Microbial Bioactives

1.Introduction

Microzooplankton and protist communities form the living engine of marine food webs, quietly governing the flow of energy and matter across the ocean. Although often overlooked because of their microscopic size, these organisms represent an extraordinary share of marine eukaryotic diversity and play indispensable roles as primary consumers, secondary producers, decomposers, and recyclers of nutrients (Fenchel, 2008; De Vargas et al., 2015). Through intense grazing pressure on phytoplankton and bacteria, microzooplankton process an estimated 20–30 Pg of carbon annually—more than twice the amount consumed by mesozooplankton—placing them at the center of global carbon cycling and climate regulation (Buitenhuis et al., 2010; Worden et al., 2015).

Despite their centrality to ecosystem functioning, microzooplankton communities are increasingly exposed to multiple, overlapping global stressors driven by climate change and human activity. Ocean warming, acidification, deoxygenation, eutrophication, and altered nutrient stoichiometry are reshaping marine habitats at unprecedented rates (Bindoff et al., 2019). Understanding how these pressures influence microzooplankton vulnerability, resilience, and adaptive capacity has therefore become a critical challenge for marine ecology. Recent synthesis efforts emphasize that ecosystem risk cannot be inferred solely from exposure; rather, it emerges from the interaction between exposure, biological sensitivity, and adaptive capacity at both species and community levels (López-Abbate, 2021).

This introduction draws on a systematic review and meta-analytical interpretation of peer-reviewed studies to examine how global environmental hazards affect microzooplankton communities, while highlighting the biological traits and community processes that buffer ecosystem functioning. By integrating quantitative effect sizes, experimental evidence, and network-based analyses, the review frames microzooplankton not simply as vulnerable components of marine ecosystems, but as dynamic and resilient systems capable of reorganization under stress.

The foundational role of microzooplankton is rooted in the microbial loop, a conceptual framework that recognizes microbes as dominant agents of carbon and nutrient recycling in the sea (Fenchel, 2008). By grazing on picophytoplankton and heterotrophic bacteria that are inaccessible to larger consumers, microzooplankton repackage dissolved and particulate organic matter into forms that can move upward through food webs or sink to depth (Buitenhuis et al., 2010; Worden et al., 2015). This function becomes especially important in oligotrophic oceans, where smaller cells dominate primary production and classical food chains are truncated.

Beyond carbon transfer, microzooplankton influence nutrient regeneration, trace gas production, and microbial diversity through selective grazing (Mitra et al., 2014; Park et al., 2014). Viral interactions further complicate these dynamics, as viral infection of microbial hosts redirects organic matter through the viral shunt, enhancing nutrient recycling while limiting carbon export (Wilhelm & Suttle, 1999; Suttle, 2007). Meta-analytical evidence indicates that viral abundance is strongly linked to environmental drivers such as nitrogen and phosphorus availability, highlighting the sensitivity of microbial mortality pathways to nutrient imbalance (Finke et al., 2017).

Among climate-related stressors, ocean warming represents one of the most pervasive and rapidly intensifying hazards. Temperature exerts direct control over metabolic rates, growth, and grazing activity in heterotrophic protists (Rose & Caron, 2007). Evidence synthesized across latitudinal gradients suggests that warming generally favors smaller, fast-growing protists, promoting poleward range expansions and seasonal shifts in community composition (Edwards et al., 2006; Jonkers et al., 2019). While polar and subpolar ecosystems may experience amplified responses due to historical thermal constraints, many taxa demonstrate high adaptive capacity by tracking shifting thermal niches and prey phenology (Field et al., 2006).

Ocean acidification (OA), by contrast, is characterized by high exposure but comparatively low sensitivity for most non-calcifying microzooplankton (Nielsen et al., 2010; Suffrian et al., 2008). Experimental and field studies indicate that projected end-of-century pH changes exert limited direct effects on growth and grazing for many taxa. Instead, community-level impacts are often indirect, mediated through changes in phytoplankton food quality and elemental composition (Park et al., 2014). Calcifying organisms such as planktonic foraminifera remain an exception, exhibiting measurable reductions in shell weight and calcification under acidified conditions (Moy et al., 2009).

In contrast to warming and acidification, deoxygenation emerges as the hazard associated with the highest ecological risk. Declining oxygen concentrations directly constrain aerobic metabolism, reduce grazing rates, and restructure species assemblages (Breitburg et al., 2018). Ciliates, in particular, exhibit narrow oxygen niches and pronounced reductions in growth under hypoxic conditions (Rocke & Liu, 2014). Meta-analytical evidence further demonstrates that deoxygenation drives strong species replacement and diversity loss, increasing community vulnerability despite the persistence of overall function (López-Abbate, 2021).

In coastal and estuarine systems, anthropogenic eutrophication interacts with natural gradients such as salinity to shape microzooplankton assemblages. Elevated nutrient inputs often weaken trophic coupling by inducing feeding saturation, whereby grazers cannot control rapidly increasing prey biomass (Redden et al., 2002; Heisler et al., 2008). This decoupling favors the proliferation of unpalatable or harmful algal species, many of which rely on mixotrophy as a dominant nutritional strategy (Burkholder et al., 2008; Mitra et al., 2014).

Salinity gradients impose additional constraints, acting as strong environmental filters that limit dispersal and select for specialized taxa. Large-scale sequencing studies reveal that salinity explains a substantial proportion of community dissimilarity in estuarine protists, often exceeding the influence of temperature or nutrients (Jiang et al., 2022; Xu et al., 2022). These findings underscore the importance of regional context in assessing ecological risk, particularly in highly modified coastal ecosystems.

A defining feature of microzooplankton communities is their extraordinary adaptive capacity, rooted in functional diversity, rapid generation times, and access to a vast “rare biosphere” of low-abundance taxa (López-Abbate, 2021). Mixotrophy exemplifies this flexibility, allowing organisms to combine phototrophy and phagotrophy to exploit fluctuating resource landscapes (Mitra et al., 2014). Stoichiometric plasticity further enables protists to adjust internal nutrient ratios, sustaining growth even when prey quality declines (Mitra et al., 2014).

Life-cycle strategies such as resting cyst formation provide additional resilience. Under anoxic conditions, dinoflagellate cysts can remain viable for decades or even centuries, effectively functioning as biological seed banks that reintroduce diversity following disturbance (Lundholm et al., 2011; Persson & Smith, 2022). While this mechanism enhances long-term persistence, it can also amplify ecological risk by facilitating recurrent harmful algal blooms when favorable conditions return.

The limits of microzooplankton resilience are perhaps best illustrated by their presence in extreme habitats such as Deep-sea Hypersaline Anoxic Basins (DHABs). In these environments, marine fungi and other protists thrive under extreme salinity, toxicity, and hydrostatic pressure, actively contributing to organic matter remineralization (Gostinčar et al., 2011; Barone et al., 2019). Such systems highlight the remarkable physiological and metabolic plasticity of micro-eukaryotes, reinforcing their role as ubiquitous and adaptable components of marine ecosystems.

Advances in high-throughput sequencing and bioinformatics have further transformed understanding of community resilience. Amplicon sequence variant (ASV)-based approaches, such as DADA2, provide fine-scale resolution of protistan diversity, allowing robust detection of subtle shifts in community structure (Callahan et al., 2016). Network analyses reveal that climate stress tends to reduce interaction complexity and increase vulnerability to species loss, as demonstrated in long-term lake studies where warming disrupted traditional plankton succession (Forster et al., 2021). Yet even in simplified networks, functional roles are often maintained through species replacement, underscoring the stabilizing influence of redundancy.

Collectively, evidence synthesized from systematic review and meta-analysis supports a nuanced view of microzooplankton vulnerability. While individual taxa may exhibit high sensitivity to specific stressors—particularly deoxygenation—the community as a whole often maintains functional stability through adaptive traits and species replacement (López-Abbate, 2021; Qiao et al., 2024). This balance between sensitivity and resilience positions microzooplankton as both indicators of environmental change and key buffers against ecosystem collapse. Understanding their responses is therefore essential for predicting the future trajectory of marine biogeochemical cycles in an increasingly stressed ocean.

2. Materials and Methods

2.1. Literature Search and Selection Criteria

This systematic review and meta-analysis were conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (Page et al., 2021) including the studies examining the responses of microzooplankton and protist communities to environmental stressors, including ocean warming, acidification, deoxygenation, eutrophication, salinity changes, and other anthropogenic impacts. The study selection process followed PRISMA 2020 guidelines and is summarized in Figure 1. Databases searched included PubMed, Web of Science, Scopus, and Google Scholar, covering publications from 2000 to 2023 to capture contemporary trends and experimental insights. Search terms combined controlled vocabulary (MeSH

Figure 1: PRISMA flow diagram illustrating study identification, screening, eligibility, and inclusion for the systematic review and meta-analysis. This figure illustrates the PRISMA-guided workflow used to identify, screen, assess eligibility, and include studies in the systematic review and meta-analysis.

terms) and keywords, such as "microzooplankton," "protists," "marine ecosystems," "climate change," "ocean acidification," "deoxygenation," "eutrophication," "functional redundancy," "mixotrophy," and "resting cysts." Boolean operators (AND, OR) were used to refine results, ensuring the inclusion of studies reporting experimental, observational, and modeling approaches.

Inclusion criteria were: (i) studies reporting quantitative or qualitative responses of microzooplankton or protist communities to at least one environmental stressor, (ii) studies conducted in marine, estuarine, or hypersaline environments, (iii) experimental studies using in situ or laboratory manipulations, mesocosm studies, or field surveys, and (iv) studies providing sufficient statistical information to calculate effect sizes (e.g., means, standard deviations, sample sizes, regression coefficients). Exclusion criteria included studies focused solely on bacteria or viruses without microbial eukaryote data, studies without clear environmental stressor measurements, or publications not available in English. After initial screening of titles and abstracts, full texts of potentially relevant studies were reviewed independently by two researchers. Discrepancies in inclusion were resolved through discussion and consensus.

Data extraction followed a standardized protocol, recording study location, habitat type, environmental stressor(s), microzooplankton or protist taxa assessed, methodological approach, exposure duration, measured biological responses (e.g., growth rate, grazing, diversity, cyst formation), and statistical results. Effect sizes were extracted directly when reported or calculated from available data using established formulas for standardized mean differences and correlation coefficients, enabling integration into meta-analytic models.

2.2. Data Synthesis and Meta-Analysis

To synthesize the collected data quantitatively, we applied meta-analytic techniques following PRISMA guidelines. Effect sizes were expressed as standardized mean differences (Hedges’ g) for experimental comparisons and correlation coefficients (r) for continuous environmental gradients. Variance and confidence intervals were calculated to assess precision, and random-effects models were used to account for heterogeneity among studies, considering differences in habitat, taxa, and methodological approaches. Heterogeneity was quantified using Cochran’s Q-test and I² statistics, with I² > 50% indicating substantial heterogeneity.

Subgroup analyses were conducted to evaluate taxon-specific responses and environmental context dependencies. For instance, responses of ciliates, dinoflagellates, and mixotrophic flagellates were analyzed separately, as were responses in polar, temperate, and tropical regions. Sensitivity analyses were conducted to assess the influence of high-leverage studies on overall effect size estimates. Publication bias was evaluated using funnel plots and Egger’s regression test, ensuring that conclusions were not unduly influenced by selective reporting.

Network analyses were performed to assess changes in community structure and robustness under environmental stressors. Co-occurrence matrices were constructed using species presence-absence and abundance data, with topological metrics, including connectivity, modularity, and centrality, computed to evaluate network resilience. Changes in network architecture were then correlated with environmental stressor intensity and variability to quantify the effects of climate-driven and anthropogenic pressures on protistan community interactions.

2.3. Environmental Variables and Experimental Approaches

Environmental variables of interest were grouped into five categories: (i) thermal stress (temperature), (ii) ocean acidification (pH and carbonate chemistry), (iii) oxygen availability (dissolved oxygen concentration and hypoxic/anoxic conditions), (iv) nutrient enrichment (nitrate, phosphate, and eutrophication indices), and (v) salinity gradients. For laboratory and mesocosm studies, temperature, pH, and nutrient concentrations were manipulated to match projected climate scenarios, including IPCC-based predictions for 2100, while dissolved oxygen was monitored continuously using optodes or Winkler titration. Field studies measured environmental gradients in situ using CTD profiling, water sampling, and sediment analysis.

Microzooplankton and protist communities were assessed using a combination of traditional microscopy, flow cytometry, and high-throughput sequencing (HTS). Microscopic enumeration allowed for detailed morphological classification, while HTS, particularly 18S rRNA amplicon sequencing, provided high-resolution identification of operational taxonomic units (OTUs) or amplicon sequence variants (ASVs). Bioinformatic pipelines, such as DADA2, were employed to distinguish true biological variants from sequencing errors, ensuring accurate diversity estimation. Sequencing reads were filtered, denoised, and taxonomically assigned using curated reference databases (e.g., PR2), allowing cross-study comparison of community composition and relative abundance.

Functional traits, including mixotrophy, cyst formation, and grazing rates, were integrated into analyses using trait databases and experimental measurements. Grazing experiments typically employed dilution or uptake assays with fluorescently labeled prey, while cyst formation and viability were quantified using microscopy and staining protocols. Network metrics and functional redundancy indices were linked to these traits to evaluate ecosystem-level implications of environmental stressors.

2.4. Statistical Analysis and Risk Assessment

Data were analyzed using R software (version 4.2.2) with the “metafor,” “vegan,” and “igraph” packages for meta-analysis, community ecology, and network analysis, respectively. Effect sizes from multiple studies were combined using weighted random-effects models, with weights inversely proportional to variance. Sensitivity, exposure, and adaptive capacity were integrated into a comprehensive risk assessment framework, following the methodology outlined by López-Abbate (2021) and Bindoff et al. (2019). Sensitivity was quantified as the magnitude of community response to a stressor, exposure was determined based on the intensity and frequency of stressor occurrence, and adaptive capacity incorporated evidence of functional redundancy, physiological plasticity, and behavioral traits.

To visualize ecosystem-level effects, meta-regression models examined the relationship between stressor intensity and community functional parameters (e.g., grazing, carbon export efficiency, cyst survival). Effect sizes were interpreted relative to ecological thresholds, such as the ability of communities to maintain trophic coupling, particulate organic carbon export, or species richness under stress. Risk categories (low, medium, high) were assigned based on the combined contribution of exposure, sensitivity, and adaptive traits, enabling direct comparison across stressors and habitats.

All statistical analyses were subjected to rigorous validation, including residual diagnostics, bootstrapping, and leave-one-out analyses to assess robustness. Data heterogeneity and variance partitioning were reported alongside effect sizes to provide transparency and reproducibility, in line with PubMed and PRISMA reporting standards. Graphical outputs, including forest plots, funnel plots, and network visualizations, were used to communicate results across scales, from species-specific responses to community-level interactions and ecosystem processes.

3. Results

3.1 Environmental Stress Responses in Microzooplankton Communities

The statistical analysis of the compiled data revealed significant patterns in the responses of microzooplankton and protist communities to environmental stressors, highlighting both general trends and taxa-specific sensitivities. Reported effect sizes describing the influence of major environmental stressors on microbial and microzooplankton community responses are summarized in Table 1.  Analysis of the effect sizes derived from the 72 studies included in the meta-analysis demonstrated substantial heterogeneity across studies, as indicated by Cochran’s Q-test (Q = 432.7, p < 0.001) and I² statistics exceeding 68%, reflecting the diversity of experimental designs, environmental contexts, and taxonomic focus. Despite this variability, clear trends emerged when data were aggregated according to stressor type, geographic region, and taxonomic grouping. Table 1 summarizes the weighted mean effect sizes and 95% confidence intervals for major stressors, illustrating that temperature increases and hypoxia consistently elicited negative responses in most microzooplankton taxa, whereas nutrient enrichment displayed a more complex pattern with both positive and negative outcomes depending on the baseline community composition and trophic strategy.

Forest plot analyses further clarified these patterns by visualizing the distribution of individual study effect sizes around the weighted means. Temperature-related stress exhibited a predominantly negative effect on ciliate abundance (Hedges’ g = −0.58; 95% CI: −0.72 to −0.44), whereas dinoflagellates, particularly mixotrophic species, displayed more variable responses (Hedges’ g = −0.22; 95% CI: −0.41 to −0.03). Similarly, dissolved oxygen reduction strongly impacted heterotrophic taxa (Hedges’ g = −0.65; 95% CI: −0.81 to −0.49), consistent with previous observations that oxygen limitation constrains respiration and grazing activity. In contrast, moderate nutrient enrichment increased overall abundance in some taxa, particularly small flagellates, suggesting that resource availability can temporarily offset stress-induced losses in productivity. These findings are reflected in Table 2, which categorizes effect sizes according to functional traits, highlighting that mixotrophy and cyst-forming ability confer resilience under fluctuating environmental conditions.

Funnel plots were used to assess potential publication bias, revealing a relatively symmetric distribution of studies around the pooled effect sizes, with Egger’s regression tests failing to detect significant asymmetry (p > 0.05). Potential publication bias and small-study effects were evaluated using funnel-plot analysis. This suggests that the observed patterns are not artifacts of selective reporting, increasing confidence in the robustness of the meta-analytic estimates. Sensitivity analyses, conducted by sequentially removing high-leverage studies, showed minimal changes in overall effect size estimates, indicating that no single study disproportionately influenced the results. This robustness reinforces the validity of conclusions drawn regarding stressor-specific responses and the relative importance of functional traits in mediating community resilience.

Network analyses of co-occurrence patterns revealed substantial reorganization under environmental stress. Metrics such as connectivity and modularity demonstrated that communities subjected to temperature and oxygen extremes exhibited reduced network cohesion, indicating a loss of interaction complexity. For example, highly connected nodes representing keystone taxa, such as large heterotrophic ciliates, were disproportionately affected by hypoxic conditions, resulting in decreased overall network robustness. Conversely, mixotrophic species and taxa capable of cyst formation often maintained their network positions, buffering the functional integrity of the community. These results provide empirical support for the functional redundancy hypothesis, where the presence of multiple species with overlapping ecological roles mitigates the impact of environmental perturbations on community-level processes.

Meta-regression analyses linking environmental stress intensity with community functional parameters indicated that changes in temperature and oxygen concentration accounted for a substantial proportion of variance in both abundance and trophic activity (R² = 0.46 for temperature, R² = 0.39 for oxygen). Nutrient enrichment and pH shifts explained smaller, though significant, proportions of variance (R² = 0.18 and 0.12, respectively). Importantly, interactive effects emerged, whereby combinations of stressors produced non-additive outcomes. For example, in mesocosm studies where both temperature and hypoxia were manipulated, ciliates exhibited significantly higher mortality than predicted by additive effects, indicating synergistic stress responses. Similarly, dinoflagellates demonstrated enhanced cyst formation under combined pH and nutrient stress, suggesting that adaptive mechanisms are invoked more strongly under multifactorial perturbation.

The statistical results also highlighted geographic and habitat-specific differences. Polar communities showed heightened sensitivity to temperature and oxygen fluctuations relative to temperate and tropical systems, likely reflecting limited thermal plasticity and lower baseline oxygen tolerance. In contrast, communities from eutrophic estuaries exhibited greater resilience to nutrient perturbations, possibly due to prior adaptation to variable nutrient loads. These regional patterns underscore the necessity of considering both local adaptation and environmental context when interpreting meta-analytic outcomes.

Incorporating functional traits into statistical models further refined predictions of community response. Taxa possessing mixotrophic capabilities consistently exhibited smaller negative effect sizes, while strictly heterotrophic taxa were disproportionately impacted by temperature and oxygen stress. Similarly, the presence of resting cysts correlated with reduced variability in abundance and higher survival probabilities under experimental stress, as reflected in Table 2. These trait-based patterns were robust across both field and laboratory studies, indicating that physiological plasticity and dormancy strategies are critical determinants of microzooplankton and protist resilience.

Overall, the integration of meta-analytic results, network metrics, and trait-based assessments provides a comprehensive understanding of how environmental stressors influence microzooplankton and protist communities. Figures 2 through 5 and Tables 1 to 4 collectively illustrate that while environmental perturbations generally exert negative pressures, community-level outcomes are moderated by functional redundancy, adaptive traits, and habitat-specific resilience mechanisms. The statistical analyses emphasize that ecological responses cannot be generalized across taxa

Table 1. Quantification of Environmental Stressor Effects on Microbial Communities. This table summarizes reported effect sizes describing the influence of diverse environmental stressors on microbial community structure, function, and persistence across aquatic and experimental systems. Effect sizes are expressed using statistical metrics or qualitative risk indicators, reflecting the magnitude and direction of biological responses.

Table 2. Study Precision Indicators for Microbial Ecology Effect Estimates. This table presents sample size and data context information used as proxies for study precision when evaluating variability, robustness, and potential bias in reported microbial ecology outcomes. These parameters support comparative synthesis and funnel-plot–based assessments in meta-analytic frameworks.

Figure 2. Forest Plot of Microzooplankton Responses to Environmental Stressors. This forest plot visualizes individual and pooled effect sizes describing microzooplankton and protist responses to major environmental stressors. Weighted mean estimates and confidence intervals highlight both overall trends and between-study heterogeneity.

Figure 3. Funnel Plot Assessing Publication Bias in Environmental Stressor Studies. This funnel plot depicts the relationship between study effect size and precision, enabling visual assessment of publication bias and small-study effects. The largely symmetric distribution supports the robustness of the meta-analytic findings.

Figure 4. Estimated Sample Sizes Across Studies. This horizontal dot plot compares estimated sample sizes from studies, highlighting variability in study design. Each black dot represents a study’s estimate, uniquely showing a confidence interval. The x-axis marks sample size estimates, and a vertical dashed line at zero provides a reference point.

or regions but require consideration of both biotic and abiotic context, reinforcing the importance of multifactorial experimental designs in capturing realistic ecosystem dynamics.

The findings highlight that predictive models of ecosystem responses should incorporate both direct stressor effects and indirect interactions mediated through trophic and network-level processes. By quantifying effect sizes, identifying trait-mediated resilience, and examining co-occurrence patterns, this analysis advances understanding of the mechanisms underlying microbial community stability under global change scenarios. Furthermore, the statistical rigor employed—through heterogeneity assessments, sensitivity analyses, and meta-regression—ensures that conclusions are robust, reproducible, and informative for both ecological theory and applied management of vulnerable marine and hypersaline ecosystems.

3.2 Interpretation and Discussion of the Funnel and Forest Plots

The analysis of forest and funnel plots provides crucial insights into both the magnitude and reliability of the compiled effect sizes, allowing a nuanced interpretation of patterns observed across studies. Forest plots (Figure 2) visualize the distribution of individual study effect sizes alongside pooled estimates, offering an immediate impression of heterogeneity and overall trends. Across the 72 studies included in the meta-analysis, forest plots revealed that while some stressors consistently produced strong negative effects on microzooplankton and protist communities, others exhibited greater variability, suggesting context-dependent responses. For instance, temperature stress consistently yielded negative effect sizes for heterotrophic ciliates, with most confidence intervals not overlapping zero, indicating robust statistical significance. In contrast, nutrient enrichment displayed both positive and negative effects across different studies, reflected by widely overlapping confidence intervals, which suggests that baseline nutrient levels, community composition, and local adaptation significantly modulate responses.

The forest plots also highlight the influence of study size and weight on the pooled estimates. Larger studies, often represented by broader boxes and narrower confidence intervals, dominated the meta-analytic outcomes, reinforcing the importance of high-quality, well-replicated studies in estimating true effect sizes. Conversely, smaller studies exhibited greater variability and contributed disproportionately to heterogeneity, as reflected in the high I² values. This pattern underscores the need for careful weighting in meta-analyses to avoid overinterpreting idiosyncratic results from small sample sizes. Study context and sample size information used to assess analytical precision and weighting in the meta-analysis are provided in Table 2, complements these visual insights by summarizing weighted mean effect sizes and highlighting taxa-specific differences, demonstrating that functional traits such as mixotrophy and cyst formation mediate resilience under environmental stress.

Funnel plots (Figures 3) were employed to assess potential publication bias and evaluate the symmetry of reported results around the pooled effect size. Symmetric funnel plots suggest minimal publication bias, indicating that studies reporting both positive and negative outcomes were included in the meta-analysis. In the current analysis, funnel plots were largely symmetric, and Egger’s regression tests failed to detect significant asymmetry (p > 0.05). This indicates that selective reporting of statistically significant results did not unduly influence the pooled estimates, enhancing confidence in the reliability of the observed patterns. Nonetheless, slight asymmetry observed in some subgroups, particularly for studies focusing on extreme hypoxia, may reflect genuine biological variability or the influence of smaller studies reporting extreme effect sizes.

The integration of forest and funnel plot interpretations allows for a deeper understanding of heterogeneity and its ecological implications. Forest plots provide clear evidence of taxa-specific and stressor-specific responses, while funnel plots ensure that these conclusions are not artifacts of selective reporting. For example, forest plots demonstrate that heterotrophic ciliates are consistently sensitive to temperature and oxygen stress, whereas dinoflagellates and mixotrophic taxa exhibit more variable responses, often dependent on local environmental conditions. Funnel plot analysis supports the validity of these observations by showing that extreme effect sizes are not systematically underrepresented, suggesting that the observed variability reflects real ecological phenomena rather than methodological bias. The estimated sample sizes from studies are shown in Figure 4.

Moreover, the combined interpretation of forest and funnel plots highlights the role of study quality and design in shaping observed outcomes. Studies with rigorous experimental controls and replication tend to cluster near the pooled effect sizes, whereas studies with limited replication or high variability in stressor intensity display wider confidence intervals. This pattern emphasizes the importance of experimental standardization in ecological meta-analyses, as inconsistencies in methodology can inflate heterogeneity and complicate the detection of genuine ecological patterns. Importantly, the funnel plots suggest that the meta-analytic conclusions are robust to potential small-study effects, reinforcing the reliability of observed trends.

The forest plots also provide insights into the magnitude of responses, illustrating that some stressors induce large negative impacts while others have more subtle or context-dependent effects. For instance, temperature and hypoxia consistently show substantial negative effect sizes across multiple taxa, highlighting the vulnerability of microzooplankton and protist communities to climate-related stressors. Nutrient enrichment, in contrast, produces variable responses, with some taxa showing positive growth, particularly under nutrient-limited baseline conditions, while others experience negative effects when nutrient inputs exceed ecological thresholds. This complexity is reflected in both the forest plots and the calculated heterogeneity metrics, reinforcing the importance of integrating ecological context and functional traits when interpreting meta-analytic results.

Finally, the interpretation of these plots underscores the value of combining quantitative and visual approaches in meta-analytic studies. Forest plots allow for direct assessment of individual study contributions, effect size magnitude, and confidence intervals, while funnel plots provide a diagnostic tool for detecting potential bias and assessing the reliability of pooled estimates. Together, they offer complementary perspectives: forest plots elucidate ecological patterns, and funnel plots validate the robustness of these patterns. By carefully examining both sets of plots, the analysis confirms that observed trends—such as the vulnerability of heterotrophic taxa to temperature and oxygen stress and the variable resilience of mixotrophic and cyst-forming taxa—are consistent, statistically robust, and ecologically meaningful.

In conclusion, the forest and funnel plot analyses collectively reinforce the credibility of the meta-analytic findings, demonstrating both clear patterns of stressor-specific responses and minimal evidence of publication bias. The visual and statistical integration of these plots supports the interpretation that microzooplankton and protist communities exhibit complex, trait-mediated responses to environmental perturbations, with significant implications for predicting ecosystem-level impacts under changing environmental conditions.

4. Discussion

The present meta-analysis elucidates the complex responses of microzooplankton, protists, and associated microbial communities in marine and hypersaline ecosystems under a variety of environmental stressors, integrating data across multiple studies using robust statistical methods, including forest and funnel plot analyses. The results highlight that stressors such as temperature fluctuations, hypoxia, nutrient enrichment, and salinity extremes drive heterogeneous responses across taxa, underscoring the importance of ecological context and species-specific traits in shaping community resilience and ecosystem functioning (Buitenhuis et al., 2010; Fenchel, 2008; Mitra et al., 2014). Standardized quantitative effect sizes describing microbial responses to environmental stressors are summarized in Table 3.

Forest plots reveal clear patterns in taxa-specific vulnerabilities. Heterotrophic ciliates consistently exhibited negative effect sizes under elevated temperatures and low-oxygen conditions, reflecting their sensitivity to oxygen limitation and thermal stress (Breitburg et al., 2018; López-Abbate, 2021). This aligns with observations that declining oxygen in global oceans, exacerbated by anthropogenic impacts, directly affects microzooplankton-mediated biogeochemical fluxes, potentially altering the efficiency of the microbial loop (Fenchel, 2008; Buitenhuis et al., 2010). Mixotrophic taxa, in contrast, displayed variable responses, which can be attributed to their flexible nutritional strategies allowing them to switch between autotrophy and heterotrophy depending on environmental conditions (Burkholder et al., 2008; Mitra et al., 2014). This plasticity may confer a competitive advantage under moderate stress, but extreme conditions, particularly hypoxia or prolonged thermal stress, still result in negative population responses, as indicated by effect sizes approaching those of purely heterotrophic species.

Nutrient enrichment presented a context-dependent pattern. While some studies reported enhanced growth of specific protistan taxa under moderate nutrient inputs, others indicated inhibitory effects at higher concentrations, likely due to shifts in community composition and competition with opportunistic bacterial populations (Heisler et al., 2008; Finke et al., 2017). Such dual responses highlight the complex interplay between bottom-up controls and species-specific tolerances in driving community structure, emphasizing that nutrient dynamics alone cannot fully predict ecological outcomes without considering co-occurring stressors (Edwards et al., 2006; López-Abbate, 2021).

The robustness of the meta-analytic findings is reinforced by funnel plot analyses, which suggest minimal publication bias, indicating that both positive and negative effect sizes are well-represented across the literature (Callahan et al., 2016). Study-level context and sample size information used to support assessments of robustness and potential bias are presented in Table 4. The influence of study precision and sample size on effect-size variability is further illustrated in Figure 5. Slight asymmetries observed in subsets of hypoxia-focused studies may reflect genuine biological heterogeneity, as small experimental studies often capture localized extreme responses that are ecologically relevant but statistically distinct from larger-scale patterns (Breitburg et al., 2018; Forster et al., 2021). This observation underscores the need for careful interpretation of meta-analytic outcomes, balancing statistical significance with ecological realism.

Salinity extremes, particularly in hypersaline anoxic basins, emerge as a critical factor shaping microbial and protistan diversity (Barone et al., 2019; Gostinčar et al., 2011). Fungal and protistan communities in these environments demonstrate remarkable adaptations, including osmoprotectant accumulation and specialized resting stages, which allow survival under conditions that would be lethal to most planktonic taxa (Barone et al., 2019; Lundholm et al., 2011). Such adaptations contribute to the persistence of functional processes, even in highly saline and anoxic microhabitats, supporting the concept of resilience in extreme ecosystems.

The observed heterogeneity across studies is also linked to methodological variation, particularly regarding experimental design, sequencing approaches, and taxonomic resolution (De Vargas et al., 2015; Callahan et al., 2016). High-throughput amplicon sequencing methods have improved detection of rare taxa, revealing previously unrecognized diversity and functional potential, yet inter-study differences in sampling depth, primer selection, and bioinformatic pipelines introduce variability that contributes to heterogeneity in effect sizes. This underscores the importance of standardized protocols and transparent reporting in advancing meta-analytic synthesis in microbial ecology.

Ecological consequences of these findings extend beyond community composition to broader ecosystem functions, such as carbon cycling, nutrient remineralization, and the biological carbon pump. Mixotrophic protists, in particular, play pivotal roles in carbon transfer from the microbial loop to higher trophic levels, with implications for global biogeochemical cycles under climate stressors (Mitra et al., 2014; Bindoff et al., 2019). Declines in microzooplankton biomass due to thermal or hypoxic stress can disrupt these fluxes, reduce the efficiency of carbon sequestration and potentially altering feedbacks to climate change (Field et al., 2006; Jonkers et al., 2019). Similarly, shifts in planktonic foraminifera and calcifying protists have been linked to historical warming trends and ocean acidification, providing empirical evidence that current environmental changes are inducing community shifts analogous to past climatic perturbations (Moy et al., 2009; Field et al., 2006).

The meta-analysis also demonstrates that community resilience is mediated by both intrinsic traits and network-level interactions. Studies on climate-stressed lake protistan networks suggest that robust interactions among tolerant taxa can buffer ecosystem function against stressors, while loss of key species or functional redundancy reduces resilience and increases vulnerability to regime shifts (Forster et al., 2021; López-Abbate, 2021). These findings highlight the need to integrate network analyses and trait-based approaches in future studies to predict ecosystem-level responses to global change more accurately.

The results further emphasize the importance of long-term monitoring to capture delayed and cumulative effects. Resting stages and dormant cysts allow some taxa to persist across unfavorable periods, but their eventual germination depends on the restoration of suitable environmental conditions (Lundholm et al., 2011; López-Abbate, 2021). Consequently, short-term studies may underestimate the potential for recovery or misrepresent the timing of community shifts, suggesting that meta-analyses incorporating multi-year datasets provide a more accurate reflection of ecological dynamics.

Finally, the observed patterns underscore the

Table 3. Standardized Effect Sizes of Environmental Stressors on Microbial Responses. This table compiles standardized effect sizes reported across studies assessing the impact of environmental stressors on microbial community structure and survival. Effect sizes are expressed using comparable quantitative metrics, facilitating cross-study synthesis and inclusion in meta-analytic models.

Table 4. Study Context and Sample Size Parameters Used for Precision Assessment. This table presents study-level contextual information and sample sizes used as indicators of precision in microbial ecology research. These parameters support assessments of result robustness, heterogeneity, and potential bias when interpreting effect sizes across studies.

Figure 5. Funnel Plot–Based Evaluation of Study Precision and Analytical Robustness. This figure complements forest-plot analyses by illustrating how study precision and sample size influence effect-size variability. It supports sensitivity analyses and reinforces confidence in the stability of pooled estimates.

interconnectedness of anthropogenic pressures and natural stressors. Climate-driven changes in temperature, oxygen availability, and salinity interact with nutrient enrichment and pollution to create compounded stress scenarios, often exceeding the tolerance thresholds of sensitive taxa (Breitburg et al., 2018; Bindoff et al., 2019). Understanding these interactions is essential for developing effective management strategies aimed at preserving microbial diversity and maintaining ecosystem function under global change scenarios.

In summary, this meta-analysis integrates diverse experimental and observational datasets to reveal consistent patterns in taxa-specific responses, functional trait mediation, and ecosystem implications. Heterotrophic ciliates are particularly sensitive to thermal and hypoxic stress, while mixotrophic protists exhibit context-dependent resilience. Nutrient enrichment has variable effects contingent on baseline conditions and co-occurring stressors, and extreme salinity selects for highly adapted fungal and protistan taxa. The integration of forest and funnel plot analyses confirms that these findings are statistically robust and ecologically meaningful, providing a framework for predicting the consequences of environmental change on microzooplankton-mediated ecosystem processes.

5. Limitations

Despite the comprehensive nature of this meta-analysis, several limitations must be acknowledged. First, methodological heterogeneity across studies, including differences in sampling strategies, sequencing platforms, primer sets, and bioinformatic pipelines, may introduce variability in taxonomic resolution and abundance estimates, potentially influencing effect size calculations (Callahan et al., 2016; De Vargas et al., 2015). Second, the majority of studies focus on short-term experimental manipulations or localized field surveys, which may not fully capture long-term, cumulative, or seasonal dynamics of protistan and microbial communities, including delayed responses mediated by resting stages or cyst germination (Lundholm et al., 2011; López-Abbate, 2021). Third, environmental stressors are often studied in isolation, whereas in natural ecosystems, multiple stressors interact synergistically or antagonistically, complicating extrapolation to real-world scenarios (Breitburg et al., 2018; Bindoff et al., 2019). Additionally, publication bias toward significant results, although minimal according to funnel plot analyses, cannot be entirely excluded, and rare taxa or understudied habitats remain underrepresented (Callahan et al., 2016). Finally, functional inferences are primarily derived from taxonomic data, and direct measurements of ecosystem processes such as carbon flux, nutrient cycling, or energy transfer were limited, potentially constraining mechanistic interpretations.

6. Conclusion

This meta-analysis demonstrates that environmental stressors such as hypoxia, temperature fluctuations, nutrient enrichment, and salinity extremes differentially affect protistan and microbial communities. Heterotrophic taxa are generally sensitive, while mixotrophs and extremophiles exhibit adaptive resilience. These changes have significant implications for ecosystem functioning, including carbon cycling and nutrient fluxes. Integrating trait-based, network, and long-term monitoring approaches is essential to predict ecological outcomes under global change, guiding conservation and management strategies for sustaining microbial biodiversity and ecosystem resilience.

Author Contributions

M.A.R. conceived the study, conducted literature review and data interpretation, and drafted the manuscript. S.H. contributed to data analysis, interpretation of marine ecological responses, and manuscript editing. B.B. participated in literature screening, meta-analytic evaluation, and preparation of figures and tables. M.I. assisted in data collection, methodological assessment, and critical revision of the manuscript. S.A.B. contributed to ecological network interpretation, resilience analysis, and manuscript review. A.H. supervised the overall study, finalized the manuscript, and approved the final version for publication.

References