
Mann-Whitney tests of protection (fished vs. reserve) effect on green abalone size, proportion of abalone larger than 150 mm, total biomass, density, aggregations, and egg production.Data for the 2021 monitoring campaign: N = 92 transects.N in MRs location (9) = 7; N in fished locations (1-8) = 85 for all biological parameters.df degrees of freedom, ind individuals.150 mm is the minimum harvest size for green abalone.Aggregation is the mean number of individuals aggregated by transects.* indicates statistical significance.

Table 3 .
Mann-Whitney tests of the site Gaviota before closure to the fishery in the 2020 monitoring campaign (N = 3) versus Gaviota 1 year after the marine reserve was established in the 2021 campaign (N = 5).df degrees of freedom, ind individuals.150 mm is the minimum harvest size for green abalone.Aggregation is the mean number of individuals aggregated by transects.