CoChat
Sign Up
Cited by 15Freshwater macroinvertebrate diversity and ecologyAquatic Ecosystems and Phytoplankton DynamicsFish Ecology and Management Studies

A meta-analysis of environmental responses to freshwater ecosystem restoration in China (1987–2018)

Fu, Hong, Xu, Jun, Zhang, Huan, García Molinos, Jorge, Zhang, Min, Klaar, Megan, Brown, Lee E.·2023·Environmental Pollution
DOI 10.1016/j.envpol.2022.120589
Open this paper in CoChat — free account

Abstract

Understanding how abiotic and biotic components respond to aquatic ecosystem restoration is pivotal for sustainable development in the face of economic development and global environmental change. However, the post-restoration monitoring and evaluation of aquatic ecosystems across large spatial and temporal scales is underfunded or not well documented, especially outside of Europe and North America. We present a meta-analysis of abiotic and biotic indices to quantify post-restoration (2 months-13 years) effects from reported aquatic restoration projects throughout the China-mainland, incorporating 39 lentic and 36 lotic ecosystems. Decreases in dissolved nutrients (total nitrogen, ammonia nitrogen and total phosphorus) post-restoration were rapid, but tended to slow down after about 9.3 years. Response ratios summarizing biodiversity responses (incorporating phytoplankton, invertebrates, vascular plants, fish and birds) typically lagged behind abiotic changes, suggesting longer timescales are needed for biotic indices to recover. Time since restoration interacted with lentic project size showing that, even with the same proportional efforts of restoration, larger lentic ecosystems responded much more slowly than smaller ones. Spatial heterogeneity, reflecting the effects of different restoration approaches (e.g., sewage interception, polluted sediment dredging, artificial wetlands, etc.), had a significantly stronger effect on biotic than abiotic indices, particularly in rivers compared to standing waters. This reflects the complexity of fluvial ecosystem dynamics and hints at a limitation in the reinstatement of ecological processes in these systems to overcome issues such as dispersal limitations. Overall, the different timelines and processes by which abiotic and biotic indices recover after restoration should be taken into account when defining restoration targets and monitoring programs. Our study illustrates the value of long-term aquatic ecosystem monitoring, especially in China given the scale and magnitude of ongoing restoration investments in the country.

AI snapshot

Written from the paper’s full text.

TL;DR

This meta-analysis of 78 Chinese freshwater restoration studies (39 lentic, 36 lotic; 157 sites; 1653 records) found generally positive abiotic and biotic responses, but biotic indices lagged behind abiotic recovery. Lentic systems responded more strongly than lotic, smaller lentic projects recovered faster, and biotic responses were more variable, especially in rivers.

Key findings

  • Restoration improved most assessed indices: NH₄⁺-N, TN, TP and BOD₅ response ratios were positive (all p < 0.001), and fish, invertebrates, phytoplankton and vascular plants were significantly higher after restoration (all p < 0.05); birds did not differ significantly.
  • Abiotic nutrient responses peaked around 8–9 years in lentic ecosystems, and the overall abiotic response peaked at 9.3 years, while biotic response ratios were still rising by the end of monitoring (n = 1130, marginal R² = 0.11, p < 0.05); lotic monitoring was ≤9 years, so peaks were often not reached.
  • Lentic ecosystems had significantly higher positive responses than lotic ecosystems (n = 1653, marginal R² = 0.10, p < 0.05), and smaller lentic project size was associated with higher response ratios (n = 1130, marginal R² = 0.14, p < 0.01), with a significant time × lentic project size interaction (p < 0.05); no such interaction was detected for lotic systems (n = 505, p = 0.29).
  • Biotic response ratios were significantly more variable than abiotic ones overall (CV = 0.23 ± 0.03 vs. 0.18 ± 0.05, p < 0.001), and especially in lotic ecosystems (CV of Δr biotic = 0.90 ± 0.39 vs. abiotic = 0.40 ± 0.08); no significant AAT10 climate-zone effect was found.

Methods

  • Systematic literature search in CNKI up to December 19, 2019, plus an equivalent ISI Web of Science search; inclusion required quantitative before/after data over at least one month, stated restoration start/end dates, and freshwater restoration. The final dataset comprised 78 studies (74 CNKI, 4 Web of Science), 36 lotic and 39 lentic systems, 157 monitored sites, and 1653 records, with monitoring from 2 months to 13 years after restoration (mean 3.69 ± 3.01 years).
  • Restoration effects were quantified as response ratios (Δr) comparing after-restoration values with before-restoration or local degraded reference values; signs were reversed for nutrients, BOD₅, and phytoplankton/Oligochaeta density so positive values indicated improvement.
  • Analyses used Wilcoxon signed-rank tests, linear mixed models (models 1–7) with AIC-based model selection, Moran’s tests for autocorrelation, Kruskal-Wallis tests and coefficients of variation; predictors included index category, start date, monitored years after restoration (quadratic term), ecosystem type, AAT10, and project size, with random effects for ecosystem ID, start date, index subcategory, and random slope dt|sites. All analyses used R 4.0.1.

Limitations

  • Stated: post-restoration monitoring is often underfunded or poorly documented; limited sample size prevented separate models for birds, fish, vascular plants, and richness/diversity data; lotic monitoring was ≤9 years, so many response peaks were not reached; available data were heterogeneous and stricter monitoring/reporting protocols are needed.
  • Stated: almost all projects were synthetic ecological restoration schemes combining measures such as sewage interception, sediment dredging, artificial wetlands, macrophyte reintroduction, fishing exclusion and riparian buffers, and individual restoration measures had too few published studies for a focused meta-analysis; >60% of numerical data were extracted from figures.
  • Evident from the text: the lack of a significant climate-zone effect may reflect the diversity of ecosystems and indices combined, and the use of local species reintroductions that blur climatic signals; the meta-analysis also combines heterogeneous restoration schemes and monitoring protocols, limiting attribution of outcomes to specific measures.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/j.envpol.2022.120589

Cites

49 of 49 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

10 of 28 cited sources checked · 1 supported · 0 flagged · accuracy not yet measured

  • Premisecitation
    § 4. Discussion · p. 8

    Organisms with different niches are influenced by physicochemical and biological factors as well as dispersal success in more complex ways, leading to greater stochasticity.

    “organisms with different niches are influenced by physicochemical and biological factors as well as dispersal success in more complex ways, leading to greater stochasticity (Cid et al., 2021; Thompson and Town­ send, 2006).”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites Cid et al., 2021 source

      The Abstract (page unavailable) says that local niche selection and biotic interactions interact with regional dispersal to shape populations and communities. The meta-system theory section (page unavailable) likewise discusses abiotic conditions, other species, and constraints on dispersal. Neither passage says these interactions lead to greater stochasticity, so that part of the claim is unsupported. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites Thompson and Townsend, 2006 source

      The Results (page unavailable) show that local ecological factors and spatial distance jointly predicted stream-invertebrate community similarity, with effects differing by dispersal ability. The Discussion (page unavailable) also says trophic groupings and species traits affect the relative importance of niche and neutral factors. The paper does not establish that these interactions lead to greater stochasticity, so that part of the claim is unsupported. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Premisecitation
    § 4. Discussion · p. 8

    The successful restoration rate of lakes from eutrophication and acidification was higher than most rivers.

    “Verdonschot et al. (2013) who qualitatively concluded that the successful restoration rate of lakes from eutrophication and acidification was higher than most rivers.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Verdonschot et al., 2013 source

      The supplied passages labeled Abstract and “3 Finished” contain bibliographic details and keywords, not the review’s findings. Further searches for eutrophication, acidification, and comparative restoration success returned no passage addressing whether lakes had a higher successful restoration rate than most rivers. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
  • Interpretationcitation
    § 4. Discussion · p. 7

    The response lag for biotic indices likely relates to dispersal and establishment limitations.

    “In contrast, the response lag for biotic indices likely relates to dispersal and establish­ ment limitations which are common, and several recent reviews of metacommunity theory and practice in freshwaters have therefore advocated for the potential reintroduction of aquatic assemblages (Cid et al., 2021; Patrick et al., 2021).”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites Cid et al., 2021 source

      In “Integrating regional-scale processes in river conservation, restoration, and monitoring” (page unavailable), the paper says dispersal limitation can affect biomonitoring and that fragmentation can impede recolonization of restored sites. It does not specifically attribute a temporal lag in biotic indices to dispersal and establishment limitations, or advocate reintroducing aquatic assemblages. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Patrick et al., 2021 source

      Section 3.2 (Restoration; page unavailable) says dispersal constraints and barriers to recolonization frequently prevent a biological response to habitat restoration, and that reintroducing whole communities has been considered. It does not specifically establish a response lag for biotic indices or identify establishment limitations as its cause. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
  • Findingown data
    § 3.1. Overall response of abiotic/biotic indices after restoration · p. 4

    The improvement of aquatic ecosystems, denoted by positive response ratios of abiotic and biotic indices, increased with time elapsed since restoration.

    “The improvement of aquatic ecosystems (denoted by positive response ratio of abiotic and biotic indices) increased with time elapsed since restoration (Fig. 3).”
    Quote in the paperSubstantial claim
  • Findingown data
    § 3.1. Overall response of abiotic/biotic indices after restoration · p. 4

    For lotic ecosystems, the limited duration of monitoring (≤ 9 years) meant the peaks of the response ratio for abiotic and biotic indices were not reached in many instances.

    “Nonetheless, the limited duration of the monitored years after restoration for lotic ecosystems ( ≤ 9 years, Fig. 3 c) meant the peaks of the response ratio for the abiotic and biotic indices were not reached in many instances, highlighting the need for longer-term monitoring efforts.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Observed post-restoration changes in abiotic and biotic indices are attributable to the restoration interventions rather than to other environmental or temporal factors.
  • May combine two claims. 17
  • Classified as probably not a substantial claim. 11

  • May combine two claims.

    Study design and conductcitation
    § 2.2. Data extraction · p. 3

    Where numerical data were not provided, data were extracted from figures using Graph digitizer software, which has been proven reliable.

    “Where numerical data were not provided in a publication, data were extracted from the figures ( &gt; 60% publications) using the Graph digi­ tizer software (Digitizelt, version 2.5, https://www.digitizeit.de). This software has been used widely in meta-analysis studies (Rasheduzzaman et al., 2020; Zhang et al., 2017), and proven to be reliable in extracting data from figures with high level of con”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Rasheduzzaman et al., 2020 source
    • cites Zhang et al., 2017 source

      In Material and Methods, under “Data collection and database for meta-analysis” (page not provided), Zhang et al. say they extracted data from figures using graph-digitizing software when raw data were unavailable. They used GetData Graph Digitizer, not the cited sentence’s DigitizeIt version 2.5, and the passage does not establish that the software was widely used or proven reliable. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites Rakap et al., 2016 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation
    § 2.4. Statistical analyses · p. 4

    Model residuals were tested for compliance with model assumptions and for spatial and temporal autocorrelation with Moran's tests.

    “Model residuals were tested for compliance with model assumptions (Crawley, 2002), and spatial and temporal autocorrelation with Moran ' s tests (Birk et al., 2020).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Crawley, 2002

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Birk et al., 2020 source

  • May combine two claims.

    Interpretationcitation
    § 4. Discussion · p. 7

    Phytoplankton concentrations were subsequently reduced significantly, linked to the decline of TP concentrations in water.

    “As a consequence, concentrations of phytoplankton were subsequently reduced significantly, linked to the decline of TP concentrations in water and probably accompanied by zooplankton and fish community struc­ ture change (Jeppesen et al., 2005).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Jeppesen et al., 2005 source

      The ‘Phytoplankton biomass and community structure’ section reports declining biomass in most shallow and deep lakes, consistent with the reported association with lower TP. It also labels the overall biomass-decline pattern as not significant (NS), so the claim that phytoplankton was reduced significantly overstates the finding. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 4. Discussion · p. 8

    An optimum annual accumulated mean daily temperature above 10°C is considered to enable more successful biodiversity recovery.

    “Whilst an optimum annual accumulated mean daily temperature above 10 ◦ C is considered to enable more successful biodiversity recovery (Dong et al., 2009)”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Dong et al., 2009 source

      The Introduction describes annual accumulated temperature above 10°C as an indicator of thermal conditions for crop growth, and §4.2 discusses its effects on cropping systems. Neither passage addresses an optimum temperature for biodiversity recovery; searches for biodiversity recovery and species diversity found no such finding in our copy. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 4. Discussion · p. 8

    In extremely warm environments, the stimulation of algal growth extends the duration of eutrophication and algal blooms, making conditions less favorable for ecosystem recovery despite attempts at restoration.

    “in extremely warm environ­ ments, the stimulation of algal growth extends the duration of eutro­ phication and algal blooms (Nazari-Sharabian et al., 2018; Xiong et al., 2016), thus making conditions less favorable for ecosystem recover despite attempts at restoration.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Nazari-Sharabian et al., 2018 source

      The paper’s A. Temperature section says warmer water accelerates cyanobacterial growth and could produce earlier, more extended periods of potential algal blooms. It also says internal nutrient release could keep water eutrophic even when external nutrient sources are restrained. It does not establish that warming extends the duration of eutrophication itself or assess ecosystem recovery after restoration; the claim also states the bloom effect more definitively than the review does. Page numbers are unavailable in our copy. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites Xiong et al., 2016 source

  • May combine two claims.

    Findingown data
    § 3.1. Overall response of abiotic/biotic indices after restoration · p. 4

    Biotic indices for fish, invertebrates, phytoplankton and vascular plants were significantly higher after restoration (all p < 0.05).

    “Biotic indices for fish, invertebrates, phytoplankton and vascular plants were significantly higher after restoration, as illustrated by generally positive response ratios (all p &lt; 0.05, Fig. 2).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § A B S T R A C T · p. 1

    Decreases in dissolved nutrients (total nitrogen, ammonia nitrogen and total phosphorus) post-restoration were rapid, but tended to slow down after about 9.3 years.

    “Decreases in dissolved nutrients (total nitrogen, ammonia nitrogen and total phosphorus) post-restoration were rapid, but tended to slow down after about 9.3 years.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Observed post-restoration changes in abiotic and biotic indices are attributable to the restoration interventions rather than to other environmental or temporal factors.

  • May combine two claims.

    Findingown data
    § 3.1. Overall response of abiotic/biotic indices after restoration · p. 4

    There was a significant interaction between time since restoration and lentic project size on the response ratio of all abiotic and biotic indices, with effects much weaker for larger project size.

    “A significant interaction between the monitored years after restora­ tion and the size of lentic project was evident for the response ratio of all the abiotic and biotic indices (n = 1130, p &lt; 0.05, Fig. 5).”
    • Also stated in: § 3.1. Overall response of abiotic/biotic indices after restoration · p. 4 · plain
      “For example, higher abiotic and biotic index responses were associated with time after restoration, but these effects were much weaker for larger project size (Fig. 5), irrespective of the number of monitored years elapsed since”
    • Also stated in: § A B S T R A C T · p. 1 · plain
      “Time since restoration interacted with lentic project size showing that, even with the same proportional efforts of restoration, larger lentic ecosystems responded much more slowly than smaller ones.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data
    § 3.1. Overall response of abiotic/biotic indices after restoration · p. 4

    For lentic ecosystems, the response ratio of abiotic indices reached its recovery peak 9.3 years after restoration, while the response ratio of biotic indices was still rising by the end of the monitored period.

    “For lentic ecosystems, the response ratio of abiotic indices reached its recovery peak 9.3 years from restoration, the response ratio of biotic indices was still rising by the end of the monitored period (n = 1130, marginal R 2 = 0.11, p &lt; 0.05, Fig. 3 b).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Observed post-restoration changes in abiotic and biotic indices are attributable to the restoration interventions rather than to other environmental or temporal factors.

  • May combine two claims.

    Findingnone given
    § 4. Discussion · p. 7

    Common projects included targeting pollution-sinks such as removal of contaminated sediment, followed by submerged macrophyte reintroduction and riparian buffer zone planting.

    “Additionally, common projects included tar­ geting pollution-sinks such as removal of contaminated sediment, fol­ lowed by submerged macrophyte reintroduction and riparian buffer zone planting.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation
    § 4. Discussion · p. 6

    Restoration projects effectively improved the abiotic and biotic conditions of aquatic ecosystems over time, except for birds.

    “our results demonstrate that restoration projects effectively improved the abiotic and biotic conditions of aquatic ecosystems over time (Fu et al., 2021; Huang et al., 2019), except for birds.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The included studies are representative of freshwater ecosystem restoration in China.

  • May combine two claims.

    Study design and conductown data
    § 2.4. Statistical analyses · p. 3

    Visual inspection of frequency histograms showed all response ratios followed non-normal distributions, so Wilcoxon signed rank tests were used.

    “Visual inspection of frequency histograms showed all response ratios of abiotic ( Δ r NH + 4 -N, Δ rTN, Δ rP, Δ r BOD) and biotic indices ( Δ r biodiversity of birds, fish, invertebrates, phytoplankton and vascular plants) followed non-normal distributions. Therefore, we used Wilcoxon signed rank tests to examine whether the median response ratios of ecosystem indices were”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data
    § 2.4. Statistical analyses · p. 4

    Only abundance/biomass data were used for models 4 to 7 due to limited richness/diversity data of each organism group.

    “Only abundance/ biomass data were used for model 4 to model 7, due to the limited richness/diversity data of each organism group.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation
    § A B S T R A C T · p. 2

    The wider extent of improvement in China's inland waters remains unknown because Zhou et al. (2017) did not consider changes in biological status, and other studies incorporating biotic indices have focused on single lakes or specific regions of China.

    “However, the wider extent of improvement remains unknown as the study did not consider any changes in biological status, whilst other studies incorpo­ rating biotic indices (e.g. plants, fish, invertebrates, etc.) have focused either on single lake (e.g. (Bai et al., 2020)) or specific regions of China (e.g. Taihu basin (Fu et al., 2021)).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Bai et al., 2020 source
    • cites Fu et al., 2021

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given
    § 2.4. Statistical analyses · p. 4

    Model selection was performed to find the most parsimonious model based on AIC.

    “For each model structure (Table 1), we performed model selection to search for the most parsimonious model based on the Akaike ' s Information Criterion (AIC).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning
    § 2.2. Data extraction · p. 3

    Several studies reported data from the same ecosystem but for different time periods; these were combined to avoid pseudo-replication.

    “Several studies reported data from the same ecosystem but for different time periods. These were combined to avoid pseudo replication.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning
    § A B S T R A C T · p. 1

    The stronger effect of spatial heterogeneity on biotic indices in rivers reflects the complexity of fluvial ecosystem dynamics and indicates a limitation in reinstating ecological processes in these systems to overcome dispersal limitations.

    “This reflects the complexity of fluvial ecosystem dynamics and hints at a limitation in the reinstatement of ecological processes in these systems to overcome issues such as dispersal limitations.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Observed post-restoration changes in abiotic and biotic indices are attributable to the restoration interventions rather than to other environmental or temporal factors.
    • Rests on an assumption: The included studies are representative of freshwater restoration projects across China mainland.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § A B S T R A C T · p. 2

    In a synthesis of river restoration projects across the USA, Bernhardt et al. (2005) highlighted the lack of long-term monitoring as a major impediment to evaluating restoration success.

    “In a synthesis of river restoration pro­ jects across the USA, Bernhardt et al. (2005) highlighted the lack of long-term monitoring as a major impediment to evaluating restoration success.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Bernhardt et al., 2005 source

  • Classified as probably not a substantial claim.

    Premisecitation
    § A B S T R A C T · p. 2

    Lotic and lentic elements of a watershed may strongly vary in the outcomes of restoration projects, depending on their specific hydrologic and biological conditions.

    “Therefore, lotic and lentic ele­ ments of a watershed may strongly vary in the outcomes of restoration projects, depending on their specific hydrologic and biological condi­ tions (Levi and McIntyre, 2020).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Levi and McIntyre, 2020 source

  • May combine two claims.

    Study design and conductown data
    § 2.2. Data extraction · p. 2

    The diversity of schemes incorporated into the analysis allows generalizations to be made about restoration effects, but individual restoration measures could not be analyzed due to too few studies.

    “The diversity of schemes incorporated into the analysis allows generalizations to be made about restoration effects, but for the feature of individual restoration measures (e.g., investments, amounts, etc.), the number of published studies typically remains too low to develop a more focused meta-analysis.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given
    § 4. Discussion · p. 8

    Augmented dispersal by incorporating species reintroduction of local plants and animals that adapt to local climate conditions will significantly blur the boundaries between natural, climatically driven processes and recovery from human modifications.

    “the practice of augmented dispersal by incorpo­ rating species reintroduction of local plants and animals that then adapt to the local climate conditions will significantly blur the boundaries between natural, climatically driven processes and recovery from human modifications.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 4. Discussion · p. 8

    High variability of the response ratio of fish, macroinvertebrates and aquatic macrophytes after river restoration was reported.

    “Kail et al. (2015) also reported the high variability of the response ratio of fish, macroinvertebrates and aquatic macrophytes after river restoration (without incorporation with abiotic indices)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kail et al., 2015 source

  • May combine two claims.

    Interpretationreasoning
    § 4. Discussion · p. 7

    Water quality improvements were typically rapid with pollutant loads reduced quickly.

    “As a consequence, water quality improvements were typically rapid with pollutant loads reduced quickly.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation
    § 4. Discussion · p. 8

    Many factors (e.g., organism group, restoration measures) can contribute to the different variability range of response ratio.

    “and indicated that many factors (e.g., organism group, restoration measures) can contribute to the different variability range of response ratio.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kail et al., 2015 source

  • May combine two claims.

    Premisecitation
    § 4. Discussion · p. 8

    Sewage interception and denitrification reduced nitrogen by >70%, but had less impact on phosphorus, illustrating the important role of sedimentary cycling.

    “Here, sewage interception and denitrification reduced nitrogen in by &gt; 70%, but had less impact on phosphorus illustrating the important role of sedimentary cycling.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Li et al., 2022

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation
    § 4. Discussion · p. 7

    Time lags of recovery of different species are highly variable because of different life-span and fecundity, with short-lived species expected to display short time-lags.

    “Finally, time lags of recovery of different species are highly variable because of different life-span and fecundity, with short-lived species expected to display short time-lags (Watts et al., 2020).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Watts et al., 2020 source
26 of 26

  • The paper relies on this without support.

    Observed post-restoration changes in abiotic and biotic indices are attributable to the restoration interventions rather than to other environmental or temporal factors.

    “Restoration works were found to be efficient at recovering fresh­ water ecosystems from their initial degraded condition, as shown by their significant effect on almost all the assessed abiotic and biotic indices except for birds (Fig. 2, Fig. S2).”

  • The paper relies on this without support.

    The included studies are representative of freshwater restoration projects across China mainland.

    “Here, we present a comprehensive national-level (China mainland) meta-analysis of the temporal trajectory of different abiotic (biological oxygen demand, nutrients like nitrogen and phosphorus) and biotic (species richness/diversity and abundance/biomass) indices, used as indicators of restoration effects.”

  • The paper relies on this without support.

    The monitored years after restoration in the included studies accurately reflect the true time since restoration.

    “the response ratio of abiotic indices declined in lentic ecosystems after about 9.3 years post-restoration”

  • The paper relies on this without support.

    The included studies are representative of freshwater ecosystem restoration in China.

    “Our study of long-term (up to 13 years) freshwater ecosystem responses following the restoration at a large spatial scale (China mainland) showed that”

  • The paper relies on this without support.

    The climatic zones (AAT10) adequately represent climatic influences on restoration outcomes.

    “No significant influence of different climatic zones (AAT10) was detected on aquatic ecosystem restoration effects in our study”
5 of 5

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Claim of a >10-year lag in lotic ecosystems exceeds the monitoring period

    Minor

    The discussion says biotic responses lagged for more than 10 years in both lentic and lotic ecosystems, but the paper states that lotic ecosystems were monitored for no more than 9 years. Its data therefore cannot establish the claimed duration for lotic ecosystems.

1 of 1

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Fu, H., Xu, J., Zhang, H., García Molinos, J., Zhang, M., Klaar, M., & Brown, L. E.
Affiliations
  • University of Leeds (GB)
  • Chinese Academy of Sciences (CN)
  • Huazhong Agricultural University (CN)
Corresponding author
  • Xu, Jun
  • Zhang, Huan
ORCID
Venue
Environmental Pollution
Publisher
Elsevier BV
Published
2023-01
Volume
316
Issue
Pt 2
Pages
120589
ISSN
0269-7491 (ISSN-L)
DOI
10.1016/j.envpol.2022.120589
OpenAlex
W4308367360
Cited by
15
References
46
Citations by year
2026: 6 · 2025: 4 · 2024: 3 · 2023: 2
FWCI
0.20
Type
journal article
Language
en
Open access
closed
Licence
Locations
  • Environmental Pollution (publishedVersion)
  • PubMed (publishedVersion)
Funders
Japan Society for the Promotion of Science (19H04314)
Topics
  • Physical Sciences › Environmental Science › Ecology › Freshwater macroinvertebrate diversity and ecology
  • Physical Sciences › Environmental Science › Environmental Chemistry › Aquatic Ecosystems and Phytoplankton Dynamics
  • Physical Sciences › Environmental Science › Nature and Landscape Conservation › Fish Ecology and Management Studies
Keywords
freshwater restoration, nutrient recovery, dissolved nitrogen, dissolved phosphorus, biodiversity recovery, spatial heterogeneity, sewage interception, sediment dredging, constructed wetlands, long-term monitoring, China
MeSH terms
Animals, China, Environmental Monitoring, Fresh Water, Nitrogen, Ecosystem, Biodiversity, Rivers

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-16.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-16.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review