https://orcid.org/0000-0002-1588-3470
Figures
Abstract
Humans have transformed ecosystems through habitat modification, harvesting, species introduction, and climate change. Changes in species distribution and composition are often thought to induce biotic homogenization, defined as an increase in the spatial similarity of species compositions through time. However, it is unclear whether homogenization is common in ocean ecosystems and if changes in similarity exhibit linear or more complex dynamics. Here, we assessed patterns of homogenization or its converse (differentiation) across more than 175,000 samples of 2,006 demersal fish species from 34 regions spanning six decades and 20% of the planet’s continental shelf area. While ten regions (29%) recorded significant homogenization, eleven (32%) recorded significant differentiation. Non-monotonic temporal fluctuations in species composition occurred in 15 regions, highlighting complex dynamics missed by before-and-after snapshots that can drive spurious conclusions about trends in similarity. Fishing pressure and temperature helped explain variance in similarity across years and regions. However, the strength and direction of these effects differed by region. Here we showed that, despite intense anthropogenic impacts on the oceans, the majority of demersal marine fish communities do not follow the global homogenization paradigm common in other realms.
Citation: Kitchel ZJ, Maureaud AA, Fredston A, Shackell N, Mérigot B, Thorson JT, et al. (2025) Marine communities do not follow the paradigm of increasing similarity through time. PLOS Clim 4(7): e0000659. https://doi.org/10.1371/journal.pclm.0000659
Editor: Kevin Tougeron, Universite Catholique de Louvain, BELGIUM
Received: December 18, 2024; Accepted: May 28, 2025; Published: July 9, 2025
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: The publicly available data used for this manuscript are from: Maureaud, A.A., Palacios-Abrantes, J., Kitchel, Z. et al. FISHGLOB_data: an integrated dataset of fish biodiversity sampled with scientific bottom-trawl surveys. Sci Data 11, 24 (2024). https://doi.org/10.1038/s41597-023-02866-w. The dataset is available for download from the GitHub repository (https://github.com/AquaAuma/FishGlob_data/tree/main/outputs/Compiled_data; https://doi.org/10.5281/zenodo.15694178) and a release available for download on Zenodo (https://zenodo.org/records/10218308).
Funding: This project was part of the FISHGLOB working group, ‘Fish biodiversity under global change: a worldwide assessment from scientific trawl surveys’, co-funded by the Centre for the Synthesis and Analysis of Biodiversity (CESAB) of the French Foundation for Research on Biodiversity (FRB), the Canadian Institute of Ecology and Evolution (CIEE), and the French embassy in Canada (to MLDP & BM). Other support was provided by U.S. National Science Foundation grants #DEB-1616821 (to ZJK & MLP), #OISE-1743711 (to ZJK & MLP), and #CBET-2137701 (MLP); the Rutgers School of Environmental and Biological Sciences (to ZJK); the Rutgers Ecology and Evolution Graduate Program (to ZJK); a Rutgers Coastal Climate Risk and Resilience Traineeship (to ZJK); the Coronavirus Response and Relief Supplemental Appropriations Act, 2021 (CRRSAA)/Higher Education Emergency Relief Fund II (HEERF II) (to ZJK); and The Natural Sciences and Engineering Research Council of Canada (NSERC) and The Social Sciences and Humanities Research Council (SSHRC) Partnership Grant (to JPA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
As ecosystems face unprecedented changes driven by human activities, communities of organisms are reorganizing across space and time [1–4]. Many studies report that communities are experiencing an increase in spatial similarity through a process termed biotic homogenization, defined more technically as a reduction in beta diversity among species assemblages across space [5–7]. Similarity increases with an increase in the proportion of shared species among assemblages, thereby causing loss of uniqueness of individual communities [5,6]. A trend in spatial beta diversity towards either homogenization or its opposite, differentiation, can transform overall ecosystem function, but homogenization can also lead to instability due to heightened synchrony among communities [7,8]. Biotic homogenization is often considered to be a widespread phenomenon [9], but most research to date has focused on terrestrial and freshwater realms [6,10–19]. We currently lack understanding for whether biotic homogenization is common across a wide range of marine ecosystems [20].
In addition, homogenization has often been characterized by comparing only two sampling events [6,21,22], limiting our ability to understand temporal dynamics and possibly leading to spurious conclusions [23,24]. Marine continental shelf ecosystems have consistent, long-term, and spatially extensive scientific monitoring programs [25], and therefore provide a unique opportunity to reveal the temporal dynamics of homogenization. Research on marine ecosystems has found evidence of biotic homogenization in a handful of cases [26–32], most often in highly modified nearshore zones such as estuaries, wetlands, and coral reefs (but see Ellingsen et al. 2015, 2020 & Magurran et al. 2015). Because marine biomes exhibit faster rates of species redistribution than terrestrial and freshwater biomes [33], we may expect that homogenization is occurring more rapidly in the ocean despite the small number of documented cases.
The mechanisms leading to homogenization in the ocean may differ from those acting in terrestrial and freshwater systems. Biotic homogenization has often been linked to the spread of invasive species across space [34], but in marine communities, high impact invasions are uncommon by comparison [35,36]; but see [37–40]). In addition, habitat heterogeneity frequently shapes spatial beta diversity by providing niche opportunities and a variety of resources to support different species assemblages [41–43]. While temperature gradients are steeper in terrestrial environments, nutrient and light availability vary dramatically more in marine environments [44,45]. Landscape homogenization (e.g., conversion of natural landscapes to farmland or the loss of structured or biogenic habitat) can also lead to biotic homogenization [42,46–50]. In the ocean, however, human impacts on many seascapes lag substantially behind impacts on terrestrial ecosystems and has been substantially less extensive to date [51].
In marine ecosystems, changes in species composition have been triggered by changes in temperature, fishing, and other factors [42,51–54]. Both press (i.e., warming or eutrophication) and pulse disturbances (i.e., a heat wave or oil spill) have led to biotic homogenization within marine communities [55–60]. Marine ectotherms are highly sensitive to water temperature due to metabolic constraints and their relative thermal specialization compared to terrestrial and freshwater species [61,62]. Declining spatial heterogeneity in temperature, therefore, may drive biotic homogenization as opportunities for niche differentiation and coexistence decline [28,63,64]. Alternatively, expansions at the leading (e.g., poleward) range edge that are faster than trailing-edge contractions in response to rising temperatures [65] could escalate species overlap and therefore homogenization. Homogenization may also result from fishing in regions where fisheries target endemic species with small range sizes [66,67]. In contrast, fishing may induce differentiation in regions where fisheries target mobile, large-bodied consumers due to the release of mesopredators that often exhibit less stable population dynamics [26,27,68]. Whether changes in temperature and fishing consistently impact spatial beta diversity in the ocean remains unclear.
Here, we used an extensive dataset of scientific bottom trawl surveys to assess the prevalence and dynamics of biotic homogenization across the coastal ocean of four continents (Tables S1-S2 in S1 File). Our primary hypotheses were that 1) marine bottom fish communities would show high prevalence of biotic homogenization because species range shifts are widespread and rapid, 2) time series of biotic homogenization would reveal complex and non-linear temporal dynamics not apparent from comparisons of two time points, 3) changes in biotic homogenization and differentiation in the ocean would be related to changes in spatial temperature heterogeneity because temperature strongly affects marine community structure, and 4) fishing would affect changes in biotic homogenization, though the effects would differ across regions because fisheries target species with different geographic range sizes and trophic levels in different regions (Fig A in S1 File). An alternative Hypothesis #3 was that a metric of average or extreme annual temperature conditions, rather than spatial temperature heterogeneity, would be related to homogenization, since marine range shifts can drive homogenization. The surveys included observations of 2,006 marine fish species across 178,531 independent samples from 34 time-series in tropical, temperate, and subpolar regions in the Americas, Europe, Africa, and Oceania. Surveys had one to six decades of consistent sampling and spatial extents from 16,000–670,000 km2. The sampling in total covered 20% of the area of shelf ecosystem habitat worldwide (Supporting Text). We tested for homogenization and differentiation by calculating annual spatial dissimilarity in each survey using beta diversity indices and estimated the effects of fishing and temperature on dissimilarity; lower beta diversity (lower dissimilarity) indicated a more homogenized community across space. We found that homogenization and differentiation were similarly common in bottom fish communities worldwide, but that most regions were not experiencing a directional change in spatial beta diversity.
Methods
Spatial beta diversity calculations
We used long-term scientific bottom trawl survey data for marine fishes recently compiled and integrated as FISHGLOB [25,69]. These are fisheries-independent surveys with standardized statistical designs. We omitted surveys that only reported metadata and those that had inconsistent sampling methods and locations through time. Surveys were only included if they had at least 10 years of sampling to facilitate detection of long term trends [70]. Seven regions were surveyed in multiple seasons and, to avoid pseudoreplication, we only included the season with the highest number of tows (Tables S1-S2 in S1 File). In total, 34 regions were included in this analysis (Fig 1a & Table S1 in S1 File). All surveys were limited to the three most-sampled months–representative of a single season–except in the case of the West Coast United States survey for which we retained four months because of consistent sampling across those months.
(a) Map of temporal trends in spatial Jaccard dissimilarity by region. (b-d) Changes in spatial community composition over the study period are visualized using non-metric multidimensional scaling (NMDS) for example regions that experienced differentiation (b; West Coast South Island, New Zealand; first = 79 tow locations, last = 65 tow locations), homogenization (c; Southeast United States; first = 77 tow locations, last = 87 tow locations), and no significant trend (d; Iceland; first = 528 tow locations, last = 529 tow locations). Example regions are labeled in (a). Each point in subfigures (b), (c), and (d) represents the species composition of an individual sampling event, with points outlined in white for the first survey year and in black for the last survey year. The NMDS ordination arranges sampling events based on their species composition and Jaccard dissimilarity, such that points closer together represent more similar communities. To aid visualization of community shifts, ellipses enclose 95% of a multivariate t-distribution fitted to the first-year (dotted line) and last-year (solid line) samples. A contraction of the ellipse suggests homogenization, while an expansion suggests differentiation. Basemap from Natural Earth [71], and map rendered using the sf and ggplot2 packages in R [72–74].
Each sample was a single tow, i.e., a drag of a bottom trawl net along the sea bottom. In the case where multiple samples occurred at the same latitude and longitude on the same day, we averaged abundance observations for all species and considered this to be a single sampling event. Additionally, we excluded low quality tows that did not match the standard area swept or duration for a given survey, which occasionally occur due to mechanical issues, oceanographic conditions, or other logistical constraints. We eliminated years, samples, and taxa unsuitable for temporal and spatial biodiversity analysis using author expertise and previous publications on survey data (see Supporting Text).
Because we were interested in temporal trends in dissimilarity, it was important to have a consistent spatial extent over time for each survey. To establish a standardized spatial extent in each region through time, we assigned each sample to a 7,774.2 km2 hexagonal spatial cell, (except for the Norwegian survey of the Barents Sea for which we used a cell size of 23,322.2 km2 due to low sampling density [25]). For each survey, we excluded years in which the survey sampled fewer than 70% of the hexagonal cells ever sampled by that survey (Fig Ba-b in S1 File, Table C in S1 File). Next, we excluded cells that were sampled in fewer than 70% of the remaining years (Fig Bc in S1 File, Table C in S1 File). Finally, we excluded regional surveys for which this standardization process excluded over 50% of the samples across the full time period (Table C in S1 File). This spatial extent standardization procedure resulted in 178,531 unique samples (tow events) across 34 regions between 1968 and 2021 (Table B-C in S1 File). We used tows as the basis for further spatial beta diversity analyses.
Each tow included species observations recorded as number of individuals or biomass, depending on the survey. We used both abundance and biomass to determine each species occurrence (presence or absence). A small number of tows (0.2%) did not include either abundance (count) or biomass (kilograms) values and were therefore removed. Cleaning and standardizing the data led some samples to have biomass or abundance values of zero for all observations. These tows were excluded from the analyses (0.03% of all tows) because dissimilarity calculations on communities with zeros across all observations are often meaningless [75].
All taxonomic names were standardized using WoRMS [76,77]. Only observations identified at the species-level were included in the analyses, leaving a total of 2,006 unique species. We performed two sensitivity tests to assess the impact of the inclusion of uncommon species and/or low abundance species on the results, as species identification quality can be lower for infrequently encountered species. First, we examined patterns in dissimilarity while excluding the bottom 15% species when ranked by abundance or biomass in each region (leaving 1,861 unique species). Second, we repeated analyses while excluding any species present in less than one-third of the years that a survey occurred (leaving 1,429 unique species).
We calculated beta diversity in each year of each region as the average pairwise dissimilarity between samples (tows) using Jaccard dissimilarity based on species occurrences (Fig A in S1 File). Therefore, the unit of analysis was the average dissimilarity value across all pairs of tows in a given region in a given year. The sample size varied between 10 years for Queen Charlotte Sound and the Rockall Plateau up to 52 years for the Northeast United States for a total of 705 samples (unique survey and year combinations). Average pairwise dissimilarity is widely used for quantifying regional heterogeneity in community composition, has the intuitive interpretation as the expected dissimilarity of a randomly selected pair of samples, and is not sensitive to differences in sample size [78–87]. Average pairwise metrics, however, do not account for patterns of co-occurrence across more than two sites [88]. Additionally, this approach does not consider how community composition varies with the geographic distance between sites [89].
To also consider species dominance, we repeated analyses using abundance-based Jaccard dissimilarity based on species abundances for the 24 of 34 surveys with species count data available [90]. We also tested using relative species abundances to remove the influence of differences in total abundance across space. Jaccard dissimilarity is highly influenced by the degree to which species are shared across sites, and therefore to differences in both richness and species turnover [91]. Jaccard has been widely used to measure community dissimilarity in community ecology, and is robust to geographic and taxonomic undersampling [92,93]. We measured dissimilarity using the `vegdist` function from the vegan package in R [75].
Testing for directional temporal trends in beta diversity
To test for an average trend in dissimilarity over time across all surveys (Hypothesis #1), we fit a linear mixed effect model using the lme4 package in R with a random slope and intercept for each survey to help account for differences in methodology across regions and repeated observations from each survey [94]. We also fit a linear model with a fixed effect interaction between survey and year to examine trends in dissimilarity for each individual survey (Hypothesis #1). Because surveys involved repeated sampling of regions through time, we compared a linear model with and without a temporal autocorrelation term for year (by survey), implemented using the nlme package in R [95].
We classified surveys with a significant negative coefficient (p < 0.05) for year as homogenizing (primary Hypothesis #1) and surveys with a significant positive coefficient as differentiating (alternative Hypothesis #1); surveys for which the coefficient + /- standard error crossed zero were classified as having no significant trend over time (null Hypothesis #1). To illustrate communities of bottom fish that underwent homogenization, differentiation, and no trend in dissimilarity, we constructed non-metric multidimensional scaling (NMDS) plots using the vegan package in R (Oksanen et al. 2022).
To assess the potential for detecting significant trends even if none existed, we compared results to a null model in which we reshuffled average annual dissimilarity values across years within surveys. This approach decoupled year from dissimilarity value and maintained correlations in abundance among species, but did not maintain temporal autocorrelations within species. We repeated this procedure 1000 times and, each time, classified surveys as homogenizing, differentiating, or not based on the same linear model approach used for the observed data. We tallied the number of homogenizing or differentiating surveys from each of the 1000 reshuffled datasets and calculated the 95th percentiles. Additionally, we compared the distribution of beta diversity trends from observed data versus the distribution from reshuffled values.
Testing for non-linear patterns in beta diversity
To illustrate non-monotonic fluctuations in dissimilarity through time in each region, we fit a generalized additive model (GAM) using the mgcv package in R with a smoother per survey [96]. We then tested for non-linear (non-monotonic) fluctuations in dissimilarity through time (Hypothesis #2) by comparing linear models with GAMs for each individual survey using Akaike Information Criteria (AICC; [97,98]) (Fig A in S1 File). We classified as non-linear those regions for which dissimilarity over the study period was better described by a GAM than by a linear model in support of primary Hypothesis #2 (∆AICC > 2). For those better described by a linear model (∆AICC > 2), we classified them as having linear trends (primary Hypothesis #2). Regions for which GAM and linear model approaches performed similarly (|∆AICC| < 2) were not classified as either (null Hypothesis #2).
Testing temperature and fishing as predictors of beta diversity
To test among potential drivers of annual dissimilarity, we built and compared a set of linear models including temperature (Hypothesis #3) and fishing (Hypothesis #4) (Fig A in S1 File). The global model included temperature, fishing, and additional variables related to potential sources of heterogeneity among surveys that were not the main focus of this study. The additional variables included survey identity, primary season of sampling (adjusted for hemisphere), average latitude of the survey, the latitudinal range of the survey, the number of species sampled, the area surveyed, the average tow depth of a survey, the range of tow depths of a survey, and the average number of tows per year within a survey (Table S4 in S1 File). We also included interactions between temperature or fishing and survey so that the relationships could differ by survey (alternative Hypothesis #3, primary Hypothesis #4). Because of repeated sampling through time, we evaluated whether including a temporal autocorrelation term for year, implemented using the nlme package [95], was favored by AICC.
All covariates were calculated as annual values per survey, and we excluded years and surveys missing any covariates (we excluded the Southern Gulf of St. Lawrence and the Rockall Plateau surveys because they were missing depth and fishing data, respectively). In total, we fit these models to 32 surveys from 1982-2019,resulting in a sample size of 628 unique year and survey combinations. We calculated the annual survey area as a concave hull surrounding all tow locations using the concaveman package in R with a concavity of 1 and length threshold of 2 [99]. Covariates were calculated for the specified year of survey sampling, except for temperature (calculated for the 12 months prior to the first observation of a survey-year) and fisheries catch (calculated for the calendar year preceding each survey-year). All numeric covariates were scaled and centered across surveys to improve model convergence, except for fisheries catch, which was scaled within a survey and was therefore representative of relative catch within a region.
Our primary temperature Hypothesis (#3, Fig A in S1 File) focused on the spatial heterogeneity of temperature [28]. However, alternative hypothesis #3 tested whether average, extreme, or the seasonal range of temperatures was a more effective predictor, since these metrics are closely linked to the species range shifts hypothesized to contribute to marine homogenization [28]. Demersal fishes respond to both extreme and average bottom temperature conditions, the range of temperatures experienced in a year, and the heterogeneity of temperature across space [100–103]. We used daily sea bottom temperature from the SODA 3.3.2 data product [104], which is a global historical reconstruction of sea temperature at multiple depths from 1980 to January of 2019 at a 1/4° resolution. As a metric of spatial heterogeneity in temperature within each survey and year (related to our primary Hypothesis #3, Fig A in S1 File), we calculated the annual mean bottom temperature for each sample location and then calculated the standard deviation across sample locations within each survey and year. For the alternative temperature hypotheses, we calculated the mean, minimum, maximum, and seasonality (maximum - minimum) for each sample location for each year. Next, we took the average of these summary statistics for each survey and year. We restricted analyses to annual regional dissimilarities between 1982 (no usable tows in 1981) and 2019 because of the availability of high resolution temperature data.
We expected bottom fish to respond most directly to bottom temperature values, but we calculated the same metrics and repeated the same analyses using the NOAA 1/4° Daily Optimum Interpolation Sea Surface Temperature (OISST) [105]. The SODA and OISST temperature products effectively capture inter-annual and decadal climate regimes (Ren et al. 2023, Mauro Vargas-Hernandez et al. 2014, Giese and Ray 2011, Huang et al. 2016) that commonly impact regional fish population dynamics [106,107].
To explore the impact of resource extraction on mean annual dissimilarity (Hypothesis #4, Fig A in S1 File), we used fisheries catch data as a proxy for fishing pressure. We extracted reconstructed annual fisheries catch in metric tons from Sea Around Us using the Large Marine Ecosystem, Exclusive Economic Zone, or Marine Ecoregion that best overlapped with a survey’s spatial extent [108] (Table S5 in S1 File). Catch values in Sea Around Us had been reconstructed using reported catch (primarily from the Food and Agriculture Organization of the United Nations) and estimates of unreported catch [108]. We included estimated landings and discards of identified marine fish. We used the total reconstructed catch (which included species that do not appear in the bottom trawl data sets) because fishing affects both target and non-target species through changes in biotic interactions and because our goal was to calculate a metric representative of relative annual fishing activity in a region [109,110].
We first compared a set of global models with AICC [97,98,111], each of which included one metric of annual temperature. From these, we selected a single temperature metric. We then used the dredge function in the MuMln package in R to compare with AICC all possible nested models constructed with the selected temperature metric [98,111]. We estimated average covariate coefficients by averaging continuous parameters included in all models with ΔAICC < 4.
Results
Prevalence of biotic homogenization and differentiation
Trends in spatial beta diversity differed substantially across 34 surveys of bottom fish in the coastal ocean (Fig 1a & C in S1 File). Overall,10 surveys (29%) recorded significant regional homogenization (Fig 1a,c), 11 surveys (32%) recorded significant differentiation (Fig 1a,b), and the other 13 surveys (38%) did not record significant trends in regional dissimilarity (Fig 1a,d). We did not find evidence of strong temporal autocorrelation (Δ AICC = 615). The null model suggested that two more regions were homogenizing than would otherwise be expected due to chance, and similarly, two more were differentiating than expected (Fig Da-b in S1 File). Therefore, we did not find evidence of Hypothesis #1, that homogenization patterns would be pervasive.
Trends in beta diversity ranged from a 6% per decade loss of spatial dissimilarity (i.e., homogenization) in the Rockall Plateau (Northeast Atlantic), to a 4% per decade gain (i.e., differentiation) in Greenland (Northwest Atlantic) (Fig 2b). The magnitude of trends in dissimilarity observed were also higher than predicted at the 95% level by the null model (Fig Dc-d in S1 File). Despite these large individual trends within surveys, we found no significant change in overall dissimilarity through time across all survey regions (slope = 0.008% + /- 0.03% SE per decade; p = 0.78; n = 705; linear mixed effects model; Fig 2a).
(a) Annual Jaccard dissimilarity for each region (colored points, n = 705) with generalized additive model (GAM) smoothers for each region (colored lines) and 95% confidence intervals (colored ribbons). A decrease in dissimilarity represents homogenization (yellow); an increase represents differentiation (pink). A lack of significant trend is shown in blue. The average linear trend across surveys (black line with 95% confidence interval in gray) is also plotted from a linear mixed effect model with a random slope and intercept for survey. (b) Coefficients and associated standard error of dissimilarity versus time for each survey from a linear model (LM) with a fixed effect interaction between survey and time in ascending order by coefficient value. Point size represents the length of the survey period. Asterisks mark surveys for which dissimilarity through time was better described by a non-linear GAM than a LM.
Our finding that neither homogenization nor differentiation were widespread across regions was not sensitive to the metric of dissimilarity, although specific survey trends differed across metrics (Fig E in S1 File). For raw abundance-based Jaccard dissimilarity, 8 regions (33%) differentiated and 8 (33%) homogenized out of the 24 regions. Seven regions (29%) exhibited a different trend as compared to the occurrence-based results. In the case of relative abundance-based Jaccard dissimilarity, eight regions (33%) differentiated, seven (29%) homogenized, and nine (38%) regions exhibited a different trend as compared to the occurrence-based results.
Sensitivity tests removing rare or low abundance species further supported the finding that instances of homogenization, differentiation, and a lack of a trend in dissimilarity were similarly common across bottom fish communities (Fig F in S1 File). When the least abundant 15% of species were removed from each survey, there were no changes in trends. However, when species present in fewer than two-thirds of years of a survey were removed, 19 regions (56%) exhibited a different dissimilarity trend as compared to the full dataset; most often a shift from either homogenization or differentiation to no trend.
We tested whether changes in gamma diversity explained the observed trends in beta diversity, because dissimilarity increased with the number of species in a region (Fig G in S1 File). However, we did not find evidence for such a relationship when Rockall Plateau (highly negative trend) was removed (p = 0.26, linear model; Fig H in S1 File, R2 = 0.04). Surveys differed in the first year (baseline) of sampling, the length of the survey period, the time of year of sampling, spatial extent, and sampling density (Table B in S1 File). However, we did not detect a relationship between the observed trend in beta diversity and the baseline year, the length of the survey period, the spatial extent, or sampling density (Fig I in S1 File). Regions with surveys occurring in the later half of the year were more likely to exhibit homogenization, but there was no significant relationship between trend and season (Fig J in S1 File). Surveys varied in sample (tow) density, but, in all regions other than the Southeast US, exhibited consistent density through time and no relationship between density and spatial beta diversity in a given year (Fig K in S1 File).
Non-linearity in beta diversity through time
These long-term surveys revealed substantial multi-annual and decadal variability, such that they varied through time non-monotonically between more homogenized and more differentiated states (Fig 2a & C in S1 File, & Table F in S1 File). For example, a decline in dissimilarity between the mid-1990s and mid-2000s off the coast of Namibia was followed by an increase continuing through the late 2010s. Additionally, the Eastern Bering Sea experienced increases in dissimilarity in the late 1980s and 2000s, followed by declines in the early 1990s and 2010s (Fig 2a & C in S1 File). We found that non-linear GAMs performed better (ΔAICC > 2) than linear models for 15 of 34 surveys (44%), partially supporting Hypothesis #2 that non-monotonic temporal dynamics would be common (Table F in S1 File, Fig C in S1 File). Linear models outperformed GAMs for seven surveys (21%), and models for the remaining twelve surveys (35%) performed similarly.
Temperature and fishing as predictors of beta diversity
We then examined the extent to which temperature and fishing explained variation in annual dissimilarity (Fig A in S1 File). Similar to dissimilarity, temporal trends in temperature and fishing differed across regions (Fig L-M in S1 File). The global model including minimum bottom temperature performed best (ΔAICC = 4.8; n = 705; Table G in S1 File), rejecting our primary Hypothesis #3 that spatial temperature heterogeneity would be most important. We carried minimum temperature forward in subsequent model comparisons.
The set of most parsimonious global models (ΔAICC < 2) included minimum temperature, fisheries catch, and survey characteristics (Table H in S1 File & Fig 3). All of the high-performing models included an interaction between survey and both fishing and temperature, suggesting that the response of dissimilarity to temperature and fishing differed by region, supporting alternative Hypothesis #3 and primary Hypothesis #4 (Table H in S1 File, Fig N-O in S1 File). The magnitude and direction of the relationship between temperature and dissimilarity varied by region. The strongest relationship was detected in the Aleutian Islands, with a decrease of 2.3 units of dissimilarity per standard deviation increase in temperature (±1.3 SE). In total, ten regions showed significant negative trends, while five regions exhibited significant positive trends (Fig 3). Similarly, the relationship between relative fishing catch and dissimilarity varied across regions. However, the magnitudes were more similar across regions, ranging from -0.02 units of dissimilarity per standard deviation increase in fishing in South Georgia (±0.007 SE) to 0.02 in the Aleutian Islands (±0.01 SE). Nine regions exhibited significant negative trends, while six exhibited significant positive trends (Fig 3). The four best performing models explained 95% of the variance in annual spatial beta diversity (Table H in S1 File). Removing temperature and removing fishing catch covariates had little effect on the amount of variance explained (93% and 94%, respectively).
= 32 surveys). Coefficients were allowed to vary by region for temperature (a) and relative fishing catch (b), but not for other characteristics (c). All variables were centered and scaled across all observations except for fishing catch, which was centered and scaled within each region. Therefore, coefficients are in units of dissimilarity over units of standard deviation of the covariate. Coefficients for which the standard error did not cross zero are in black and others in gray. Season is not displayed as it was included in the model as a factor.
Models including surface instead of bottom temperature performed similarly and led to similar conclusions, although maximum temperature out-performed other temperature metrics (ΔAICC = 3.4) and was therefore carried forward in subsequent analyses (Table I-J in S1 File & Fig P in S1 File). For all temperature metrics, models without a temporal AR term performed better than those with this term (Table G & I in S1 File).
Discussion
While biotic homogenization is a common expectation and finding in terrestrial and freshwater ecosystems [11,16,112,113], we found that marine fish communities are not consistently homogenizing through time despite rapid and extensive species range shifts. Instead, we revealed complex multi-annual fluctuations in the heterogeneity of community composition through time. The high temporal resolution of scientific surveys on continental shelves around the world also allowed us to detect region-specific effects of both fishing and temperature on the biotic homogeneity on some, but not all demersal fish communities.
Prevalence and temporal dynamics of homogenization
Demersal fish communities were more likely to exhibit no trend in spatial beta diversity than either homogenization or differentiation, reflective of other recent synthesis work [114]. While some regions, such as the Southeast United States and the Barents Sea have homogenized in the past two to four decades, others, such as Greenland and the Scotian Shelf have differentiated. Homogenization and differentiation of individual regions has been previously described (Ellingsen et al. 2020, Siwertsson et al. 2024, Ellingsen et al. 2015), but the relative prevalence of these patterns across continents has not previously been apparent. Marked regional differences in trends of spatial beta diversity over time—ranging from sharp declines to rapid increases in heterogeneity—highlight the critical role of context during periods of rapid environmental change [2,115,116]. The specific species composition, regional environmental conditions, and legacy of human impact shape both the types of disturbances a system encounters and how it responds. These findings illustrate the importance of comparing trends across diverse ecosystems to comprehensively assess global change, rather than focusing solely on those demonstrating dramatic change in community structure.
Despite widespread expectations that communities are consistently homogenizing [11,16,112,113], we found that change within marine fish communities is a highly dynamic process, regularly fluctuating between periods of higher and lower dissimilarity. Some regions exhibited distinct periods of more homogenized and more differentiated community composition, a phenomenon also observed in communities of freshwater diatoms [117], and plants across biomes [118–121]. For example, the Eastern Bering Sea and Sub-Antarctic New Zealand did not experience significant directional change in beta diversity, and yet these regions experienced swings of 10–12% in dissimilarity within just a decade. This variation across years is dramatic compared to what is known from other ecosystems. For example, the variation we observed across decades was two to three times greater than the homogenization observed among plants (3%) and birds (4%) across the centuries since human settlement, though we caution that differences in sampling, scale, and metrics also affect these comparisons [122,123].
The high temporal variability of spatial beta diversity also highlights the importance of the baseline effect in shaping observed trends, wherein the first observation has a strong influence on observed patterns (Navarrete et al. 2010, Edwards et al. 2010, Werner et al. 2020). The baseline effect is particularly strong when only two time points are available for assessing trends, which is how most homogenization trends have been detected to date [124,125]. Two time points are also unable to detect more complex dynamics. The relatively long time series (>10 years) with high temporal resolution (sampling every one to three years) examined here helps to minimize the impact of the starting year (Navarrete et al. 2010, Edwards et al. 2010). While we did not observe any obvious relationships between beta diversity trend and baseline year of sampling, the starting year may still have an impact on the trend we detect, especially in the case of shorter time series. For instance, the Aleutian Islands (North Pacific) exhibited no overall trend between 1983 and 2018. However, a time series beginning in 2000 would have suggested a strong pattern of homogenization supported by 19 years of observations. Understanding temporal variability of spatial beta diversity will be important for evaluating whether baseline effects may be biasing conclusions about homogenization trends in other ecosystems.
Predictors of spatial beta diversity
A large sample size and diverse regions allowed us to test mechanisms that may drive patterns of homogenization and differentiation [42]. Temperature and fishing are known to strongly influence marine species population dynamics, community composition, and geographic distributions [42,51–54], and here, we found that these factors are strongly associated with the temporal dynamics of biotic homogenization and differentiation in a subset of regions.
After testing a range of temperature metrics, we found that minimum annual temperature rather than temperature heterogeneity was the best predictor of annual dissimilarity. This finding matches recent work demonstrating that climate extremes shape species distributions more than average conditions [126]. Minimum temperature in a year can directly (i.e., thermal tolerance; [127]) or indirectly (i.e., predation; [128]) impact species range shifts. In the ocean, cold temperatures have long been known to act as a control on species distributions [129]. The lower bound of temperature extremes are currently increasing at a faster rate than the upper bounds [130], and we therefore anticipate a parallel change in the heterogeneity of bottom fish communities across space.
The influence of temperature on homogenization, however, was strongly context-dependent. For example, off of the west coast of the South Island of New Zealand, warmer years were more homogenized, while in South Georgia (South Atlantic), warmer years were more differentiated. One explanation may be that homogenization is more likely to occur at ecotones experiencing warming. Levels of homogenization are highest when species’ establishments are common, regardless of whether or not they are paired with local extirpations of endemic species [131]. The introduction of warmer-water associated species with high dispersal capacity may initially lead to an increase of uniqueness across space as novel species accumulate poleward of the ecotone, then later drive a decline in uniqueness as those species spread more widely across the region. This ecotone-related phenomenon may explain homogenization in two northwest Atlantic regions–the Northeast and Southeast United States [132–134] These regions sit poleward of biogeographic breaks at Cape Hatteras and Cape Canaveral, respectively.
Similar to previous studies [135,136], we found that fishing also shaped fish community composition. As we hypothesized, this relationship between dissimilarity and fishing varied across region. In the Southeast United States and Newfoundland (North Atlantic), highly fished years were followed by differentiation, while highly fished years in Greenland and the North Sea were instead followed by homogenization. This regional variation suggests that the effect of fishing is more highly context dependent than previously appreciated, and possibly shaped by the distribution of fishing across space, food web structure, and the trophic level of species targeted in the system [52,137,138]. When fishing primarily targets dominant widespread predators, an increase in harvesting is often matched with a decrease in the spatial similarity of species compositions through time [26,27]. In previous studies, declining populations and shrinking ranges of top groundfish predators has led to differentiation through a release of mesopredators with more heterogeneous distributions [27,68,139,140]; this process may also be at play in the demersal communities we studied. In contrast, homogenization may ensue if fishing in a region targets relatively rare species such as sharks and rays [141].
While long in the context of ecological research, the study periods included here only represent observations from the last ~60 years. Therefore, we could not detect if the current state of demersal fish communities are homogenized or differentiated in comparison to communities before the realized impacts of anthropogenic climate change and resource extraction in the ocean. Marine resource extraction in some of these regions has occurred for thousands of years [142–144], but technological advancements and a rise in global food demand contributed to a rapid industrialization of commercial fishing following World War II [145–147]. This period of overexploitation was followed by a decline in global catch of many high value demersal fish stocks beginning in the 1970s in some regions due to increased management intensity in an attempt to compensate for overfishing [148]. Although we used reconstructed catch as a proxy to represent trends in fishing over time, we encourage future researchers to adopt a species- or region-specific approach to capture the complex history of fishing more accurately. Finally, investigation into the mechanisms and pathways leading to changes in dissimilarity, including lagged and indirect responses to oceanographic conditions and resource extraction, would improve our understanding of how communities respond to multiple stressors [149].
The drivers of homogenization and differentiation in ecosystems across realms have often been expressed as directional [28,34,112], but our results emphasize that they need not be so [27,150]. Instead, the degree of homogeneity may more commonly fluctuate through time and be related to environmental conditions (e.g., climatic oscillations and change) and direct human impacts (e.g., fishing). The variability in marine fish community composition across years further highlights the importance of long term observations at high temporal resolution that allow us to disentangle cyclic variability from long term directional trends (Hughes et al. 2017).
Considerations of metric and scale
Our main finding that homogenization and differentiation are relatively uncommon was robust to beta diversity metric. However, both the way in which communities are defined (species presence/absence vs. raw abundance vs. relative abundance), and the metric used to calculate dissimilarity led to differences in how some regions were classified. Including observations of species abundances in calculations can be more representative of the population dynamics underlying changes in community composition [131]. As one example, while occurrence observations led Northern Ireland to be classified as differentiating, abundance observations led the region to be classified as exhibiting no trend. This suggests that changes in community composition are muted in this region when species are weighed by abundance, and therefore, changes are stemming from relatively rare species. As another example, in the Bay of Biscay (North Atlantic), occurrence-based analyses classified the region as homogenizing, while abundance-based analyses classified the region as differentiating, suggesting that while changes in the distribution of uncommon species has led to homogenization across space, changes in the distribution of species making up a substantial portion of the overall community have led to differentiation. We also note that beta diversity trends are sensitive to spatial scale [131,151–153], and therefore our findings for regional communities may differ when assessed at a sub-regional or global scale. While we found that on average, community heterogeneity was higher for larger regions, we did not find a relationship between the size of a region and dissimilarity trend. We did not assess how distance-decay of community similarity varies across surveys, but acknowledge that this spatial pattern likely plays a role in homogenization dynamics and encourage future researchers to explore that intersection [89,154].
Trends in spatial beta diversity were also sensitive to survey-specific characteristics and sampling methodologies. While longer surveys are more likely to detect species gains and losses and therefore directional trends in dissimilarity [131], we did not detect a relationship between sampling period and likelihood of homogenization or differentiation. Surveys vary in their ability to detect and identify uncommon species. While removing species of low abundance did not change our results, removing species that were caught infrequently through time led many regions to exhibit no trend in dissimilarity and a few regions to exhibit trends different than those exhibited using the full dataset. As one example, while Chile exhibited no trend in dissimilarity using the full dataset, the region homogenized when infrequently caught species were removed from the analyses. Future work could explore the role of detectability and uncommon species in homogenization dynamics.
Additionally, we found that higher differentiation was paired with higher regional species richness, reflecting the widely recognized positive relationship between gamma and beta diversity [155]. Due to seasonal migrational patterns and variability in environmental conditions across the year, trends in beta diversity can vary depending on when communities are sampled [156]. We observed lower dissimilarity values and more regions homogenizing later in the year, although this likely reflects more surveys occurring later in the year because we found no significant differences in dissimilarity trends across season of sampling. While rare in ecological datasets, consistency in sampling methodology for long-time series is essential for detecting patterns through time and for performing time-series analysis [25,157,158]. We found no relationship between sampling density and dissimilarity trend, and for all but one region, sampling density was consistent through time. In the Southeast United States survey, an increase in sampling density over time coincided with a decrease in beta diversity. However, this interaction runs contrary to common assumptions that higher sampling density would lead to differentiation as more unique niches are represented.
Conclusion
Our findings demonstrate that, despite significant human impacts on the oceans, most demersal marine fish communities do not conform to the widespread homogenization trend observed in other ecosystems. The observed heterogeneity in marine ecosystems indicates that effective conservation planning should be tailored to regional trends and changes, rather than relying on global proxies [159]. Informing local strategies, in turn, relies on effective systems for monitoring these changes, which can include not only bottom trawl surveys, but also eDNA, sonar, and other technologies.
We found that multi-annual swings between more homogenous and more differentiated community composition have been common and that both temperature and fishing have been key drivers of these changes. Examining temporal dynamics in other marine ecosystems and in terrestrial and freshwater ecosystems will be important for understanding whether large fluctuations are also common in these realms. Future studies exploring the dynamics of functional and phylogenetic dissimilarity across time and space—in all systems, not only marine—will further deepen our knowledge on how structuring factors, such as climate and direct human impacts induce changes in species assembly [160–163].
Supporting information
S1 File. Supporting figures, tables, text, and references.
Contains Figs A-P, Tables A-J, supporting text, and supporting references.
https://doi.org/10.1371/journal.pclm.0000659.s001
(DOCX)
Acknowledgments
We are grateful to the bottom trawl survey data providers, managers, technicians, and volunteers for the use of these data, and their generosity in sharing regional expertise and perspective. We specifically acknowledge Fisheries New Zealand and the Department of Forestry, Fisheries, and the Environment (DFFE) of South Africa for permission to use their data. We thank T. S. Dencker, M. Lindegren, R. Frelat, and the Global Change Research Group (Rutgers University and UC Santa Cruz) for conversations along the way and advice on early analyses. Thank you to A. Gruss (National Institute of Water and Atmospheric Research Ltd. New Zealand) and S. Friedman (National Oceanic and Atmospheric Administration) for providing technical reviews. Finally, thank you to C. Free for tremendous guidance and support through the revision process.
References
- 1. Boivin NL, Zeder MA, Fuller DQ, Crowther A, Larson G, Erlandson JM, et al. Ecological consequences of human niche construction: Examining long-term anthropogenic shaping of global species distributions. Proc Natl Acad Sci U S A. 2016;113(23):6388–96. pmid:27274046
- 2. Elahi R, O’Connor MI, Byrnes JEK, Dunic J, Eriksson BK, Hensel MJS, et al. Recent Trends in Local-Scale Marine Biodiversity Reflect Community Structure and Human Impacts. Curr Biol. 2015;25(14):1938–43. pmid:26166784
- 3. Pandolfi JM, Staples TL, Kiessling W. Increased extinction in the emergence of novel ecological communities. Science. 2020;370(6513):220–2. pmid:33033218
- 4. Williams JW, Jackson ST. Novel climates, no-analog communities, and ecological surprises. Frontiers in Ecology and the Environment. 2007;5(9):475–82.
- 5. McKinney M, Lockwood J. Biotic homogenization: a few winners replacing many losers in the next mass extinction. Trends Ecol Evol. 1999;14(11):450–3. pmid:10511724
- 6. Nielsen TF, Sand‐Jensen K, Dornelas M, Bruun HH. More is less: net gain in species richness, but biotic homogenization over 140 years. Ecol Lett. 2019;22:1650–7.
- 7. Olden JD, Leroy Poff N, Douglas MR, Douglas ME, Fausch KD. Ecological and evolutionary consequences of biotic homogenization. Trends Ecol Evol. 2004;19(1):18–24. pmid:16701221
- 8. Wang S, Loreau M, de Mazancourt C, Isbell F, Beierkuhnlein C, Connolly J, et al. Biotic homogenization destabilizes ecosystem functioning by decreasing spatial asynchrony. Ecology. 2021;102(6):e03332. pmid:33705570
- 9. Li D, Olden JD, Lockwood JL, Record S, McKinney ML, Baiser B. Changes in taxonomic and phylogenetic diversity in the Anthropocene. Proc Biol Sci. 2020;287(1929):20200777. pmid:32546087
- 10. Peoples BK, Davis AJS, Midway SR, Olden JD, Stoczynski L. Landscape-scale drivers of fish faunal homogenization and differentiation in the eastern United States. Hydrobiologia. 2020;847(18):3727–41.
- 11. Baiser B, Olden JD, Record S, Lockwood JL, McKinney ML. Pattern and process of biotic homogenization in the New Pangaea. Proc Biol Sci. 2012;279(1748):4772–7. pmid:23055062
- 12. Newbold T, Hudson LN, Contu S, Hill SLL, Beck J, Liu Y, et al. Widespread winners and narrow-ranged losers: Land use homogenizes biodiversity in local assemblages worldwide. PLoS Biol. 2018;16(12):e2006841. pmid:30513079
- 13. Pearse WD, Cavender‐Bares J, Hobbie SE, Avolio ML, Bettez N, Roy Chowdhury R, et al. Homogenization of plant diversity, composition, and structure in North American urban yards. Ecosphere. 2018;9(2).
- 14. Petsch DK. Causes and consequences of biotic homogenization in freshwater ecosystems. Internat Rev Hydrobiol. 2016;101(3–4):113–22.
- 15. Monchamp M-E, Spaak P, Domaizon I, Dubois N, Bouffard D, Pomati F. Homogenization of lake cyanobacterial communities over a century of climate change and eutrophication. Nat Ecol Evol. 2018;2(2):317–24. pmid:29230026
- 16. Rahel FJ. Biogeographic barriers, connectivity and homogenization of freshwater faunas: it’s a small world after all. Freshwater Biology. 2007;52(4):696–710.
- 17. Karp DS, Frishkoff LO, Echeverri A, Zook J, Juárez P, Chan KMA. Agriculture erases climate-driven β-diversity in Neotropical bird communities. Glob Chang Biol. 2018;24(1):338–49. pmid:28833924
- 18. Keith SA, Newton AC, Morecroft MD, Bealey CE, Bullock JM. Taxonomic homogenization of woodland plant communities over 70 years. Proceedings of the Royal Society B: Biological Sciences. 2009;276:3539–44.
- 19. McCune JL, Vellend M. Gains in native species promote biotic homogenization over four decades in a human‐dominated landscape. Journal of Ecology. 2013;101(6):1542–51.
- 20. Dornelas M, Chase JM, Gotelli NJ, Magurran AE, McGill BJ, Antão LH, et al. Looking back on biodiversity change: lessons for the road ahead. Philos Trans R Soc Lond B Biol Sci. 2023;378(1881):20220199. pmid:37246380
- 21. Rooney TP, Wiegmann SM, Rogers DA, Waller DM. Biotic Impoverishment and Homogenization in Unfragmented Forest Understory Communities. Conservation Biology. 2004;18(3):787–98.
- 22. Radomski PJ, Goeman TJ. The Homogenizing of Minnesota Lake Fish Assemblages. Fisheries. 1995;20(7):20–3.
- 23. Stuble KL, Bewick S, Fisher M, Forister ML, Harrison SP, Shapiro AM, et al. The promise and the perils of resurveying to understand global change impacts. Ecological Monographs. 2021;91(2).
- 24. Brown CJ, O’Connor MI, Poloczanska ES, Schoeman DS, Buckley LB, Burrows MT, et al. Ecological and methodological drivers of species’ distribution and phenology responses to climate change. Glob Chang Biol. 2016;22(4):1548–60. pmid:26661135
- 25. A Maureaud A, Frelat R, Pécuchet L, Shackell N, Mérigot B, Pinsky ML, et al. Are we ready to track climate-driven shifts in marine species across international boundaries? - A global survey of scientific bottom trawl data. Glob Chang Biol. 2021;27(2):220–36. pmid:33067925
- 26. Ellingsen KE, Yoccoz NG, Tveraa T, Frank KT, Johannesen E, Anderson MJ, et al. The rise of a marine generalist predator and the fall of beta diversity. Glob Chang Biol. 2020;26(5):2897–907. pmid:32181966
- 27. Ellingsen KE, Anderson MJ, Shackell NL, Tveraa T, Yoccoz NG, Frank KT. The role of a dominant predator in shaping biodiversity over space and time in a marine ecosystem. J Anim Ecol. 2015;84(5):1242–52. pmid:25981204
- 28. Magurran AE, Dornelas M, Moyes F, Gotelli NJ, McGill B. Rapid biotic homogenization of marine fish assemblages. Nat Commun. 2015;6:8405. pmid:26400102
- 29. Pawluk M, Fujiwara M, Martinez-Andrade F. Climate change linked to functional homogenization of a subtropical estuarine system. Ecol Evol. 2022;12(4):e8783. pmid:35432937
- 30. Zhang Y, Zhang L, Kang Y, Li Y, Chen Z, Li R, et al. Biotic homogenization increases with human intervention: implications for mangrove wetland restoration. Ecography. 20221;2022(4).
- 31. Burman S, Aronson R, van Woesik R. Biotic homogenization of coral assemblages along the Florida reef tract. Mar Ecol Prog Ser. 2012;467:89–96.
- 32. Shiganova T. Biotic Homogenization of Inland Seas of the Ponto-Caspian. Annu Rev Ecol Evol Syst. 2010;41(1):103–25.
- 33. Lenoir J, Bertrand R, Comte L, Bourgeaud L, Hattab T, Murienne J, et al. Species better track climate warming in the oceans than on land. Nat Ecol Evol. 2020;4(8):1044–59. pmid:32451428
- 34. Petsch DK, Bertoncin AP dos S, Ortega JCG, Thomaz SM. Non‐native species drive biotic homogenization, but it depends on the realm, beta diversity facet and study design: a meta‐analytic systematic review. Oikos. 2022;2022(6).
- 35. Arndt E, Marchetti MP, Schembri PJ. Ecological impact of alien marine fishes: insights from freshwater systems based on a comparative review. Hydrobiologia. 2018;817(1):457–74.
- 36.
Ricciardi A, Macisaac HJ. Impacts of Biological Invasions on Freshwater Ecosystems. 1st ed. In: Richardson DM. Fifty Years of Invasion Ecology. 1st ed. Wiley; 2010. 211–224. https://doi.org/10.1002/9781444329988.ch16
- 37. Ballew NG, Bacheler NM, Kellison GT, Schueller AM. Invasive lionfish reduce native fish abundance on a regional scale. Sci Rep. 2016;6:32169. pmid:27578096
- 38. Galil BS. Loss or gain? Invasive aliens and biodiversity in the Mediterranean Sea. Mar Pollut Bull. 2007;55(7–9):314–22. pmid:17222869
- 39. Campbell MD, Pollack AG, Thompson K, Switzer T, Driggers WB, Hoffmayer ER, et al. Rapid spatial expansion and population increase of invasive lionfish (Pterois spp.) observed on natural habitats in the northern Gulf of Mexico. Biol Invasions. 2022;24(1):93–105.
- 40. D‘Amen M, Smeraldo S, Azzurro E. Salinity, not only temperature, drives tropical fish invasions in the Mediterranean Sea, and surface-only variables explain it better. Coral Reefs. 2023;42(2):467–72.
- 41. Regolin AL, Ribeiro MC, Martello F, Melo GL, Sponchiado J, Campanha LF de C, et al. Spatial heterogeneity and habitat configuration overcome habitat composition influences on alpha and beta mammal diversity. Biotropica. 2020;52(5):969–80.
- 42. Rolls RJ, Deane DC, Johnson SE, Heino J, Anderson MJ, Ellingsen KE. Biotic homogenisation and differentiation as directional change in beta diversity: synthesising driver-response relationships to develop conceptual models across ecosystems. Biol Rev Camb Philos Soc. 2023;98(4):1388–423. pmid:37072381
- 43. Veech JA, Crist TO. Habitat and climate heterogeneity maintain beta‐diversity of birds among landscapes within ecoregions. Global Ecology and Biogeography. 2007;16(5):650–6.
- 44. Boyd PW, Cornwall CE, Davison A, Doney SC, Fourquez M, Hurd CL, et al. Biological responses to environmental heterogeneity under future ocean conditions. Glob Chang Biol. 2016;22(8):2633–50. pmid:27111095
- 45. Reusch TBH, Boyd PW. Experimental evolution meets marine phytoplankton. Evolution. 2013;67(7):1849–59. pmid:23815643
- 46. Gossner MM, Lewinsohn TM, Kahl T, Grassein F, Boch S, Prati D, et al. Land-use intensification causes multitrophic homogenization of grassland communities. Nature. 2016;540(7632):266–9. pmid:27919075
- 47. Hiddink JG, Jennings S, Kaiser MJ, Queirós AM, Duplisea DE, Piet GJ. Cumulative impacts of seabed trawl disturbance on benthic biomass, production, and species richness in different habitats. Can J Fish Aquat Sci. 2006;63:721–36.
- 48. McKinney ML. Urbanization as a major cause of biotic homogenization. Biological Conservation. 2006;127(3):247–60.
- 49. Thrush SF, Gray JS, Hewitt JE, Ugland KI. Predicting the effects of habitat homogenization on marine biodiversity. Ecological Applications. 2006;16(5):1636–42.
- 50. Thrush SF, Halliday J, Hewitt JE, Lohrer AM. The effects of habitat loss, fragmentation, and community homogenization on resilience in estuaries. Ecol Appl. 2008;18(1):12–21. pmid:18372552
- 51. McCauley DJ, Pinsky ML, Palumbi SR, Estes JA, Joyce FH, Warner RR. Marine defaunation: animal loss in the global ocean. Science. 2015;347(6219):1255641. pmid:25593191
- 52. Essington TE, Beaudreau AH, Wiedenmann J. Fishing through marine food webs. Proc Natl Acad Sci U S A. 2006;103(9):3171–5. pmid:16481614
- 53. Pauly D, Christensen V V, Dalsgaard J, Froese R, Torres F Jr. Fishing down marine food webs. Science. 1998;279(5352):860–3. pmid:9452385
- 54. Pinsky ML, Worm B, Fogarty MJ, Sarmiento JL, Levin SA. Marine taxa track local climate velocities. Science. 2013;341(6151):1239–42. pmid:24031017
- 55. Araújo FG, de Azevedo MCC, Guedes APP. Inter-decadal changes in fish communities of a tropical bay in southeastern Brazil. Regional Studies in Marine Science. 2016;3:107–18.
- 56. Bianchi CN, Azzola A, Parravicini V, Peirano A, Morri C, Montefalcone M. Abrupt Change in a Subtidal Rocky Reef Community Coincided with a Rapid Acceleration of Sea Water Warming. Diversity. 2019;11(11):215.
- 57. Chihoub S, Christaki U, Chelgham S, Amara R, Ramdane Z, Zebboudj A, et al. Coastal eutrophication as a potential driver of functional homogenization of copepod species assemblages in the Mediterranean Sea. Ecological Indicators. 2020;115:106388.
- 58. McClain CR, Nunnally C, Benfield MC. Persistent and substantial impacts of the Deepwater Horizon oil spill on deep-sea megafauna. R Soc Open Sci. 2019;6(8):191164. pmid:31598269
- 59. Richardson LE, Graham NAJ, Pratchett MS, Eurich JG, Hoey AS. Mass coral bleaching causes biotic homogenization of reef fish assemblages. Glob Chang Biol. 2018;24(7):3117–29. pmid:29633512
- 60. Sarker SK, Matthiopoulos J, Mitchell SN, Ahmed ZU, Mamun MBA, Reeve R. 1980s-2010s: The world’s largest mangrove ecosystem is becoming homogeneous. Biol Conserv. 2019;236:79–91. pmid:31496538
- 61. Deutsch C, Ferrel A, Seibel B, Pörtner H-O, Huey RB. Ecophysiology. Climate change tightens a metabolic constraint on marine habitats. Science. 2015;348(6239):1132–5. pmid:26045435
- 62. Faizal M, Rafiuddin Ahmed M. On the ocean heat budget and ocean thermal energy conversion. Int J Energy Res. 2011;35(13):1119–44.
- 63. Questad EJ, Foster BL. Coexistence through spatio-temporal heterogeneity and species sorting in grassland plant communities. Ecol Lett. 2008;11(7):717–26. pmid:18445035
- 64.
Tilman D, Pacala S. The Maintenance of Species Richness in Plant Communities. In: Ricklefs RE, Schluter D. Species Diversity in Ecological Communities. Chicago: University of Chicago Press; 1993. 13–25. https://www.cedarcreek.umn.edu/biblio/fulltext/t1193.pdf
- 65. Fredston-Hermann A, Selden R, Pinsky M, Gaines SD, Halpern BS. Cold range edges of marine fishes track climate change better than warm edges. Glob Chang Biol. 2020;26(5):2908–22. pmid:32037696
- 66. Coll M, Navarro J, Palomera I. Ecological role, fishing impact, and management options for the recovery of a Mediterranean endemic skate by means of food web models. Biological Conservation. 2013;157:108–20.
- 67. Friedlander AM, Ballesteros E, Beets J, Berkenpas E, Gaymer CF, Gorny M, et al. Effects of isolation and fishing on the marine ecosystems of Easter Island and Salas y Gómez, Chile. Aquatic Conservation. 2013;23(4):515–31.
- 68. Eriksson BK, Sieben K, Eklöf J, Ljunggren L, Olsson J, Casini M, et al. Effects of altered offshore food webs on coastal ecosystems emphasize the need for cross-ecosystem management. Ambio. 2011;40(7):786–97. pmid:22338716
- 69. Maureaud AA, Palacios-Abrantes J, Kitchel Z, Mannocci L, Pinsky ML, Fredston A, et al. FISHGLOB_data: an integrated dataset of fish biodiversity sampled with scientific bottom-trawl surveys. Sci Data. 2024;11(1):24. pmid:38177193
- 70. Cusser S, Helms J 4th, Bahlai CA, Haddad NM. How long do population level field experiments need to be? Utilising data from the 40-year-old LTER network. Ecol Lett. 2021;24(5):1103–11. pmid:33616295
- 71. Massiocotte P, South A. rnaturalearth: World Map Data from Natural Earth. 2023. https://CRAN.R-project.org/package=rnaturalearth
- 72. Pebesma E. Simple features for R: standardized support for spatial vector data. R J. 2018;10:439–46.
- 73.
Pebesma E, Bivand R. Spatial Data Science: With Applications in R. New York: Chapman and Hall/CRC; 2023. https://doi.org/10.1201/9780429459016
- 74.
Wickham H. ggplot2. Cham: Springer International Publishing; 2016. https://doi.org/10.1007/978-3-319-24277-4
- 75. Oksanen J, Simpson GL, Blanchet FG, Kindt R, Legendre P, Minchin PR, et al. vegan: Community Ecology Package (2.6-4). 2022. https://CRAN.R-project.org/package=vegan
- 76.
Ahyong S, Boyko CB, Bailly N, Bernot J, Bieler R, Brandão SN, et al. World Register of Marine Species (WoRMS). WoRMS Editorial Board; 2023. https://www.marinespecies.org
- 77. Chamberlain S. worrms: World Register of Marine Species (WoRMS) Client (0.4.2). 2020. https://CRAN.R-project.org/package=worrms
- 78. Arroyo-Correa B, Jordano P, Bartomeus I. Intraspecific variation in species interactions promotes the feasibility of mutualistic assemblages. Ecol Lett. 2023;26(3):448–59. pmid:36688287
- 79. García‐Navas V, Sattler T, Schmid H, Ozgul A. Temporal homogenization of functional and beta diversity in bird communities of the Swiss Alps. Diversity and Distributions. 2020;26(8):900–11.
- 80. Li S-P, Cadotte MW, Meiners SJ, Pu Z, Fukami T, Jiang L. Convergence and divergence in a long-term old-field succession: the importance of spatial scale and species abundance. Ecol Lett. 2016;19(9):1101–9. pmid:27373449
- 81. Marion ZH, Fordyce JA, Fitzpatrick BM. Pairwise beta diversity resolves an underappreciated source of confusion in calculating species turnover. Ecology. 2017;98(4):933–9. pmid:28134975
- 82. Salces-Castellano A, Patiño J, Alvarez N, Andújar C, Arribas P, Braojos-Ruiz JJ, et al. Climate drives community-wide divergence within species over a limited spatial scale: evidence from an oceanic island. Ecol Lett. 2020;23(2):305–15. pmid:31762170
- 83. Soininen J, Heino J, Wang J. A meta‐analysis of nestedness and turnover components of beta diversity across organisms and ecosystems. Global Ecol Biogeogr. 2018;27(1):96–109.
- 84. Strandberg NA, Steinbauer MJ, Walentowitz A, Gosling WD, Fall PL, Prebble M, et al. Floristic homogenization of South Pacific islands commenced with human arrival. Nat Ecol Evol. 2024;8(3):511–8. pmid:38225430
- 85. Tatsumi S, Strengbom J, Čugunovs M, Kouki J. Partitioning the colonization and extinction components of beta diversity across disturbance gradients. Ecology. 2020;101(12):e03183. pmid:32892360
- 86. Vale CG, Arenas F, Barreiro R, Piñeiro‐Corbeira C. Understanding the local drivers of beta‐diversity patterns under climate change: The case of seaweed communities in Galicia, North West of the Iberian Peninsula. Diversity and Distributions. 2021;27(9):1696–705.
- 87. Vannette RL, Fukami T. Dispersal enhances beta diversity in nectar microbes. Ecol Lett. 2017;20(7):901–10. pmid:28597955
- 88. Baselga A. Separating the two components of abundance‐based dissimilarity: balanced changes in abundance vs. abundance gradients. Methods Ecol Evol. 2013;4(6):552–7.
- 89. Soininen J, McDonald R, Hillebrand H. The distance decay of similarity in ecological communities. Ecography. 2007;30(1):3–12.
- 90. Legendre P, De Cáceres M. Beta diversity as the variance of community data: dissimilarity coefficients and partitioning. Ecol Lett. 2013;16(8):951–63. pmid:23809147
- 91. Koleff P, Gaston KJ, Lennon JJ. Measuring beta diversity for presence–absence data. Journal of Animal Ecology. 2003;72(3):367–82.
- 92. Anderson MJ, Ellingsen KE, McArdle BH. Multivariate dispersion as a measure of beta diversity. Ecol Lett. 2006;9(6):683–93. pmid:16706913
- 93. Schroeder PJ, Jenkins DG. How robust are popular beta diversity indices to sampling error?. Ecosphere. 2018;9(2).
- 94.
Bates D, Maechler M, Bolker B, Walker S, Christensen RHB. lme4: Linear Mixed-Effects Models using “Eigen” and S4. 2022.
- 95.
Pinheiro J, Bates D, DebRoy D. nlme: Linear and Nonlinear Mixed Effects Models. 2022.
- 96. Wood S. mgcv (1.8-42). 2023. https://cran.r-project.org/web/packages/mgcv/index.html
- 97.
Akaike H. Information Theory and an Extension of the Maximum Likelihood Principle. In: Parzen E, Tanabe K, Kitagawa G. Selected Papers of Hirotugu Akaike. New York, NY: Springer; 1998. 199–213. https://doi.org/10.1007/978-1-4612-1694-0_15
- 98. Burnham KP, Anderson DR, Huyvaert KP. AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons. Behav Ecol Sociobiol. 2011;65(1):23–35.
- 99.
Gombin J, Vaidyanathan R, Agafonkin V. Concaveman: A very fast 2D concave hull algorithm. 2020.
- 100. Alabia ID, García Molinos J, Hirata T, Mueter FJ, Hirawake T, Saitoh S-I. Marine biodiversity refugia in a climate-sensitive subarctic shelf. Glob Chang Biol. 2021;27(14):3299–311. pmid:33899298
- 101. Figueira WF, Booth DJ. Increasing ocean temperatures allow tropical fishes to survive overwinter in temperate waters. Global Change Biology. 2010;16(2):506–16.
- 102. White ER, Hastings A. Seasonality in ecology: Progress and prospects in theory. Ecological Complexity. 2020;44:100867.
- 103. Nielsen JM, Rogers LA, Brodeur RD, Thompson AR, Auth TD, Deary AL, et al. Responses of ichthyoplankton assemblages to the recent marine heatwave and previous climate fluctuations in several Northeast Pacific marine ecosystems. Glob Chang Biol. 2020;27(3):506–20. pmid:33107157
- 104. Carton JA, Giese BS. A Reanalysis of Ocean Climate Using Simple Ocean Data Assimilation (SODA). Monthly Weather Review. 2008;136(8):2999–3017.
- 105. Huang B, Liu C, Banzon V, Freeman E, Graham G, Hankins B, et al. Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version 2.1. Journal of Climate. 2021;34(8):2923–39.
- 106. Lindegren M, Checkley DM Jr, Koslow JA, Goericke R, Ohman MD. Climate-mediated changes in marine ecosystem regulation during El Niño. Glob Chang Biol. 2018;24(2):796–809. pmid:29156088
- 107. Mantua NJ, Hare SR, Zhang Y, Wallace JM, Francis RC. A Pacific Interdecadal Climate Oscillation with Impacts on Salmon Production. Bull Amer Meteor Soc. 1997;78(6):1069–79.
- 108. Sea Around Us Concepts, Design and Data. 2020. https://seaaroundus.org
- 109. Nye JA, Gamble RJ, Link JS. The relative impact of warming and removing top predators on the Northeast US large marine biotic community. Ecological Modelling. 2013;264:157–68.
- 110.
Jennings S, Kaiser MJ. The Effects of Fishing on Marine Ecosystems. In. : Blaxter JHS, Southward AJ, Tyler PA, editors. Advances in Marine Biology. Academic Press; 1998. 201–352. https://doi.org/10.1016/S0065-2881(08)60212-6
- 111. Bartón K. MuMIn: Multi-Model Inference. 2023. https://cran.r-project.org/web/packages/MuMIn/MuMIn.pdf
- 112. Arce-Peña NP, Arroyo-Rodríguez V, Avila-Cabadilla LD, Moreno CE, Andresen E. Homogenization of terrestrial mammals in fragmented rainforests: the loss of species turnover and its landscape drivers. Ecol Appl. 2022;32(1):e02476. pmid:34653282
- 113. Cazelles K, Bartley T, Guzzo MM, Brice M-H, MacDougall AS, Bennett JR, et al. Homogenization of freshwater lakes: Recent compositional shifts in fish communities are explained by gamefish movement and not climate change. Glob Chang Biol. 2019;25(12):4222–33. pmid:31502733
- 114. Blowes SA, McGill B, Brambilla V, Chow CFY, Engel T, Fontrodona-Eslava A, et al. Synthesis reveals approximately balanced biotic differentiation and homogenization. Sci Adv. 2024;10(8):eadj9395. pmid:38381832
- 115. Gruner DS, Bracken MES, Berger SA, Eriksson BK, Gamfeldt L, Matthiessen B, et al. Effects of experimental warming on biodiversity depend on ecosystem type and local species composition. Oikos. 2017;126(1):8–17.
- 116. Nagelkerken I, Connell SD. Ocean acidification drives global reshuffling of ecological communities. Glob Chang Biol. 2022;28(23):7038–48. pmid:36172974
- 117. Benito X, Vilmi A, Luethje M, Carrevedo ML, Lindholm M, Fritz SC. Spatial and Temporal Ecological Uniqueness of Andean Diatom Communities Are Correlated With Climate, Geodiversity and Long-Term Limnological Change. Front Ecol Evol. 2020;8.
- 118. Britton AJ, Beale CM, Towers W, Hewison RL. Biodiversity gains and losses: Evidence for homogenisation of Scottish alpine vegetation. Biological Conservation. 2009;142(8):1728–39.
- 119. Lindholm M, Alahuhta J, Heino J, Hjort J, Toivonen H. Changes in the functional features of macrophyte communities and driving factors across a 70-year period. Hydrobiologia. 2020;847(18):3811–27.
- 120. Lindholm M, Alahuhta J, Heino J, Toivonen H. No biotic homogenisation across decades but consistent effects of landscape position and pH on macrophyte communities in boreal lakes. Ecography. 2019;43(2):294–305.
- 121. Pinceloup N, Poulin M, Brice M-H, Pellerin S. Vegetation changes in temperate ombrotrophic peatlands over a 35 year period. PLoS One. 2020;15(2):e0229146. pmid:32053706
- 122.
Gould E, Fraser HS, Parker TH, Nakagawa S, Griffith SC, Vesk PA, et al. Same data, different analysts: variation in effect sizes due to analytical decisions in ecology and evolutionary biology. 2023.
- 123. Rosenblad KC, Sax DF. A new framework for investigating biotic homogenization and exploring future trajectories: oceanic island plant and bird assemblages as a case study. Ecography. 2017;40(9):1040–9.
- 124. Olden JD, Rooney TP. On defining and quantifying biotic homogenization. Glob Ecol Biogeogr. 2006;15:113–20.
- 125. Wayman JP, Sadler JP, Martin TE, Graham LJ, White HJ, Tobias JA, et al. Unravelling the complexities of biotic homogenization and heterogenization in the British avifauna. J Anim Ecol. 2024;93(9):1288–302. pmid:39120041
- 126. Germain SJ, Lutz JA. Climate extremes may be more important than climate means when predicting species range shifts. Climatic Change. 2020;163(1):579–98.
- 127. Leriorato JC, Nakamura Y, Uy WH. Cold thermal tolerance as a range-shift predictive trait: an essential link in the disparity of occurrence of tropical reef fishes in temperate waters. Mar Biol. 2021;168(6).
- 128. Morley JW, Batt RD, Pinsky ML. Marine assemblages respond rapidly to winter climate variability. Glob Chang Biol. 2017;23(7):2590–601. pmid:27885755
- 129. Dana JD. On an isothermal oceanic chart, illustrating the geographical distribution of marine animals. Am J Sci Arts. 1853;66:153–67, 325–7, 391–2.
- 130.
Su W, Marvel K, Delgado R, Aarons S, Chatterjee A, Garcia ME, et al. Climate Trends. Fifth Natl Clim Assess. U.S. Global Change Research Program: Washington, DC; 2023. https://nca2023.globalchange.gov/chapter/2/
- 131. Olden JD, Poff NL. Toward a mechanistic understanding and prediction of biotic homogenization. Am Nat. 2003;162(4):442–60. pmid:14582007
- 132. Beaugrand G, Edwards M, Hélaouët P. An ecological partition of the Atlantic Ocean and its adjacent seas. Progress in Oceanography. 2019;173:86–102.
- 133. Cavanaugh KC, Dangremond EM, Doughty CL, Williams AP, Parker JD, Hayes MA, et al. Climate-driven regime shifts in a mangrove-salt marsh ecotone over the past 250 years. Proc Natl Acad Sci U S A. 2019;116(43):21602–8. pmid:31591236
- 134. Walker BK, Fisco Becker D, Williams GJ, Kilfoyle AK, Smith SG, Kozachuk A. Regional reef fish assemblage maps provide baseline biogeography for tropicalization monitoring. Sci Rep. 2024;14(1):7893. pmid:38570549
- 135. Free CM, Mangin T, Molinos JG, Ojea E, Burden M, Costello C, et al. Realistic fisheries management reforms could mitigate the impacts of climate change in most countries. PLoS One. 2020;15(3):e0224347. pmid:32134926
- 136. Gaines SD, Costello C, Owashi B, Mangin T, Bone J, Molinos JG, et al. Improved fisheries management could offset many negative effects of climate change. Sci Adv. 2018;4(8):eaao1378. pmid:30167455
- 137. Branch TA, Watson R, Fulton EA, Jennings S, McGilliard CR, Pablico GT, et al. The trophic fingerprint of marine fisheries. Nature. 2010;468(7322):431–5. pmid:21085178
- 138. Garrison L. Fishing effects on spatial distribution and trophic guild structure of the fish community in the Georges Bank region. ICES Journal of Marine Science. 2000;57(3):723–30.
- 139. Casini M, Blenckner T, Möllmann C, Gårdmark A, Lindegren M, Llope M, et al. Predator transitory spillover induces trophic cascades in ecological sinks. Proc Natl Acad Sci U S A. 2012;109(21):8185–9. pmid:22505739
- 140. Ritchie EG, Johnson CN. Predator interactions, mesopredator release and biodiversity conservation. Ecol Lett. 2009;12(9):982–98. pmid:19614756
- 141. Dulvy NK, Pacoureau N, Rigby CL, Pollom RA, Jabado RW, Ebert DA, et al. Overfishing drives over one-third of all sharks and rays toward a global extinction crisis. Curr Biol. 2021;31(21):4773-4787.e8. pmid:34492229
- 142. Bess R. New Zealand’s indigenous people and their claims to fisheries resources. Marine Policy. 2001;25(1):23–32.
- 143. Bolster WJ. Putting the Ocean in Atlantic History: Maritime Communities and Marine Ecology in the Northwest Atlantic, 1500–1800. The American Historical Review. 2008;113(1):19–47.
- 144. Castañeda RA, Burliuk CMM, Casselman JM, Cooke SJ, Dunmall KM, Forbes LS, et al. A Brief History of Fisheries in Canada. Fisheries. 2020;45(6):303–18.
- 145. Srinivasan UT, Watson R, Rashid Sumaila U. Global fisheries losses at the exclusive economic zone level, 1950 to present. Marine Policy. 2012;36(2):544–9.
- 146. Standal D, Hersoug B. Shaping technology, building society; the industrialization of the Norwegian cod fisheries. Marine Policy. 2015;51:66–74.
- 147. Glantz MH. Science, politics and economics of the Peruvian anchoveta fishery. Marine Policy. 1979;3(3):201–10.
- 148. Hilborn R, Hively DJ, Loke NB, de Moor CL, Kurota H, Kathena JN, et al. Global status of groundfish stocks. Fish and Fisheries. 2021;22(5):911–28.
- 149. Rastetter EB, Ohman MD, Elliott KJ, Rehage JS, Rivera‐Monroy VH, Boucek RE, et al. Time lags: insights from the U.S. Long Term Ecological Research Network. Ecosphere. 2021;12(5).
- 150. Tatsumi S, Iritani R, Cadotte MW. Temporal changes in spatial variation: partitioning the extinction and colonisation components of beta diversity. Ecol Lett. 2021;24(5):1063–72. pmid:33715273
- 151. Chase JM, McGill BJ, McGlinn DJ, May F, Blowes SA, Xiao X, et al. Embracing scale-dependence to achieve a deeper understanding of biodiversity and its change across communities. Ecol Lett. 2018;21(11):1737–51. pmid:30182500
- 152. Chase JM, McGill BJ, Thompson PL, Antão LH, Bates AE, Blowes SA, et al. Species richness change across spatial scales. Oikos. 2019;128(8):1079–91.
- 153. Keil P, Biesmeijer JC, Barendregt A, Reemer M, Kunin WE. Biodiversity change is scale-dependent: an example from Dutch and UK hoverflies (Diptera, Syrphidae). Ecography. 2010;34(3):392–401.
- 154. Morlon H, Chuyong G, Condit R, Hubbell S, Kenfack D, Thomas D, et al. A general framework for the distance-decay of similarity in ecological communities. Ecol Lett. 2008;11(9):904–17. pmid:18494792
- 155. Kraft NJB, Comita LS, Chase JM, Sanders NJ, Swenson NG, Crist TO, et al. Disentangling the drivers of β diversity along latitudinal and elevational gradients. Science. 2011;333(6050):1755–8. pmid:21940897
- 156. Li Y, Ma S, Li J, Liu S, Tian Y. Difference in seasonal shift of spatial homogenization between taxonomic and functional structure in demersal fish communities. Estuarine, Coastal and Shelf Science. 2023;295:108561.
- 157. Hughes BB, Beas-Luna R, Barner AK, Brewitt K, Brumbaugh DR, Cerny-Chipman EB, et al. Long-Term Studies Contribute Disproportionately to Ecology and Policy. BioScience. 2017;67(3):271–81.
- 158. Jansen T, Kristensen K, Payne M, Edwards M, Schrum C, Pitois S. Long-term retrospective analysis of mackerel spawning in the North Sea: a new time series and modeling approach to CPR data. PLoS One. 2012;7(6):e38758. pmid:22737221
- 159. Jupiter SD, Cohen PJ, Weeks R, Tawake A, Govan H. Locally-managed marine areas: multiple objectives and diverse strategies. Pac Conserv Biol. 2014;20(2):165.
- 160. Baiser B, Lockwood JL. The relationship between functional and taxonomic homogenization. Glob Ecol Biogeogr. 2011;20:134–44.
- 161. Harrison T, Gibbs J, Winfree R. Phylogenetic homogenization of bee communities across ecoregions. Global Ecol Biogeogr. 2018;27(12):1457–66.
- 162. Rocha‐Santos L, Mayfield MM, Lopes AV, Pessoa MS, Talora DC, Faria D, et al. The loss of functional diversity: A detrimental influence of landscape‐scale deforestation on tree reproductive traits. Journal of Ecology. 2020;108(1):212–23.
- 163. Saladin B, Pellissier L, Graham CH, Nobis MP, Salamin N, Zimmermann NE. Rapid climate change results in long-lasting spatial homogenization of phylogenetic diversity. Nat Commun. 2020;11(1):4663. pmid:32938914