Environmental Evidence最新文献

{"title":"What evidence exists on how biodiversity is affected by the adoption of carbon footprint-reducing agricultural practices? A systematic map.","authors":"Stuart Rowlands, Julia Casperd, Michael R F Lee, Scott Kirby, Nicola Randall","doi":"10.1186/s13750-025-00372-7","DOIUrl":"10.1186/s13750-025-00372-7","url":null,"abstract":"<p><strong>Background: </strong>The global agriculture sector is expected to contribute towards carbon net zero by adopting interventions to reduce/offset greenhouse gas emissions and increase carbon sequestration/removal. Many of these interventions require change to land management and agriculturally associated habitats, subsequently impacting biodiversity. This relationship is important as the Convention on Biological Diversity has also pledged to reverse nature decline. To understand this relationship, a systematic map was developed to collate evidence relating to the impacts of carbon footprint reducing interventions on agriculturally associated biodiversity. This systematic map collated studies from temperate farming systems including northern Europe, North America and New Zealand.</p><p><strong>Methods: </strong>A protocol was published to define the methodology. Potentially relevant articles were identified by searching three academic databases using a predefined search string. Also, nine organisational websites were searched using key words. All potentially relevant articles were exported into EPPI-Reviewer-Web. Following deduplication, the remaining articles were screened at title and abstract level, partially with the aide of machine learning, before full text screening and extraction of metadata.</p><p><strong>Review findings: </strong>Screening began with 67,617 articles that ended with an evidence base of 820 primary research studies and 82 reviews. The evidence base includes studies from 1978 to April 2024, of which 81% were studies that lasted less than 5 years. Whilst microorganisms (n = 328), arthropods (n = 190), worms (n = 121) and plants (n = 118) were well represented in the evidence base, other groups such as birds (n = 32), gastropods (n = 16), mammals (n = 13), amphibians (n = 1) and reptiles (n = 1) were represented less well. The most studied interventions were to increase soil organic carbon through reduced tillage (n = 227) and cover cropping (n = 136). However, there were less than five studies in total for the following land management objectives: avoiding soil compaction (n = 2), precision farming (n = 2) and renewable energy production. Study authors reported carbon footprint-reducing practices to positively impact biodiversity in 65% of studies, to have mixed effects in 11%, negative in 8% and no effect in 16% of studies. As no critical appraisal was carried out on the included studies, we recommend further study validation and synthesis in order to support these findings.</p><p><strong>Conclusions: </strong>The evidence base has highlighted evidence clusters and gaps on how farming practices that can reduce the carbon footprint of a farm impacts agriculturally associated biodiversity. There are many areas for further research including studies investigating the long-term relationship of interventions that alter habitats over a long period such as rewetting peat soils and increasing tree cover. Future research sh","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"16"},"PeriodicalIF":5.2,"publicationDate":"2025-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12514805/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145276470","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"What evidence exists on the interlinkages between ecological and societal impacts of borealisation of the arctic? A systematic map protocol.","authors":"Kate Baker, Vera Helene Hausner, Jennifer Ramsay, Helen C Wheeler","doi":"10.1186/s13750-025-00367-4","DOIUrl":"10.1186/s13750-025-00367-4","url":null,"abstract":"<p><strong>Background: </strong>As the global climate rapidly warms, one pervasive impact is the \"borealisation\" of the Arctic. Borealisation occurs when the species, communities and ecological processes of the Arctic transform to resemble that of more boreal lower latitudes. Such change is likely to have profound impacts on the diverse communities and cultures of the Arctic. Some of these impacts are starting to be documented, however this evidence has not been synthesised systematically. This systematic map protocol will therefore address the research question: \"What evidence exists on the interlinkages between ecological and societal impacts of borealisation of the Arctic?\" Additionally, this systematic map will support two current assessments of the Arctic Council working groups on the societal and ecological impacts of climate change in the Arctic, thus responding to policy relevant questions posed by Arctic governments.</p><p><strong>Methods: </strong>Following guidelines set out by the Collaboration for Environmental Evidence (CEE), a search of literature, both peer reviewed and grey, will be performed using a range of bibliographic databases, websites and search engines. The search strategy will use a pre-defined search string with Boolean operators. The search results will be screened for relevance according to specific inclusion and exclusion criteria. This will be done in two stages - firstly a screen of titles and abstracts, then a full text screening of eligible articles. At both stages, articles will be excluded if they fail to meet all eligibility criteria or if they meet exclusion criteria. Next, articles that are eligible after full text screening will be coded. At both the screening and coding stages, two reviewers will independently assess a defined number of articles to ensure inter-reviewer reliability and resolve differences. This evidence will then form a searchable database with accompanying visual outputs. A narrative output will outline the range and distribution of evidence, identify potential bias, knowledge clusters and gaps, and will explore areas for further research.</p>","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"15"},"PeriodicalIF":5.2,"publicationDate":"2025-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12317435/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144765663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"What is the evidence for the impacts of airborne anthropogenic noise on wildlife? A systematic map update.","authors":"Léa Terray, Benjamin Petiteau, Guillaume Dutilleux, Sylvie Vanpeene, Pamela Amiard, Romain Sordello, Yorick Reyjol","doi":"10.1186/s13750-025-00368-3","DOIUrl":"10.1186/s13750-025-00368-3","url":null,"abstract":"<p><strong>Background: </strong>Noise from human activities is a major concern for wildlife, with numerous studies demonstrating significant impacts. In 2020, Sordello and collaborators systematically mapped the literature on the impacts of anthropogenic noise on wildlife up to 2018. Since then, research on this topic has continued to grow steadily. To reflect these developments, we present an updated systematic map encompassing studies published through 2023, exclusively focused on airborne noise.</p><p><strong>Methods: </strong>The method follows the a priori protocol published by Sordello and collaborators in 2019. The present work includes literature searches by Sordello et al. (2020) and a complementary search update performed on 2020-2023. Literature from Sordello et al. (2020) was re-screened to align with the updated scope, now restricted to airborne noise. For the update, both peer-reviewed and grey literature were retrieved from Scopus, the Web of Science Core Collection, and Google Scholar. Titles, abstracts, and full texts were screened by eligibility criteria, and included articles were coded. We included all wild terrestrial or semi-aquatic taxonomic groups, and anthropogenic noise from various sources (e.g., transport, urban, recreational) was considered, along with all relevant outcomes (e.g., behaviour, reproduction, physiology).</p><p><strong>Review findings: </strong>Sordello et al. (2020) provided 1,794 articles, of which 466 were retained after re-screening the full texts. The search update yielded 13,698 citations, resulting in 397 relevant articles. A total of 863 articles were included in the map (665 primary research studies, 196 reviews and meta-analyses, 2 modelling papers). Most studies have been conducted in the USA. Birds are the most studied taxonomic group (64%), followed by mammals (22%); transportation is the most studied source of noise (43%), followed by urban noise (24%); behaviour (27%) and vocal communication (25%) are the most studied outcomes.</p><p><strong>Conclusion: </strong>The map represents an updated state of the art on the impact of airborne anthropogenic noise on wildlife and can serve as a starting point for further syntheses of evidence. Three clusters of knowledge were identified as suitable candidates for future syntheses: (1) What is the impact of anthropogenic noise on mammals' behaviour? (2) What is the impact of anthropogenic noise on birds' reproductive success? (3) What is the impact of anthropogenic noise on species richness and diversity? In addition, the knowledge gaps identified may be used to inform future research and address the apparent imbalance in the published research: many taxonomic groups are still understudied (e.g., especially reptiles and arachnids), many potential sources of noise disturbance are neglected (e.g., recreational and military noise) and the impacts of noise are unevenly studied between taxonomic groups.</p>","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"14"},"PeriodicalIF":5.2,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12297495/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144718933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mapping large bodies of research in environmental sciences: insights from compiling evidence on the recovery and reuse of nutrients found in human excreta and domestic wastewater.","authors":"Robin Harder","doi":"10.1186/s13750-025-00366-5","DOIUrl":"10.1186/s13750-025-00366-5","url":null,"abstract":"<p><p>Mapping evidence on a particular research topic among others aims to provide a comprehensive overview of the topic along with a searchable database of relevant literature. When attempting to map large bodies of research, mappers may soon find themselves in a situation where the resources available for the mapping are incommensurate to the number of studies to be handled. This typically requires either a narrower scope of the map or a streamlined mapping process. Grounded in a comparison of five evidence maps on the topic of recovery and reuse of nutrients found in human excreta and domestic wastewater-some of them systematic, some not-the present paper sets out to quantify the potential effect of procedural differences on mapping outcomes. Ultimately, the goal is to discern the factors that matter most for comprehensive and balanced mapping outcomes. This exploration suggests that a good search strategy is key when mapping large bodies of research, especially so when terminology is barely standardized. The paper also sheds light to an issue that could be described as differential search term sensitivity and specificity (compound search terms that are not equally sensitive and specific across all subdomains of the map) and that may deserve more attention in evidence mapping. Drawing from my experiences from compiling the online evidence platform Egestabase, the paper sketches how this issue might be mitigated. In addition, the paper outlines several measures that can help achieve substantial efficiency gains, and offers reflections on how to set priorities and navigate tradeoffs when a standard systematic mapping process appears not to be viable and not strictly necessary.</p>","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"13"},"PeriodicalIF":3.4,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12261714/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144638473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The impact of aminoglycoside exposure on soil and plant root-associated microbiota: a meta-analysis.","authors":"Jessica L Coates, Ashanti J Lawson, Kathleen Bostick, Mentewab Ayalew","doi":"10.1186/s13750-025-00365-6","DOIUrl":"10.1186/s13750-025-00365-6","url":null,"abstract":"<p><strong>Background: </strong>Exposure to aminoglycosides, a class of potent bactericidal antibiotics naturally produced by soil microorganisms and commonly used in agriculture, has the potential to cause shifts in the population dynamics of microorganisms that impact plant and soil health. In particular, aminoglycoside exposure could result in alterations of the soil and plant root-associated bacterial species diversity and richness due to their potent inhibitory action on microbial growth, the creation of selective conditions for the proliferation of antibiotic-resistant bacteria, or a reduction in the ability to suppress soil pathogens. Previous studies have attempted to understand the relationship between aminoglycoside exposure and the plant-associated microbiota with varying results. Thus, this systematic review aims to survey all relevant published data to answer the question, \"What is the impact of aminoglycoside exposure on the soil and plant root-associated microbiota?\"</p><p><strong>Methods: </strong>We searched 5 academic databases and 1 specialist organization database for scientific journal publications written in any language. Articles were included based on the criteria described in Coates et al., 2022. Included studies were subject to critical appraisal using the CEE Critical Appraisal Tool Version 0.2 (Prototype) to evaluate their susceptibility to confounding factors, misclassification bias, selection bias, attrition bias, reporting bias and analysis bias. Studies deemed to be high risk based on critical appraisal results were excluded from further analysis. Descriptive data analysis was performed for studies considered low or unclear for risk of bias. Meta-analyses were conducted for antibiotic resistance and microbial diversity.</p><p><strong>Review findings: </strong>Out of 8370 screened records, 50 articles fulfilled the search criteria, and from these, 13 studies were included in meta-analysis. Most studies investigated the impact of aminoglycoside exposure on soil microbiota (93%) in a laboratory setting (62%), primarily from the United States (32%), China (24%), France, Switzerland and Germany (8%). A limited number of studies investigated the impact of aminoglycoside exposure on disease suppression, so it was excluded from meta-analysis. Therefore, our synthesis primarily details the impact of aminoglycoside exposure on the microbial diversity and antibiotic resistance of the soil microbiota. Overall, exposure to aminoglycosides did not result in a significant change in the microbial diversity. However, soil use, pH, and type of aminoglycoside used could be potential modifiers. Additionally, we observed an average 7% of the microbial population exhibiting resistance to aminoglycosides, with the relationship between the exposure concentration and the selection concentration emerging as a potential modifier.</p><p><strong>Conclusions: </strong>Current research is limited by gaps in understanding the relationship between","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"12"},"PeriodicalIF":3.4,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12243174/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144610029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"What evidence exists on wild bee trends in Germany? A systematic map.","authors":"Anne-Christine Mupepele, Niels Hellwig, Petra Dieker, Alexandra-Maria Klein","doi":"10.1186/s13750-025-00364-7","DOIUrl":"10.1186/s13750-025-00364-7","url":null,"abstract":"<p><strong>Background: </strong>Wild bees have attracted growing attention from both the scientific community and civil society, alongside increasing evidence of biodiversity losses. Declining wild bee populations threaten both the quality and quantity of pollination, which also affect crop production and are therefore critically important for human wellbeing. Landscape homogenisation, land use changes, land use intensity, and climate change are driving the decline. Despite concerns about the wild bee decline, knowledge of wild bee population patterns and long-term trends across Germany remains limited. Here, we present a systematic map, including a newly developed comprehensive database that compiles available data on temporal trends in wild bee communities across Germany. Our goal is to provide an overview of the frequency of wild bee trend studies over time and the land use types and geographical areas they have covered.</p><p><strong>Methods: </strong>Our search for data on wild bee trends was conducted in November 2020 and included peer-reviewed literature (from Web of Science databases and Scopus) and grey literature in English and in German. After screening the literature by title and abstract, relevant data were extracted from eligible studies. All eligible studies included data on wild bee taxa from at least two years at the same site within Germany and using the same sampling method. The database consists of data sheets on studies (bibliographic context), on covariates (methodological and spatio-temporal context) and on data (bee species sampled at a specific time on a specific location - exemplarily for two German regions).</p><p><strong>Review findings: </strong>The database contains 382 studies out of 24,486 initial records. Nearly 75% of the full texts screened did not include field data on wild bees from at least two different years and were therefore excluded. Studies date back to the 1880s, with a consistently high number of studies since the 1990s. Most studies were published in German-language journals of entomological societies in Germany. Data originate from different types of land use throughout the country, mostly from southern and north-western Germany and from urban areas.</p><p><strong>Conclusions: </strong>The systematic map shows that there is a lack of long-term monitoring studies. Moreover, there are research gaps in land use categories and federal states, which hinder more regional or land use-specific analyses. We encourage researchers and practitioners to use the database for further analyses on bee trends and their drivers, and the potential success of mitigation measures.</p>","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"11"},"PeriodicalIF":3.4,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12178071/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144334138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effectiveness of perches in promoting bird-mediated seed dispersal for natural forest regeneration: a systematic review.","authors":"Jelaine L Gan, Matthew J Grainger, Mark David Foster Shirley, Sheena Davis, Molly Watson, Shreya Dube, Marion Pfeifer","doi":"10.1186/s13750-025-00363-8","DOIUrl":"10.1186/s13750-025-00363-8","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Assisted Natural Regeneration (ANR) is an increasingly popular cost-effective approach to restore forests for climate change mitigation and biodiversity conservation. One ANR strategy is the use of perches to attract avian seed dispersers to degraded landscapes for increased seed supply and seedling establishment. This systematic review sought to determine the effectiveness of artificial, semi-natural, and natural perches in promoting natural forest regeneration, specifically in driving four outcomes: seed richness, seed density, seedling richness, and seedling density.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;In September 2023, we searched for studies in eight bibliographic sources, which include an organizational library and a web-based search engine, using a refined search string in English. After deduplication, we conducted double screening at title and abstract, then at full text level to check for eligibility (e.g., compared perches versus control). The final list of studies underwent critical appraisal based on risk of bias and method validity, then data extraction. We used Hedges' g as our effect size and fitted each outcome in a three-level meta-regression model. We also tested the effect of matrix type, bioregion, and precipitation variation as modifiers, and conducted sensitivity analysis based on risk of bias and method validity.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Review findings: &lt;/strong&gt;After screening, we accepted a total of 396 studies in 79 articles for the review. The majority of these studies examined seed (49%) and seedling density (28%) outcomes over richness, using mostly natural perches (68%) and, to a lesser frequency, artificial and semi-natural perches. Most studies that included distance to forest edge as a factor reported no effects (n = 68). We then analysed 333 studies in a meta-analysis. Results showed that natural perches had overall positive effects in increasing seed and seedling density and richness, while artificial and semi-natural perches were effective only for seed outcomes. We found high heterogeneity in our models, with perch effectiveness affected by matrix type, bioregion, precipitation variation, method specificity, as well as study quality. In general, perches showed robust positive effects in shrublands and grasslands in tropical, subtropical, and mediterranean biomes.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;The use of perches can be an effective ANR strategy to increase seed rain and seedling establishment in a variety of degraded landscapes. We recommend that natural perches be preserved in the matrix, but in areas lacking these natural features, to explore the use of artificial or semi-natural perches to increase seed rain and follow it up with additional treatments, such as soil amelioration and weeding, to improve seedling establishment. Due to insufficient data, we could not analyse the effect of distance to forest edge as a moderator in our meta-regression models. This gap can be addressed by e","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"10"},"PeriodicalIF":3.4,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12166613/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144295080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating generative AI for qualitative data extraction in community-based fisheries management literature.","authors":"S Spillias, K M Ollerhead, M Andreotta, R Annand-Jones, F Boschetti, J Duggan, D B Karcher, C Paris, R J Shellock, R Trebilco","doi":"10.1186/s13750-025-00362-9","DOIUrl":"10.1186/s13750-025-00362-9","url":null,"abstract":"<p><p>Uptake of AI tools in knowledge production processes is rapidly growing. In this pilot study, we explore the ability of generative AI tools to reliably extract qualitative data from a limited sample of peer-reviewed documents in the context of community-based fisheries management (CBFM) literature. Specifically, we evaluate the capacity of multiple AI tools to analyse 33 CBFM papers and extract relevant information for a systematic literature review, comparing the results to those of human reviewers. We address how well AI tools can discern the presence of relevant contextual data, whether the outputs of AI tools are comparable to human extractions, and whether the difficulty of question influences the performance of the extraction. While the AI tools we tested (GPT4-Turbo and Elicit) were not reliable in discerning the presence or absence of contextual data, at least one of the AI tools consistently returned responses that were on par with human reviewers. These results highlight the potential utility of AI tools in the extraction phase of evidence synthesis for supporting human-led reviews, while underscoring the ongoing need for human oversight. This exploratory investigation provides initial insights into the current capabilities and limitations of AI in qualitative data extraction within the specific domain of CBFM, laying groundwork for future, more comprehensive evaluations across diverse fields and larger datasets.</p>","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"9"},"PeriodicalIF":3.4,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12128520/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144200573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Verifying authors' claims to have conducted a Systematic Review? A checklist for journal editors and peer reviewers.","authors":"Andrew S Pullin, Biljana Macura","doi":"10.1186/s13750-025-00361-w","DOIUrl":"10.1186/s13750-025-00361-w","url":null,"abstract":"","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"8"},"PeriodicalIF":3.4,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12076875/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144026113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Testing the utility of GPT for title and abstract screening in environmental systematic evidence synthesis.","authors":"Björn Nykvist, Biljana Macura, Maria Xylia, Erik Olsson","doi":"10.1186/s13750-025-00360-x","DOIUrl":"https://doi.org/10.1186/s13750-025-00360-x","url":null,"abstract":"<p><p>In this paper we show that OpenAI's Large Language Model (LLM) GPT perform remarkably well when used for title and abstract eligibility screening of scientific articles and within a (systematic) literature review workflow. We evaluated GPT on screening data from a systematic review study on electric vehicle charging infrastructure demand with almost 12,000 records using the same eligibility criteria as human screeners. We tested 3 different versions of this model that were tasked to distinguishing between relevant and irrelevant content by responding with a relevance probability between 0 and 1. For the latest GPT-4 model (tested in November 2023) and probability cutoff 0.5 the recall rate is 100%, meaning no relevant papers were missed and using this mode for screening would have saved 50% of the time that would otherwise be spent on manual screening. Experimenting with a higher cut of threshold can save more time. With threshold chosen so that recall is still above 95% for GPT-4 (where up to 5% of relevant papers might be missed), the model could save 75% of the time spent on manual screening. If automation technologies can replicate manual screening by human experts with effectiveness, accuracy, and precision, the work and cost savings are significant. Furthermore, the value of a comprehensive list of relevant literature, rather quickly available at the start of a research project, is hard to understate. However, as this study only evaluated the performance on one systematic review and one prompt, we caution that more test and methodological development is needed, and outline the next steps to properly evaluate rigor and effectiveness of LLMs for eligibility screening.</p>","PeriodicalId":48621,"journal":{"name":"Environmental Evidence","volume":"14 1","pages":"7"},"PeriodicalIF":3.4,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12016299/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144023071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}