Tracking phases of the fall armyworm (Spodoptera frugiperda) invasion across multiple continents using news media

Tracking phases of the fall armyworm (Spodoptera frugiperda) invasion across multiple continents using news media | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Tracking phases of the fall armyworm (Spodoptera frugiperda) invasion across multiple continents using news media Kathryn Bjorklund, Melissa A. Barton, Stefan Daume, Peter Søgaard Jørgensen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7393018/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Understanding the dynamics of cascading social-ecological impacts associated with emerging pests and pathogens is key for addressing the challenges they introduce in an ever more interconnected and rapidly changing world. Here, we used topic modeling of digital news articles to investigate the potential cascading social-ecological impacts associated with the ongoing fall armyworm invasion of multiple geographic regions. We identified regional thematic shifts in the popular news media discourse surrounding the fall armyworm invasion. In the news discourse in Oceania, we discerned a stronger focus on invasion preparation than in regions like Africa and Asia. Additionally, we observed a common biological invasion phase pattern across regions, with Africa distinguished by a longer and proportionally larger impacts-related phase. These regional variations illuminate localized priorities in addressing this invasive species. By highlighting the significance of applying machine learning techniques to news articles to identify and describe cascading social-ecological impacts of emerging pests and pathogens, we can improve our understanding of these patterns and inform more targeted management and mitigation strategies. social-ecological impacts fall armyworm Spodoptera frugiperda text analysis digital news articles Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Emergent infectious diseases and pest species can have cascading consequences on society, that extend far beyond their original impacts. The COVID-19 pandemic is a recent example of this with wide-ranging, longer-term consequences that include wildlife trade bans, economic stimulus policies, and increases in global poverty, all of which in turn produce knock-on effects (Diffenbaugh 2022 ). Similarly, a coffee leaf rust (Hemileia vastatrix) epidemic in Central America that initially devastated coffee crops later prompted a mass domestic and international migration of smallholder farmers seeking alternative economic opportunities (Dupre et al. 2022 ). Another notable case is that of the fall armyworm Spodoptera frugiperda (FAW), an agricultural pest native to the Americas (Lu and Adang 1996 ), that was introduced to Africa in 2016, likely as a stowaway on commercial aircraft (Day et al. 2017 ). In part due to its ability to migrate long distances on prevailing winds (Day et al. 2017 ), FAW spread rapidly across three continents over the span of six years (Fig. 1 ), contributing to widespread crop damage, primarily affecting maize (Prasanna et al. 2022 ). FAW in Africa is responsible for an estimated 9.4 billion USD in yield losses per annum, making it the most damaging invasive species on the continent (Eschen et al. 2021 ). These losses threaten the food security and livelihoods of smallholder farmers in particular (Devi 2018 ; Tambo et al. 2020 ). There remains a need for broader understanding of the cascading impacts linked to emerging pests and pathogens and their drivers and triggers, as well as the development of effective methods for studying them. In this paper, we aim to address these gaps by exploring the recent rapid spread of the fall armyworm, and we introduce a mixed-methods approach incorporating topic modeling of digital text corpora to study such impacts. In the context of the fall armyworm case study, we examine the types of impacts—ecological, social, and/or combined—that occur in each region’s digital news discourse. Additionally, we identify regional variation in mentions of governmental involvement and pest management strategies. Methods Data collection We selected digital news media as the data source for our investigation into the fall armyworm invasion, leveraging its real-time nature and broad coverage. News media sources also provide diverse perspectives on the issue and serve as gauges of public interest and concern (Tateosian et al. 2023 ). We extracted news articles from the freely accessible Google News platform, which aggregates news articles from thousands of sources worldwide and stores web news content extending back to 2003. We retrieved URLs of articles published between 1 January 2016 and 9 December 2022 from the Google News RSS feed. We used the Latin name “ Spodoptera frugiperda ” and the English common name “armyworm” as search terms to obtain approximately 1,100 URLs for English-language articles. We used the tool Extractor API ( https://extractorapi.com/ ) to obtain the full texts, domains, and publication dates. Structural topic model In this study, we employed a structural topic model (STM), a machine learning tool that facilitates uncovering latent topics within large text collections. Topic modeling is an unsupervised classification method that takes a document collection as input and returns two distributions which characterize the prevalence of topics in a given document and the probability with which specific words are associated with a topic. An STM combines probabilistic modeling with regression analysis and additionally integrates metadata to provide a deeper contextual understanding of topics (Roberts et al. 2019 ). To support this approach, we annotated a news article corpus with metadata. First, we reviewed each article to identify its focal country or region and corresponding continent, enabling geographic analysis of topics and themes. Second, we established a FAW entry date for each country referenced, drawing on multiple sources (Appendix 1), and calculated an “invasion year” (the number of years at time of publication since FAW first entered a country) for each article by subtracting the country-specific FAW entry date from the article publication date. We used this invasion year as metadata to track the evolution of the themes in public discourse as the FAW invasion progressed. Finally, we eliminated duplicate articles and those regarding countries where FAW is considered endemic. The refined news article corpus encompasses content from 45 of the 87 countries where FAW has emerged. We fitted the STM using complete news articles (N = 570) as documents and included continent and invasion year as covariates in order to explore regional variances and thematic shifts over time in FAW discourse. During text data preprocessing, we removed English stop words and a custom list of geographic and FAW-specific terms to refine topic analysis by eliminating ubiquitous, and thus, less informative words. We selected a K value of 33 topics for the STM based on iterative model diagnostics and interpretation (Appendix 2). Interpreting the structural topic model In STM, like other topic models, topics are characterized by the probability (Prob) of words belonging to a specific topic, indicating the strength of association. In addition, STM offers a metric quantifying the combined frequency and exclusivity of words within a topic (FREX) (Roberts et al. 2019 ). We interpreted topics generated by the STM inductively, emphasizing the Prob and FREX of word assemblages, combined with a manual interpretation of the five most representative documents for each topic (Appendix 3). We then categorized topics into nine broader themes: 1) ‘Invasive Species’ to capture more general and introductory discussion of FAW, 2) ‘Surveillance and Prevention’ to group discourse surrounding proactive measures of different regions, 3) ‘Research and Development’ to group evidence of investment in solution-oriented efforts, 4) ‘Conventional Control’ and 5) ‘Agroecological Control’ to distinguish between reactive approaches to FAW management across regions, 6) ‘Government’ to evaluate involvement of governmental bodies in addressing FAW, 7) ‘Agriculture’ to illustrate wide-ranging, sectoral discourse around FAW, and 8) ‘Crop Impacts’ and 9) ‘Social Impacts’ to differentiate the types of impacts observed in different regions (Appendix 4). Prediction theme prevalence across biological invasion phases We further contextualized identified themes by formulating predictions about their likely occurrence and distribution across the phases of the biological invasion process, as described by Welsh et al. ( 2021 ) (Fig. 2 ). In the first phase ‘Preventing Establishment’, we expect the theme ‘Surveillance and Prevention’ to be prevalent as it focuses on proactive measures used to prevent and monitor potential invasions. In the second phase, following the detection of a pest, an ‘Incursion Response’ occurs, during which efforts to identify, eradicate, contain and manage the pest invasion take place, as presented by Welsh et al. ( 2021 ). We expect the themes ‘Conventional Control’, ‘Agroecological Control’, ‘Research and Development’ to be prevalent in this phase, as many regions initially rely on synthetic insecticides to address immediate threats of pests (Bale et al. 2008 ; Brévault and Bouyer 2014 ) or employ sustainable pest management trials and practices to reduce potential damage. Similarly, increased funding for Research and Development and scientific advancements are likely to occur in response to the invasion (Fig. 2 ). Lastly, Welsh et al.’s biological invasion process includes a ‘Pest Impacts’ phase, where a pest’s direct and indirect effects, such as yield losses, economic burdens, and social consequences like job losses and market disruptions, are experienced (Welsh et al. 2021 ). We categorized the ‘Crop Impacts’ and ‘Social Impacts’ themes as relevant to this phase, as they address the broader consequences of the invasion (Fig. 2 ). Finally, we expect the themes ‘Invasive Species,’ ‘Government’ and ‘Agriculture’ to occur across the three phases (Fig. 2 ), as they encompass transitional or overarching elements such as pest biology, governmental involvement and agricultural metrics. Calculating the prevalence of themes by continent and by continent over time To capture contextual variability and reduce noise in the model, we aggregated news articles by ‘continent’ and by both ‘continent’ and ‘invasion year.’ For each grouping, we calculated the mean topic proportions to understand how dominant or frequent certain topics were in different continents or during different years. We then added the nine predefined themes as a variable to the groupings, which allowed for the categorization of related topics into broader themes. We then summed the mean topic proportions for topics that fell under the same theme to calculate the cumulative proportions of each theme (Fig. 3 ). Calculating biological invasion phase of best fit by continent over time To assess how closely observed themes aligned with predicted theme-phase patterns, we classified each observed region-year pairing into a corresponding biological invasion phase. First, we normalized the observed theme proportions by rescaling them within each region-theme group to their local maxima. For each region and year, we then calculated the mean absolute distance between the observed normalized and predicted theme values. We assigned each observation to the phase with the minimum distance. Results Overall temporal thematic trends Over the duration of the invasion, and despite the length of this period varying between continents, we find similar patterns in temporal trends of themes (Fig. 3 ). Overall themes show different temporal patterns that in part confirm some of the predictions of our idealized model combining expected news themes with invasion phases (Fig. 2 ). Confirming our prediction, all continents see an early emphasis on ‘Surveillance and Prevention’ with one or both of the two control-oriented themes and the ‘Research and Development’ theme peaking later (‘Conventional Control’, ‘Agroecological Control’, and ‘Research and Development.’ (Fig. 3 )). In contrast to our predictions, ‘Crop Impacts’ and ‘Social Impacts’ do not show a tendency for higher prevalence later in an invasion sequence, but in general occur in parallel to the control and ‘Research and Development’ themes, mentioned above. Among the themes that we predicted to occur across phases, the ‘Government’ theme confirms this pattern, while the ‘Agriculture’ and ‘Invasive Species’ themes occurred too low a prevalence to find any conclusive patterns (Fig. 3 ). Thematic variation between continents Across continents, there is variation in the prominence of themes. Africa features a thematic emphasis on both control types (‘Conventional Control’ and ‘Agroecological Control’), as well as ‘Surveillance and Prevention’, ‘Research and Development’, and ‘Government’ (Fig. 3 ). ‘Social Impacts’, ‘Conventional Control’, ‘Agroecological Control’, and ‘Government’ are most prevalent in Asia’s discourse. However, the continent exhibits greater thematic breadth and less depth relative to other continents. For Oceania, ‘Surveillance and Prevention’ along with ‘Crop Impacts’ dominate both its intra- and inter-continental discourse (Fig. 3 ). Progression of invasion phases by region Consistent with our idealized model, all three continents exhibit early emphasis on a Surveillance phase (Fig. 4 ). A progression from a Surveillance phase to an Impact one, followed by a Control phase, is evident across regions. In Africa, however, the Impact phase is more pronounced and sustained, while in Asia it appears to a lesser extent (Fig. 4 ). Discussion Limitations of data sources and methods An important consideration in this study is the extent to which the captured themes and their temporal patterns can be interpreted as cascading impacts. The captured themes do not necessarily reflect the exact reactions on the ground or measures being implemented. Instead, they indicate the importance or focus placed on these themes by media outlets. In the early stages of invasion, this focus may include concerns about potential or anticipated impacts that have not yet materialized. As such, some themes that appear early in the invasion timeline may reflect expectations about future consequences, rather than observed outcomes. The overall global trends are also likely to be heavily influenced by countries or regions with a larger number of articles available in English. Additionally, a theme may signify different practices or strategies depending on governmental, institutional or geographical contexts. Although we identified overarching themes within regional discourse, not all countries are proportionally represented in available news articles, so country-level variation could not be assessed. The more recent invasion of Asia and Oceania also affects the quantity of available news coverage, as well as availability of FAW reporting data underlining assessment of impacts. This is particularly the case for China, Taiwan, India and South-East Asian countries, where documentation of yield losses due to FAW is limited (Overton et al. 2021 ). Another challenge to identifying cascading impacts triggered by the introduction of FAW or other pests is that concerns about market access may incentivize countries to delay or avoid reporting invasive species. In Asia there have been reporting delays and inconsistencies that are known to have affected the perceived invasion pattern of FAW, supporting the “out of Africa” thesis (Kenis et al. 2023 ). Media theme patterns in relation to invasion model The observed phase trends may reflect an interplay between real-world invasion dynamics and media production practices. The progression of phases partly aligns with our idealized model predictions, particularly regarding an early Surveillance phase. However, the subsequent emergence of an Impact phase followed by a Control phase may indicate that consequences of the invasion unfolded rapidly and that control efforts were reactive, that early coverage included anticipated outcomes, or that it relied on anecdotal evidence and preliminary data. Alternatively, media outlets initially emphasized these topics, but attention subsequently declined following the issue-attention cycle, or the outbreak itself was effectively controlled. Interpreting continental thematic variation Continental variations in these media representations of fall armyworm invasions appear to reflect the actual conditions and responses occurring within these regions. For Oceania, the prominence of the theme ‘Surveillance & Prevention’ aligns with the region’s demonstrated biosecurity measures and infrastructure. Australia and New Zealand, in particular, have rigorous biosecurity protocols and well-established institutions stemming from their need to protect their unique, geographically isolated ecosystems and significant agricultural sectors from invasive species. These measures are often regarded as more stringent than those of many other countries (Stone 2021 ). Additionally, the earlier outbreaks of FAW in Africa and Asia may have provided countries in Oceania advance warning, potentially increasing regional discussions around prevention and preparedness. Both pest control methods themes are more predominantly emphasized in the discourse surrounding Africa and Asia. In Africa, prior to the invasion of FAW, the use of chemical insecticides among maize producers was minimal (Kenis et al. 2023 ). Post-invasion, synthetic pesticides have become the most common management method for FAW in Africa (Niassy et al. 2021 ). The government of Ghana, for instance, after having suffered financial losses, reacted by widely deploying insecticides (Safo et al. 2023 ). Additionally, the adoption of genetically modified crops on the African continent expanded beyond South Africa with the introduction of programs such as TELA maize in Nigeria (Orchardson 2024 ). Similarly, in China, the invasion has led to increased pesticide expenditure among smallholder farmers, with local maize growers in Yunnan primarily relying on pesticides for FAW management (Yang et al. 2021 ). In a survey conducted involving farmers from several South Asian countries, respondents reported using an assortment of management practices, with chemical pesticides being the dominant strategy (Khan et al. 2023 ). The theme of ‘Agroecological Control’ also emerges in the discourse of these regions, potentially driven by development efforts. For example, following the FAW invasion in Sub-Saharan Africa, governments and international and non-governmental organizations discussed training farmers and agricultural stakeholders on constructing integrated pest management (IPM) approaches for FAW control and management (Matova et al. 2020 ) . With regard to impact types, ‘Social Impacts’ proves more pronounced than ‘Crop Impacts’ in Africa and Asia, where subsistence farming is widespread and harm to crops has more direct social consequences (Giller et al. 2021 ). In both regions, FAW has disrupted smallholder farming systems, exacerbated household hunger, reduced incomes, and created other negative impacts on human health and the environment (Midega et al. 2023 ). Extending news-driven frameworks to cascading impacts Given that several temporal and regional thematic trends correspond to an existing biological invasion framework and real-world phenomena, this analytical approach demonstrates potential for identifying cascading impacts. This capability is increasingly needed, particularly considering the ongoing migration of FAW from North Africa toward Southern Europe, a process already documented by recent research (Wang et al. 2023 ; Kartakis et al. 2025 ). Utilizing structural topic modeling or comparable analytical methods focused on news media could provide timely monitoring of this evolving threat in Europe, thus supporting improved response strategies. Existing news-scraping tools such as PADI-web (Platform for Automated extraction of Disease Information from the web) could be broadened beyond their current focus on animal and plant pathogens to track other pest species as well as their downstream impacts (Valentin et al. 2021 ; Roche et al. 2024 ). However, the current study is limited in its scope due to the focus on a single pest species and English-language sources. Future research should look to incorporate a wider range of pest species and a multilingual dataset to strengthen the robustness and applicability of the findings. Additionally, future studies could conduct a form of linguistic analysis, such as examining verb tense, to distinguish between projected and reported events in the text, facilitating the exploration of actual cascading effects. Conclusions Structural topic model (STM) analysis reveals regional variations in popular news media discourse surrounding the FAW invasion, highlighting the influence of geographical context on the thematic evolution of FAW-related narratives. In particular, in Oceania, the continent where the FAW invasion is more recent, the discourse focuses more on preparatory actions compared to Africa and Asia. Additionally, all regions follow a broadly similar progression of biological invasion phases: Surveillance, Impact, and Control. Africa features a longer and proportionally more substantial Impact phase, providing insight into the cascading dynamics of such events. When combined with other data sources, findings from structural topic modeling of news media can contribute to a broader understanding of the cascading impacts of emerging agricultural pests. Understanding these regional differences in discourse, and the underlying social-ecological impacts they reflect, is necessary for developing targeted and effective strategies to manage the consequences of FAW invasions. This need is particularly pressing as FAW has the potential to establish overwintering populations in Southern Europe. Applying STM or similar news media analysis methods to monitor evolving discourse in Europe could enhance preparedness and inform timely responses to this growing threat. Declarations Author contributions KB conducted the initial research for this study as part of the thesis requirement for the Social-ecological Resilience for Sustainable Development (SERSD) MSc program at the Stockholm Resilience Centre, Stockholm University, Stockholm, Sweden. All authors contributed to the study conception and design. KB performed material preparation and data collection. All authors performed data analysis. KB wrote the first draft of the manuscript and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Funding We acknowledge funding from the Erling-Persson Family Foundation (PSJ, Grant number: N/A), the European Union (ERC, PSJ, INFLUX, Grant number: 101039376), the IKEA foundation (PSJ, Grant number: N/A), the Marianne and Marcus Wallenberg Foundation (PSJ, Grant number: N/A), and the Swedish Research Council for Sustainable Development Formas (SD, Grant number: 2020-01586). The funders had no role in the design, analysis, or writing of this paper. Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authorities can be held responsible for them. Data availability The full news articles used in this study are subject to copyright and cannot be shared. The code to reproduce the dataset is available at https://github.com/kathrynbjorklund/tracking-faw-invasion-phases-via-news-media. Competing interests The authors have no relevant financial or non-financial interests to disclose. Ethics approval This study analyzed publicly available news articles and did not involve human participants, animals, or sensitive data. Therefore, no formal ethics review was required. Consent to participate Not applicable. Consent to publish Not applicable. References Bale JS, van Lenteren JC, Bigler F (2008) Biological control and sustainable food production. Philos Trans R Soc Lond B Biol Sci 363:761–776. https://doi.org/10.1098/rstb.2007.2182 Brévault T, Bouyer J (2014) From Integrated to System-Wide Pest Management: Challenges for Sustainable Agriculture. Outlook Pest Man 25:212–213. https://doi.org/10.1564/v25_jun_05 Day R, Abrahams P, Bateman M et al (2017) Fall Armyworm: Impacts and Implications for Africa. Outlooks Pest Manage 28:196–201. https://doi.org/10.1564/v28_oct_02 Devi S (2018) Fall armyworm threatens food security in southern Africa. Lancet 391:727. https://doi.org/10.1016/S0140-6736(18)30431-8 Diffenbaugh NS (2022) COVID-19 and the Environment: Short-Run and Potential Long-Run Impacts. Annu Rev Environ Resour 47:65–90. https://doi.org/10.1146/annurev-environ-120920-125207 Dupre SI, Harvey CA, Holland MB (2022) The impact of coffee leaf rust on migration by smallholder coffee farmers in Guatemala. World Dev 156:105918. https://doi.org/10.1016/j.worlddev.2022.105918 Eschen R, Beale T, Bonnin JM et al (2021) Towards estimating the economic cost of invasive alien species to African crop and livestock production. CABI Agric Bioscience 2:18. https://doi.org/10.1186/s43170-021-00038-7 Giller KE, Delaune T, Silva JV et al (2021) The future of farming: Who will produce our food? Food Sec 13:1073–1099. https://doi.org/10.1007/s12571-021-01184-6 Kartakis S, Horrocks KJ, Cingiz K et al (2025) Migration extent and potential economic impact of the fall armyworm in Europe. Sci Rep 15:17405. https://doi.org/10.1038/s41598-025-02595-7 Kenis M, Benelli G, Biondi A et al (2023) Invasiveness, biology, ecology, and management of the fall armyworm, Spodoptera frugiperda. entomologia 43:187–241. https://doi.org/10.1127/entomologia/2022/1659 Khan FZA, Paudel S, Saeed S et al (2023) Mitigating the impact of the invasive fall armyworm: evidence from South Asian farmers and policy recommendations. Int J Pest Manage 1–9. https://doi.org/10.1080/09670874.2023.2205834 Lu Y, Adang MJ (1996) Distinguishing Fall Armyworm (Lepidoptera: Noctuidae) Strains Using a Diagnostic Mitochondrial Dna Marker. Fla Entomol 48–48. https://doi.org/10.2307/3495753 Matova PM, Kamutando CN, Magorokosho C et al (2020) Fall-armyworm invasion, control practices and resistance breeding in Sub-Saharan Africa. Crop Sci 60:2951–2970. https://doi.org/10.1002/csc2.20317 Midega C, Hadi B, McGuire S et al (2023) Fall armyworm: measuring damage and loss caused by a novel invasive pest as a guide to sustainable management. FAO Niassy S, Agbodzavu MK, Kimathi E et al (2021) Bioecology of fall armyworm Spodoptera frugiperda (J. E. Smith), its management and potential patterns of seasonal spread in Africa. PLoS ONE 16:e0249042. https://doi.org/10.1371/journal.pone.0249042 Orchardson E (2024) TELA Maize Project. In: CIMMYT. https://www.cimmyt.org/projects/tela-maize-project/ . Accessed 20 Jan 2025 Overton K, Maino JL, Day R et al (2021) Global crop impacts, yield losses and action thresholds for fall armyworm ( Spodoptera frugiperda ): A review. Crop Prot 145:105641. https://doi.org/10.1016/j.cropro.2021.105641 Prasanna BM, Bruce A, Beyene Y et al (2022) Host plant resistance for fall armyworm management in maize: relevance, status and prospects in Africa and Asia. Theor Appl Genet 135:3897–3916. https://doi.org/10.1007/s00122-022-04073-4 Roberts ME, Stewart BM, Tingley D (2019) stm: An R Package for Structural Topic Models. J Stat Soft 91. https://doi.org/10.18637/jss.v091.i02 Roche M, Rabatel J, Trevennec C, Pieretti I (2024) PADI-web for Plant Health Surveillance. Intelligent Information Systems. Springer, Cham, pp 148–156. https://doi.org/10.1007/978-3-031-61000-4_17 Safo A, Avicor SW, Baidoo PK et al (2023) Farmers’ knowledge, experience and management of fall armyworm in a major maize producing municipality in Ghana. Cogent Food Agric 9:2184006. https://doi.org/10.1080/23311932.2023.2184006 Stone M (2021) Biosecurity is critical to New Zealand’s national security, economy and way of life. N Z Vet J 69:309–312. https://doi.org/10.1080/00480169.2021.1965076 Tambo JA, Kansiime MK, Mugambi I et al (2020) Understanding smallholders’ responses to fall armyworm ( Spodoptera frugiperda ) invasion: Evidence from five African countries. Sci Total Environ 740:140015. https://doi.org/10.1016/j.scitotenv.2020.140015 Tateosian LG, Saffer A, Walden-Schreiner C, Shukunobe M (2023) Plant pest invasions, as seen through news and social media. Comput Environ Urban Syst 100:101922. https://doi.org/10.1016/j.compenvurbsys.2022.101922 Valentin S, Arsevska E, Rabatel J et al (2021) PADI-web 3.0: A new framework for extracting and disseminating fine-grained information from the news for animal disease surveillance. One Health 13:100357. https://doi.org/10.1016/j.onehlt.2021.100357 Wang J, Huang Y, Huang L et al (2023) Migration risk of fall armyworm (Spodoptera frugiperda) from North Africa to Southern Europe. Front Plant Sci 14:1141470. https://doi.org/10.3389/fpls.2023.1141470 Welsh MJ, Turner JA, Epanchin-Niell RS et al (2021) Approaches for estimating benefits and costs of interventions in plant biosecurity across invasion phases. Ecol Appl 31:e02319. https://doi.org/10.1002/eap.2319 Yang X, Wyckhuys KAG, Jia X et al (2021) Fall armyworm invasion heightens pesticide expenditure among Chinese smallholder farmers. J Environ Manage 282:111949. https://doi.org/10.1016/j.jenvman.2021.111949 Supplementary Files ESM1.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 29 Aug, 2025 Reviewers invited by journal 28 Aug, 2025 Editor invited by journal 21 Aug, 2025 Editor assigned by journal 18 Aug, 2025 First submitted to journal 17 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7393018","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":506867000,"identity":"9fbb18c8-d447-4984-8cb9-917d694b4ae7","order_by":0,"name":"Kathryn Bjorklund","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/klEQVRIie3PMUvDQBTA8XcI5xLs+g6hn+GkkCqKn6VHwC5BBKHU7bqcizin3yIfIfAgWaJdU9pBCcRV94AmpUMQru0oeH84eMP9uHcALtcfDAH4ZpCQANN3zZTpfeSoS2Qz5cnBBLakGO0m4nGRlg9mDUNG6Vck6fZk+VEh1FdWcuoFx2cvpoILnQbzWNK9WIU+MnNjJX0IuJgZApnkA/YmScWrkCPTZCe98jdZZu1i3/bFsPtK3JICfASe2L8flb7Qr+S1f2GRHKt5Hg7OlQmsBBeqEnpC/SESsafppXrOsvfis762kk2MgwfYXWW0GzTVzenpvddcLpfrn/YDAtlbVlZ6NzEAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-1334-8626","institution":"Stockholm Resilience Centre","correspondingAuthor":true,"prefix":"","firstName":"Kathryn","middleName":"","lastName":"Bjorklund","suffix":""},{"id":506867001,"identity":"211c386f-8022-4990-8427-49fcbd2b62de","order_by":1,"name":"Melissa A. Barton","email":"","orcid":"","institution":"Stockholm Resilience Centre","correspondingAuthor":false,"prefix":"","firstName":"Melissa","middleName":"A.","lastName":"Barton","suffix":""},{"id":506867002,"identity":"b5193e08-10bf-41c6-a3a5-0ceede05c69d","order_by":2,"name":"Stefan Daume","email":"","orcid":"","institution":"Stockholm Resilience Centre","correspondingAuthor":false,"prefix":"","firstName":"Stefan","middleName":"","lastName":"Daume","suffix":""},{"id":506867003,"identity":"dc25f585-6acb-43c7-8a50-56e934f56bad","order_by":3,"name":"Peter Søgaard Jørgensen","email":"","orcid":"","institution":"Stockholm Resilience Centre","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"Søgaard","lastName":"Jørgensen","suffix":""}],"badges":[],"createdAt":"2025-08-17 14:46:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7393018/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7393018/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90598203,"identity":"d8dcb6ee-ae15-43aa-a189-95f43c7f7633","added_by":"auto","created_at":"2025-09-04 14:05:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":111754,"visible":true,"origin":"","legend":"\u003cp\u003eMap of spread of fall armyworm (FAW). The boundaries of larger countries India, China and Australia have been subdivided to depict more specific areas targeted. FAW data compiled from various sources can be found in Appendix 1. Map was constructed using QGIS version 3.36.1 with spatial data from DIVA-GIS (https://diva-gis.org/) and Natural Earth (https://www.naturalearthdata.com/)\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7393018/v1/027249a6b11521cd2a57fed6.png"},{"id":90598206,"identity":"e1ec21e3-a158-4c41-8182-a1de5036438e","added_by":"auto","created_at":"2025-09-04 14:05:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":44786,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of predicted theme distributions mapped onto the phases of the biological invasion process described by Welsh et al., 2021. Themes such as ‘Surveillance and Prevention’ corresponds strongly with the initial ‘Preventing Establishment’ phase; ‘Conventional Control,’ ‘Research and Development’ and ‘Agroecological Control’ with the ‘Incursion Response’ phase; ‘Crop Impacts’ and ‘Social Impacts’ apply to the ‘Pest Impacts’ phase; and ‘Invasive Species,’ ‘Government’ and ‘Agriculture’ constitute cross-cutting themes that span multiple stages of the invasion process\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7393018/v1/77dc4f8a049fe881d0725132.png"},{"id":90598205,"identity":"591b8664-ea91-49b9-bcf3-eb50b1e64eb3","added_by":"auto","created_at":"2025-09-04 14:05:04","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":70841,"visible":true,"origin":"","legend":"\u003cp\u003eThematic prevalence for articles involving Africa (N = 300), Asia (N = 214) and Oceania (N = 55) over year since fall armyworm entry\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7393018/v1/48bf14163725562c7740a921.png"},{"id":90599109,"identity":"06c377b3-b578-4e7d-a347-e05462cc8dfc","added_by":"auto","created_at":"2025-09-04 14:13:04","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":73836,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Best-fitting biological invasion phase over time by continent, based on thematic prevalence in news coverage. Each tile represents a year-continent combination and is colored by the phase (Surveillance, Control, or Impact) whose predicted theme profile most closely matched the observed distribution of themes in the news data for that region and year. The matching was based on the minimum mean absolute distance between observed normalized theme values and predicted phase profiles. Tile shading indicates the proximity of the match, with darker shades representing closer alignment between observed and predicted values. (b) Temporal trends in the proximity of thematic prevalence in news coverage to each phase by continent. (c) Continent-year observations plotted by proximity to pairs of phases\u003c/p\u003e","description":"","filename":"Figure4a4c.png","url":"https://assets-eu.researchsquare.com/files/rs-7393018/v1/ac0e520ecf95d130e94435a8.png"},{"id":90600902,"identity":"df765ce3-6fa9-44a1-86ad-19a3e01c8d89","added_by":"auto","created_at":"2025-09-04 14:37:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":969909,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7393018/v1/2477f14c-584a-4ef3-9ec3-09d7a5bb922d.pdf"},{"id":90598209,"identity":"785951dc-5d4c-4618-bb04-2cb23ce7d3ce","added_by":"auto","created_at":"2025-09-04 14:05:05","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":439477,"visible":true,"origin":"","legend":"","description":"","filename":"ESM1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7393018/v1/d0a2e4d05c1f6e43aae533e9.pdf"}],"financialInterests":"","formattedTitle":"Tracking phases of the fall armyworm (Spodoptera frugiperda) invasion across multiple continents using news media","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEmergent infectious diseases and pest species can have cascading consequences on society, that extend far beyond their original impacts. The COVID-19 pandemic is a recent example of this with wide-ranging, longer-term consequences that include wildlife trade bans, economic stimulus policies, and increases in global poverty, all of which in turn produce knock-on effects (Diffenbaugh \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Similarly, a coffee leaf rust (Hemileia vastatrix) epidemic in Central America that initially devastated coffee crops later prompted a mass domestic and international migration of smallholder farmers seeking alternative economic opportunities (Dupre et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAnother notable case is that of the fall armyworm \u003cem\u003eSpodoptera frugiperda\u003c/em\u003e (FAW), an agricultural pest native to the Americas (Lu and Adang \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1996\u003c/span\u003e), that was introduced to Africa in 2016, likely as a stowaway on commercial aircraft (Day et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). In part due to its ability to migrate long distances on prevailing winds (Day et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), FAW spread rapidly across three continents over the span of six years (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), contributing to widespread crop damage, primarily affecting maize (Prasanna et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). FAW in Africa is responsible for an estimated 9.4\u0026nbsp;billion USD in yield losses per annum, making it the most damaging invasive species on the continent (Eschen et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These losses threaten the food security and livelihoods of smallholder farmers in particular (Devi \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Tambo et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThere remains a need for broader understanding of the cascading impacts linked to emerging pests and pathogens and their drivers and triggers, as well as the development of effective methods for studying them. In this paper, we aim to address these gaps by exploring the recent rapid spread of the fall armyworm, and we introduce a mixed-methods approach incorporating topic modeling of digital text corpora to study such impacts. In the context of the fall armyworm case study, we examine the types of impacts\u0026mdash;ecological, social, and/or combined\u0026mdash;that occur in each region\u0026rsquo;s digital news discourse. Additionally, we identify regional variation in mentions of governmental involvement and pest management strategies.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eData collection\u003c/h2\u003e\u003cp\u003eWe selected digital news media as the data source for our investigation into the fall armyworm invasion, leveraging its real-time nature and broad coverage. News media sources also provide diverse perspectives on the issue and serve as gauges of public interest and concern (Tateosian et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eWe extracted news articles from the freely accessible Google News platform, which aggregates news articles from thousands of sources worldwide and stores web news content extending back to 2003. We retrieved URLs of articles published between 1 January 2016 and 9 December 2022 from the Google News RSS feed. We used the Latin name \u0026ldquo;\u003cem\u003eSpodoptera frugiperda\u003c/em\u003e\u0026rdquo; and the English common name \u0026ldquo;armyworm\u0026rdquo; as search terms to obtain approximately 1,100 URLs for English-language articles. We used the tool Extractor API (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://extractorapi.com/\u003c/span\u003e\u003cspan address=\"https://extractorapi.com/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to obtain the full texts, domains, and publication dates.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eStructural topic model\u003c/h3\u003e\n\u003cp\u003eIn this study, we employed a structural topic model (STM), a machine learning tool that facilitates uncovering latent topics within large text collections. Topic modeling is an unsupervised classification method that takes a document collection as input and returns two distributions which characterize the prevalence of topics in a given document and the probability with which specific words are associated with a topic. An STM combines probabilistic modeling with regression analysis and additionally integrates metadata to provide a deeper contextual understanding of topics (Roberts et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTo support this approach, we annotated a news article corpus with metadata. First, we reviewed each article to identify its focal country or region and corresponding continent, enabling geographic analysis of topics and themes. Second, we established a FAW entry date for each country referenced, drawing on multiple sources (Appendix 1), and calculated an \u0026ldquo;invasion year\u0026rdquo; (the number of years at time of publication since FAW first entered a country) for each article by subtracting the country-specific FAW entry date from the article publication date. We used this invasion year as metadata to track the evolution of the themes in public discourse as the FAW invasion progressed. Finally, we eliminated duplicate articles and those regarding countries where FAW is considered endemic. The refined news article corpus encompasses content from 45 of the 87 countries where FAW has emerged.\u003c/p\u003e\u003cp\u003eWe fitted the STM using complete news articles (N\u0026thinsp;=\u0026thinsp;570) as documents and included continent and invasion year as covariates in order to explore regional variances and thematic shifts over time in FAW discourse. During text data preprocessing, we removed English stop words and a custom list of geographic and FAW-specific terms to refine topic analysis by eliminating ubiquitous, and thus, less informative words. We selected a K value of 33 topics for the STM based on iterative model diagnostics and interpretation (Appendix 2).\u003c/p\u003e\n\u003ch3\u003eInterpreting the structural topic model\u003c/h3\u003e\n\u003cp\u003eIn STM, like other topic models, topics are characterized by the probability (Prob) of words belonging to a specific topic, indicating the strength of association. In addition, STM offers a metric quantifying the combined frequency and exclusivity of words within a topic (FREX) (Roberts et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). We interpreted topics generated by the STM inductively, emphasizing the Prob and FREX of word assemblages, combined with a manual interpretation of the five most representative documents for each topic (Appendix 3). We then categorized topics into nine broader themes: 1) \u0026lsquo;Invasive Species\u0026rsquo; to capture more general and introductory discussion of FAW, 2) \u0026lsquo;Surveillance and Prevention\u0026rsquo; to group discourse surrounding proactive measures of different regions, 3) \u0026lsquo;Research and Development\u0026rsquo; to group evidence of investment in solution-oriented efforts, 4) \u0026lsquo;Conventional Control\u0026rsquo; and 5) \u0026lsquo;Agroecological Control\u0026rsquo; to distinguish between reactive approaches to FAW management across regions, 6) \u0026lsquo;Government\u0026rsquo; to evaluate involvement of governmental bodies in addressing FAW, 7) \u0026lsquo;Agriculture\u0026rsquo; to illustrate wide-ranging, sectoral discourse around FAW, and 8) \u0026lsquo;Crop Impacts\u0026rsquo; and 9) \u0026lsquo;Social Impacts\u0026rsquo; to differentiate the types of impacts observed in different regions (Appendix 4).\u003c/p\u003e\n\u003ch3\u003ePrediction theme prevalence across biological invasion phases\u003c/h3\u003e\n\u003cp\u003eWe further contextualized identified themes by formulating predictions about their likely occurrence and distribution across the phases of the biological invasion process, as described by Welsh et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In the first phase \u0026lsquo;Preventing Establishment\u0026rsquo;, we expect the theme \u0026lsquo;Surveillance and Prevention\u0026rsquo; to be prevalent as it focuses on proactive measures used to prevent and monitor potential invasions.\u003c/p\u003e\u003cp\u003eIn the second phase, following the detection of a pest, an \u0026lsquo;Incursion Response\u0026rsquo; occurs, during which efforts to identify, eradicate, contain and manage the pest invasion take place, as presented by Welsh et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). We expect the themes \u0026lsquo;Conventional Control\u0026rsquo;, \u0026lsquo;Agroecological Control\u0026rsquo;, \u0026lsquo;Research and Development\u0026rsquo; to be prevalent in this phase, as many regions initially rely on synthetic insecticides to address immediate threats of pests (Bale et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Br\u0026eacute;vault and Bouyer \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) or employ sustainable pest management trials and practices to reduce potential damage. Similarly, increased funding for Research and Development and scientific advancements are likely to occur in response to the invasion (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eLastly, Welsh et al.\u0026rsquo;s biological invasion process includes a \u0026lsquo;Pest Impacts\u0026rsquo; phase, where a pest\u0026rsquo;s direct and indirect effects, such as yield losses, economic burdens, and social consequences like job losses and market disruptions, are experienced (Welsh et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). We categorized the \u0026lsquo;Crop Impacts\u0026rsquo; and \u0026lsquo;Social Impacts\u0026rsquo; themes as relevant to this phase, as they address the broader consequences of the invasion (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFinally, we expect the themes \u0026lsquo;Invasive Species,\u0026rsquo; \u0026lsquo;Government\u0026rsquo; and \u0026lsquo;Agriculture\u0026rsquo; to occur across the three phases (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), as they encompass transitional or overarching elements such as pest biology, governmental involvement and agricultural metrics.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eCalculating the prevalence of themes by continent and by continent over time\u003c/h3\u003e\n\u003cp\u003eTo capture contextual variability and reduce noise in the model, we aggregated news articles by \u0026lsquo;continent\u0026rsquo; and by both \u0026lsquo;continent\u0026rsquo; and \u0026lsquo;invasion year.\u0026rsquo; For each grouping, we calculated the mean topic proportions to understand how dominant or frequent certain topics were in different continents or during different years. We then added the nine predefined themes as a variable to the groupings, which allowed for the categorization of related topics into broader themes. We then summed the mean topic proportions for topics that fell under the same theme to calculate the cumulative proportions of each theme (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eCalculating biological invasion phase of best fit by continent over time\u003c/h2\u003e\u003cp\u003eTo assess how closely observed themes aligned with predicted theme-phase patterns, we classified each observed region-year pairing into a corresponding biological invasion phase. First, we normalized the observed theme proportions by rescaling them within each region-theme group to their local maxima. For each region and year, we then calculated the mean absolute distance between the observed normalized and predicted theme values. We assigned each observation to the phase with the minimum distance.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eOverall temporal thematic trends\u003c/h3\u003e\n\u003cp\u003eOver the duration of the invasion, and despite the length of this period varying between continents, we find similar patterns in temporal trends of themes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Overall themes show different temporal patterns that in part confirm some of the predictions of our idealized model combining expected news themes with invasion phases (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Confirming our prediction, all continents see an early emphasis on \u0026lsquo;Surveillance and Prevention\u0026rsquo; with one or both of the two control-oriented themes and the \u0026lsquo;Research and Development\u0026rsquo; theme peaking later (\u0026lsquo;Conventional Control\u0026rsquo;, \u0026lsquo;Agroecological Control\u0026rsquo;, and \u0026lsquo;Research and Development.\u0026rsquo; (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e)). In contrast to our predictions, \u0026lsquo;Crop Impacts\u0026rsquo; and \u0026lsquo;Social Impacts\u0026rsquo; do not show a tendency for higher prevalence later in an invasion sequence, but in general occur in parallel to the control and \u0026lsquo;Research and Development\u0026rsquo; themes, mentioned above. Among the themes that we predicted to occur across phases, the \u0026lsquo;Government\u0026rsquo; theme confirms this pattern, while the \u0026lsquo;Agriculture\u0026rsquo; and \u0026lsquo;Invasive Species\u0026rsquo; themes occurred too low a prevalence to find any conclusive patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eThematic variation between continents\u003c/h2\u003e\u003cp\u003eAcross continents, there is variation in the prominence of themes. Africa features a thematic emphasis on both control types (\u0026lsquo;Conventional Control\u0026rsquo; and \u0026lsquo;Agroecological Control\u0026rsquo;), as well as \u0026lsquo;Surveillance and Prevention\u0026rsquo;, \u0026lsquo;Research and Development\u0026rsquo;, and \u0026lsquo;Government\u0026rsquo; (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). \u0026lsquo;Social Impacts\u0026rsquo;, \u0026lsquo;Conventional Control\u0026rsquo;, \u0026lsquo;Agroecological Control\u0026rsquo;, and \u0026lsquo;Government\u0026rsquo; are most prevalent in Asia\u0026rsquo;s discourse. However, the continent exhibits greater thematic breadth and less depth relative to other continents. For Oceania, \u0026lsquo;Surveillance and Prevention\u0026rsquo; along with \u0026lsquo;Crop Impacts\u0026rsquo; dominate both its intra- and inter-continental discourse (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eProgression of invasion phases by region\u003c/h2\u003e\u003cp\u003eConsistent with our idealized model, all three continents exhibit early emphasis on a Surveillance phase (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). A progression from a Surveillance phase to an Impact one, followed by a Control phase, is evident across regions. In Africa, however, the Impact phase is more pronounced and sustained, while in Asia it appears to a lesser extent (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eLimitations of data sources and methods\u003c/h2\u003e\u003cp\u003eAn important consideration in this study is the extent to which the captured themes and their temporal patterns can be interpreted as cascading impacts. The captured themes do not necessarily reflect the exact reactions on the ground or measures being implemented. Instead, they indicate the importance or focus placed on these themes by media outlets. In the early stages of invasion, this focus may include concerns about potential or anticipated impacts that have not yet materialized. As such, some themes that appear early in the invasion timeline may reflect expectations about future consequences, rather than observed outcomes. The overall global trends are also likely to be heavily influenced by countries or regions with a larger number of articles available in English.\u003c/p\u003e\u003cp\u003eAdditionally, a theme may signify different practices or strategies depending on governmental, institutional or geographical contexts. Although we identified overarching themes within regional discourse, not all countries are proportionally represented in available news articles, so country-level variation could not be assessed. The more recent invasion of Asia and Oceania also affects the quantity of available news coverage, as well as availability of FAW reporting data underlining assessment of impacts. This is particularly the case for China, Taiwan, India and South-East Asian countries, where documentation of yield losses due to FAW is limited (Overton et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAnother challenge to identifying cascading impacts triggered by the introduction of FAW or other pests is that concerns about market access may incentivize countries to delay or avoid reporting invasive species. In Asia there have been reporting delays and inconsistencies that are known to have affected the perceived invasion pattern of FAW, supporting the \u0026ldquo;out of Africa\u0026rdquo; thesis (Kenis et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eMedia theme patterns in relation to invasion model\u003c/h2\u003e\u003cp\u003eThe observed phase trends may reflect an interplay between real-world invasion dynamics and media production practices. The progression of phases partly aligns with our idealized model predictions, particularly regarding an early Surveillance phase. However, the subsequent emergence of an Impact phase followed by a Control phase may indicate that consequences of the invasion unfolded rapidly and that control efforts were reactive, that early coverage included anticipated outcomes, or that it relied on anecdotal evidence and preliminary data. Alternatively, media outlets initially emphasized these topics, but attention subsequently declined following the issue-attention cycle, or the outbreak itself was effectively controlled.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eInterpreting continental thematic variation\u003c/h2\u003e\u003cp\u003eContinental variations in these media representations of fall armyworm invasions appear to reflect the actual conditions and responses occurring within these regions. For Oceania, the prominence of the theme \u0026lsquo;Surveillance \u0026amp; Prevention\u0026rsquo; aligns with the region\u0026rsquo;s demonstrated biosecurity measures and infrastructure. Australia and New Zealand, in particular, have rigorous biosecurity protocols and well-established institutions stemming from their need to protect their unique, geographically isolated ecosystems and significant agricultural sectors from invasive species. These measures are often regarded as more stringent than those of many other countries (Stone \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Additionally, the earlier outbreaks of FAW in Africa and Asia may have provided countries in Oceania advance warning, potentially increasing regional discussions around prevention and preparedness.\u003c/p\u003e\u003cp\u003eBoth pest control methods themes are more predominantly emphasized in the discourse surrounding Africa and Asia. In Africa, prior to the invasion of FAW, the use of chemical insecticides among maize producers was minimal (Kenis et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Post-invasion, synthetic pesticides have become the most common management method for FAW in Africa (Niassy et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The government of Ghana, for instance, after having suffered financial losses, reacted by widely deploying insecticides (Safo et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Additionally, the adoption of genetically modified crops on the African continent expanded beyond South Africa with the introduction of programs such as TELA maize in Nigeria (Orchardson \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Similarly, in China, the invasion has led to increased pesticide expenditure among smallholder farmers, with local maize growers in Yunnan primarily relying on pesticides for FAW management (Yang et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In a survey conducted involving farmers from several South Asian countries, respondents reported using an assortment of management practices, with chemical pesticides being the dominant strategy (Khan et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The theme of \u0026lsquo;Agroecological Control\u0026rsquo; also emerges in the discourse of these regions, potentially driven by development efforts. For example, following the FAW invasion in Sub-Saharan Africa, governments and international and non-governmental organizations discussed training farmers and agricultural stakeholders on constructing integrated pest management (IPM) approaches for FAW control and management (Matova et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) .\u003c/p\u003e\u003cp\u003eWith regard to impact types, \u0026lsquo;Social Impacts\u0026rsquo; proves more pronounced than \u0026lsquo;Crop Impacts\u0026rsquo; in Africa and Asia, where subsistence farming is widespread and harm to crops has more direct social consequences (Giller et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In both regions, FAW has disrupted smallholder farming systems, exacerbated household hunger, reduced incomes, and created other negative impacts on human health and the environment (Midega et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eExtending news-driven frameworks to cascading impacts\u003c/h2\u003e\u003cp\u003eGiven that several temporal and regional thematic trends correspond to an existing biological invasion framework and real-world phenomena, this analytical approach demonstrates potential for identifying cascading impacts. This capability is increasingly needed, particularly considering the ongoing migration of FAW from North Africa toward Southern Europe, a process already documented by recent research (Wang et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Kartakis et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Utilizing structural topic modeling or comparable analytical methods focused on news media could provide timely monitoring of this evolving threat in Europe, thus supporting improved response strategies. Existing news-scraping tools such as PADI-web (Platform for Automated extraction of Disease Information from the web) could be broadened beyond their current focus on animal and plant pathogens to track other pest species as well as their downstream impacts (Valentin et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Roche et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHowever, the current study is limited in its scope due to the focus on a single pest species and English-language sources. Future research should look to incorporate a wider range of pest species and a multilingual dataset to strengthen the robustness and applicability of the findings. Additionally, future studies could conduct a form of linguistic analysis, such as examining verb tense, to distinguish between projected and reported events in the text, facilitating the exploration of actual cascading effects.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eStructural topic model (STM) analysis reveals regional variations in popular news media discourse surrounding the FAW invasion, highlighting the influence of geographical context on the thematic evolution of FAW-related narratives. In particular, in Oceania, the continent where the FAW invasion is more recent, the discourse focuses more on preparatory actions compared to Africa and Asia. Additionally, all regions follow a broadly similar progression of biological invasion phases: Surveillance, Impact, and Control. Africa features a longer and proportionally more substantial Impact phase, providing insight into the cascading dynamics of such events. When combined with other data sources, findings from structural topic modeling of news media can contribute to a broader understanding of the cascading impacts of emerging agricultural pests.\u003c/p\u003e\u003cp\u003eUnderstanding these regional differences in discourse, and the underlying social-ecological impacts they reflect, is necessary for developing targeted and effective strategies to manage the consequences of FAW invasions. This need is particularly pressing as FAW has the potential to establish overwintering populations in Southern Europe. Applying STM or similar news media analysis methods to monitor evolving discourse in Europe could enhance preparedness and inform timely responses to this growing threat.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions\u0026nbsp;\u003c/strong\u003eKB conducted the initial research for this study as part of the thesis requirement for the Social-ecological Resilience for Sustainable Development (SERSD) MSc program at the Stockholm Resilience Centre, Stockholm University, Stockholm, Sweden. All authors contributed to the study conception and design. KB performed material preparation and data collection. All authors performed data analysis. KB wrote the first draft of the manuscript and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003eWe acknowledge funding from the Erling-Persson Family Foundation (PSJ, Grant number: N/A), the European Union (ERC, PSJ, INFLUX, Grant number: 101039376), the IKEA foundation (PSJ, Grant number: N/A), the Marianne and Marcus Wallenberg Foundation (PSJ, Grant number: N/A), and the Swedish Research Council for Sustainable Development\u0026nbsp;Formas (SD, Grant number:\u0026nbsp;2020-01586). The funders had no role in the design, analysis, or writing of this paper. Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authorities can be held responsible for them.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u0026nbsp;\u003c/strong\u003eThe full news articles used in this study are subject to copyright and cannot be shared. The code to reproduce the dataset is available at https://github.com/kathrynbjorklund/tracking-faw-invasion-phases-via-news-media.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003eThe authors have no relevant financial or non-financial interests to disclose.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u0026nbsp;\u003c/strong\u003eThis study analyzed publicly available news articles and did not involve human participants, animals, or sensitive data. Therefore, no formal ethics review was required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u0026nbsp;\u003c/strong\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to publish\u0026nbsp;\u003c/strong\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBale JS, van Lenteren JC, Bigler F (2008) Biological control and sustainable food production. Philos Trans R Soc Lond B Biol Sci 363:761\u0026ndash;776. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rstb.2007.2182\u003c/span\u003e\u003cspan address=\"10.1098/rstb.2007.2182\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBr\u0026eacute;vault T, Bouyer J (2014) From Integrated to System-Wide Pest Management: Challenges for Sustainable Agriculture. Outlook Pest Man 25:212\u0026ndash;213. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1564/v25_jun_05\u003c/span\u003e\u003cspan address=\"10.1564/v25_jun_05\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDay R, Abrahams P, Bateman M et al (2017) Fall Armyworm: Impacts and Implications for Africa. Outlooks Pest Manage 28:196\u0026ndash;201. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1564/v28_oct_02\u003c/span\u003e\u003cspan address=\"10.1564/v28_oct_02\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDevi S (2018) Fall armyworm threatens food security in southern Africa. Lancet 391:727. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0140-6736(18)30431-8\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(18)30431-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiffenbaugh NS (2022) COVID-19 and the Environment: Short-Run and Potential Long-Run Impacts. Annu Rev Environ Resour 47:65\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev-environ-120920-125207\u003c/span\u003e\u003cspan address=\"10.1146/annurev-environ-120920-125207\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDupre SI, Harvey CA, Holland MB (2022) The impact of coffee leaf rust on migration by smallholder coffee farmers in Guatemala. World Dev 156:105918. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.worlddev.2022.105918\u003c/span\u003e\u003cspan address=\"10.1016/j.worlddev.2022.105918\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEschen R, Beale T, Bonnin JM et al (2021) Towards estimating the economic cost of invasive alien species to African crop and livestock production. CABI Agric Bioscience 2:18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s43170-021-00038-7\u003c/span\u003e\u003cspan address=\"10.1186/s43170-021-00038-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiller KE, Delaune T, Silva JV et al (2021) The future of farming: Who will produce our food? Food Sec 13:1073\u0026ndash;1099. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12571-021-01184-6\u003c/span\u003e\u003cspan address=\"10.1007/s12571-021-01184-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKartakis S, Horrocks KJ, Cingiz K et al (2025) Migration extent and potential economic impact of the fall armyworm in Europe. Sci Rep 15:17405. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-025-02595-7\u003c/span\u003e\u003cspan address=\"10.1038/s41598-025-02595-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKenis M, Benelli G, Biondi A et al (2023) Invasiveness, biology, ecology, and management of the fall armyworm, Spodoptera frugiperda. entomologia 43:187\u0026ndash;241. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1127/entomologia/2022/1659\u003c/span\u003e\u003cspan address=\"10.1127/entomologia/2022/1659\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKhan FZA, Paudel S, Saeed S et al (2023) Mitigating the impact of the invasive fall armyworm: evidence from South Asian farmers and policy recommendations. Int J Pest Manage 1\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/09670874.2023.2205834\u003c/span\u003e\u003cspan address=\"10.1080/09670874.2023.2205834\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLu Y, Adang MJ (1996) Distinguishing Fall Armyworm (Lepidoptera: Noctuidae) Strains Using a Diagnostic Mitochondrial Dna Marker. Fla Entomol 48\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/3495753\u003c/span\u003e\u003cspan address=\"10.2307/3495753\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatova PM, Kamutando CN, Magorokosho C et al (2020) Fall-armyworm invasion, control practices and resistance breeding in Sub-Saharan Africa. Crop Sci 60:2951\u0026ndash;2970. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/csc2.20317\u003c/span\u003e\u003cspan address=\"10.1002/csc2.20317\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMidega C, Hadi B, McGuire S et al (2023) Fall armyworm: measuring damage and loss caused by a novel invasive pest as a guide to sustainable management. FAO\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNiassy S, Agbodzavu MK, Kimathi E et al (2021) Bioecology of fall armyworm Spodoptera frugiperda (J. E. Smith), its management and potential patterns of seasonal spread in Africa. PLoS ONE 16:e0249042. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0249042\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0249042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOrchardson E (2024) TELA Maize Project. In: CIMMYT. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cimmyt.org/projects/tela-maize-project/\u003c/span\u003e\u003cspan address=\"https://www.cimmyt.org/projects/tela-maize-project/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Accessed 20 Jan 2025\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOverton K, Maino JL, Day R et al (2021) Global crop impacts, yield losses and action thresholds for fall armyworm (\u003cem\u003eSpodoptera frugiperda\u003c/em\u003e): A review. Crop Prot 145:105641. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cropro.2021.105641\u003c/span\u003e\u003cspan address=\"10.1016/j.cropro.2021.105641\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePrasanna BM, Bruce A, Beyene Y et al (2022) Host plant resistance for fall armyworm management in maize: relevance, status and prospects in Africa and Asia. Theor Appl Genet 135:3897\u0026ndash;3916. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00122-022-04073-4\u003c/span\u003e\u003cspan address=\"10.1007/s00122-022-04073-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoberts ME, Stewart BM, Tingley D (2019) stm: An \u003cem\u003eR\u003c/em\u003e Package for Structural Topic Models. J Stat Soft 91. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18637/jss.v091.i02\u003c/span\u003e\u003cspan address=\"10.18637/jss.v091.i02\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoche M, Rabatel J, Trevennec C, Pieretti I (2024) PADI-web for Plant Health Surveillance. Intelligent Information Systems. Springer, Cham, pp 148\u0026ndash;156. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-031-61000-4_17\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-61000-4_17\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSafo A, Avicor SW, Baidoo PK et al (2023) Farmers\u0026rsquo; knowledge, experience and management of fall armyworm in a major maize producing municipality in Ghana. Cogent Food Agric 9:2184006. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/23311932.2023.2184006\u003c/span\u003e\u003cspan address=\"10.1080/23311932.2023.2184006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStone M (2021) Biosecurity is critical to New Zealand\u0026rsquo;s national security, economy and way of life. N Z Vet J 69:309\u0026ndash;312. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00480169.2021.1965076\u003c/span\u003e\u003cspan address=\"10.1080/00480169.2021.1965076\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTambo JA, Kansiime MK, Mugambi I et al (2020) Understanding smallholders\u0026rsquo; responses to fall armyworm (\u003cem\u003eSpodoptera frugiperda\u003c/em\u003e) invasion: Evidence from five African countries. Sci Total Environ 740:140015. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2020.140015\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2020.140015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTateosian LG, Saffer A, Walden-Schreiner C, Shukunobe M (2023) Plant pest invasions, as seen through news and social media. Comput Environ Urban Syst 100:101922. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.compenvurbsys.2022.101922\u003c/span\u003e\u003cspan address=\"10.1016/j.compenvurbsys.2022.101922\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eValentin S, Arsevska E, Rabatel J et al (2021) PADI-web 3.0: A new framework for extracting and disseminating fine-grained information from the news for animal disease surveillance. One Health 13:100357. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.onehlt.2021.100357\u003c/span\u003e\u003cspan address=\"10.1016/j.onehlt.2021.100357\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang J, Huang Y, Huang L et al (2023) Migration risk of fall armyworm (Spodoptera frugiperda) from North Africa to Southern Europe. Front Plant Sci 14:1141470. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fpls.2023.1141470\u003c/span\u003e\u003cspan address=\"10.3389/fpls.2023.1141470\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWelsh MJ, Turner JA, Epanchin-Niell RS et al (2021) Approaches for estimating benefits and costs of interventions in plant biosecurity across invasion phases. Ecol Appl 31:e02319. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/eap.2319\u003c/span\u003e\u003cspan address=\"10.1002/eap.2319\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang X, Wyckhuys KAG, Jia X et al (2021) Fall armyworm invasion heightens pesticide expenditure among Chinese smallholder farmers. J Environ Manage 282:111949. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2021.111949\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2021.111949\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"biological-invasions","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binv","sideBox":"Learn more about [Biological Invasions](https://www.springer.com/journal/10530)","snPcode":"10530","submissionUrl":"https://submission.nature.com/new-submission/10530/3","title":"Biological Invasions","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"social-ecological impacts, fall armyworm, Spodoptera frugiperda, text analysis, digital news articles","lastPublishedDoi":"10.21203/rs.3.rs-7393018/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7393018/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eUnderstanding the dynamics of cascading social-ecological impacts associated with emerging pests and pathogens is key for addressing the challenges they introduce in an ever more interconnected and rapidly changing world. Here, we used topic modeling of digital news articles to investigate the potential cascading social-ecological impacts associated with the ongoing fall armyworm invasion of multiple geographic regions. We identified regional thematic shifts in the popular news media discourse surrounding the fall armyworm invasion. In the news discourse in Oceania, we discerned a stronger focus on invasion preparation than in regions like Africa and Asia. Additionally, we observed a common biological invasion phase pattern across regions, with Africa distinguished by a longer and proportionally larger impacts-related phase. These regional variations illuminate localized priorities in addressing this invasive species. By highlighting the significance of applying machine learning techniques to news articles to identify and describe cascading social-ecological impacts of emerging pests and pathogens, we can improve our understanding of these patterns and inform more targeted management and mitigation strategies.\u003c/p\u003e","manuscriptTitle":"Tracking phases of the fall armyworm (Spodoptera frugiperda) invasion across multiple continents using news media","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-04 14:05:00","doi":"10.21203/rs.3.rs-7393018/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2025-08-29T10:45:29+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-28T10:58:55+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Biological Invasions","date":"2025-08-21T09:36:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-18T09:47:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Biological Invasions","date":"2025-08-17T10:45:56+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"biological-invasions","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"binv","sideBox":"Learn more about [Biological Invasions](https://www.springer.com/journal/10530)","snPcode":"10530","submissionUrl":"https://submission.nature.com/new-submission/10530/3","title":"Biological Invasions","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e4625b53-0a2e-4fca-b3f9-f8c6b6fa0ce1","owner":[],"postedDate":"September 4th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-01-03T13:54:49+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-04 14:05:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7393018","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7393018","identity":"rs-7393018","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}