{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp11_mechanism_context_external20_20260804","research_id":"eoa_inverse_innovation_exp11_external_scrutiny_20260804","cell_id":"invariant_mode_decomposition_design__biology_ecology","opaque_id":"invariant_mode_decomposition_design__biology_ecology__C","search_lanes":{"direct_problem":{"queries":["dynamic mode decomposition lake phytoplankton dissolved oxygen hypoxia early warning","mesocosm nutrient pulse phytoplankton bloom hypoxia dissolved oxygen experiment","multivariate early warning signals eutrophication lake regime shift mesocosm"],"source_ids":["SRC1","SRC2","SRC3","SRC8"],"no_result_note":null},"closest_prior_art":{"queries":["\"dynamic mode decomposition\" ecology ecosystem","\"dynamic mode decomposition\" phytoplankton","\"Koopman\" lake ecosystem eutrophication","\"dynamic mode decomposition\" \"dissolved oxygen\""],"source_ids":["SRC4","SRC5","SRC8"],"no_result_note":null},"historical_terminology":{"queries":["ecological early warning \"dominant eigenvalue\" Jacobian multivariate time series","lake eutrophication \"leading eigenvalue\" early warning","multivariate autoregressive model eigenvalue early warning ecosystem transition","ecosystem stability generalized modeling eigenvalues eutrophication plankton oxygen"],"source_ids":["SRC2","SRC3","SRC4"],"no_result_note":null},"products_practices_standards":{"queries":["site:epa.gov cyanobacteria monitoring early warning dissolved oxygen chlorophyll lake managers guidance","site:who.int cyanobacterial blooms monitoring dissolved oxygen chlorophyll guidance freshwater","ISO 17289 dissolved oxygen optical sensor water quality standard official","commercial lake harmful algal bloom early warning system dissolved oxygen chlorophyll sensor product"],"source_ids":["SRC6","SRC7","SRC8"],"no_result_note":null},"non_english_regional":{"queries":["mesocosmos eutrofización hipoxia \"oxígeno disuelto\" alerta temprana","Frühwarnsystem Eutrophierung Sauerstoff Mesokosmos Algenblüte","湖泊 富营养化 缺氧 预警 多变量 模态 分解","湖沼 富栄養化 貧酸素 早期警戒 メソコスム","\"富营养化浅水湖泊藻源性湖泛的短期数值预报方法\""],"source_ids":["SRC8"],"no_result_note":null},"composition_subproblems":{"queries":["Koopman operator harmful algal bloom dissolved oxygen ecosystem early warning","mesocosm nutrient pulse phytoplankton bloom hypoxia dissolved oxygen experiment","\"Physics Informed Modeling of Ecosystem Respiration via Dynamic Mode Decomposition with Control Input\"","\"Early warning signals to anticipate critical transitions in high-dimensional systems\""],"source_ids":["SRC2","SRC4","SRC5","SRC8"],"no_result_note":null}},"sources":[{"source_id":"SRC1","title":"What Are the Effects of HABs","url":"https://www.epa.gov/habs/what-are-effects-habs","publisher":"United States Environmental Protection Agency","date_or_year":"Updated 2026-06-23","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["High-biomass algal blooms can produce hypoxia or anoxia and increase mortality of plants, invertebrates, shellfish, and fish.","Bloom-driven oxygen loss can alter sediment biochemistry and reinforce ecosystem degradation."]},{"source_id":"SRC2","title":"Time scale of resilience loss: Implications for managing critical transitions in water quality","url":"https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0223366","publisher":"PLOS ONE","date_or_year":"2019-10-07","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["A whole-lake experiment used daily nutrient fertilization to induce an algal transition and halted fertilization after an early-warning signal was detected.","High-frequency chlorophyll measurements and a reference lake supported prospective comparison of warning behavior.","The study found autocorrelation-warning behavior broadly robust across tested aggregation scales, demonstrating a feasible warning-triggered intervention precedent."]},{"source_id":"SRC3","title":"Early warning signals have limited applicability to empirical lake data","url":"https://www.nature.com/articles/s41467-023-43744-8","publisher":"Nature Communications","date_or_year":"2023","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Across empirical multi-trophic lake records, early-warning performance was strongly dependent on preprocessing, temporal resolution, and indicator choice.","Multivariate expanding-window methods were only modestly better than chance on average, and no tested indicator was consistently reliable across true-positive and true-negative scenarios.","The study identifies multivariate Jacobian and autoregressive stability approaches as promising but cautions that ecosystem-specific validation is necessary."]},{"source_id":"SRC4","title":"Early warning signals to anticipate critical transitions in high-dimensional systems","url":"https://sciety.org/articles/activity/10.21203/rs.3.rs-7865028/v1","publisher":"Research Square preprint, indexed by Sciety","date_or_year":"2025-11-11","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["MM-DEV estimates a multivariate dynamical-eigenvalue warning after nonlinear dimension reduction.","The method was validated on a controlled methanogenic-ecosystem collapse experiment using chemical and microbial time series.","It was also applied to Lake Zurich plankton data and reportedly anticipated a food-web regime shift, making it close mechanism-level ecological prior art."]},{"source_id":"SRC5","title":"Physics Informed Modeling of Ecosystem Respiration via Dynamic Mode Decomposition with Control Input","url":"https://iclr.cc/virtual/2024/22022","publisher":"International Conference on Learning Representations, ML4RS Workshop","date_or_year":"2024","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Dynamic mode decomposition with control input has been applied to ecosystem-respiration data.","The method separates autonomous dynamics from observed environmental inputs and permits controlled-input analysis.","Its reported predictive accuracy was comparable to other methods, supporting technical feasibility but not hypoxia-warning or intervention efficacy."]},{"source_id":"SRC6","title":"EPA Tools for Waterbody Managers to Monitor for and Respond to CyanoHABs","url":"https://www.epa.gov/habs/epa-tools-waterbody-managers-monitor-and-respond-cyanohabs","publisher":"United States Environmental Protection Agency","date_or_year":"Updated 2026-05-18","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Waterbody managers are an explicitly identified user group for bloom-monitoring and response tools.","EPA maintains planning, monitoring, forecasting, advisory, and response resources for cyanobacterial blooms."]},{"source_id":"SRC7","title":"ISO 17289:2014 — Water quality — Determination of dissolved oxygen — Optical sensor method","url":"https://www.iso.org/standard/59515.html","publisher":"International Organization for Standardization","date_or_year":"2014; confirmed 2026","source_type":"OFFICIAL_STANDARD","language":"English","claims_supported":["ISO 17289 specifies an optical dissolved-oxygen measurement method suitable for laboratory, field, and continuous monitoring.","The standard addresses low-oxygen measurements and algal-growth supersaturation, supporting standardized mesocosm instrumentation and quality control."]},{"source_id":"SRC8","title":"蓝藻颗粒物堆积预警模拟技术 (Cyanobacterial particulate accumulation early-warning simulation technology)","url":"https://bee.wuxi.gov.cn/doc/2021/03/01/3207769.shtml","publisher":"Wuxi Municipal Bureau of Ecology and Environment; technology supplied by Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences","date_or_year":"2021-03-01","source_type":"FIRST_PARTY_PRODUCT","language":"Chinese","claims_supported":["An operational Taihu platform combines ecological-dynamic and weather models with field monitoring and remote sensing to forecast chlorophyll and dissolved oxygen over seven days.","The platform estimates bloom and black-bloom probability, location, area, and risk level, then issues routine reports to management departments.","From 2017 to 2020 it reportedly issued more than 160 reports, predicted more than ten small black-bloom events, supported emergency response, and was extended to other drinking-water sources.","This falsifies any broad implication that operational practice is limited to independent fixed thresholds, although it does not use transition-operator modes or modal intervention projection."]}],"problem_evidence":{"status":"PARTLY_SUPPORTED","finding":"The underlying hazard is well supported: nutrient-driven algal transitions can precede severe oxygen loss and organism harm, and prospective warning is an acknowledged management problem. Nutrient-manipulation experiments demonstrate warning-triggered cessation of fertilization, while empirical lake studies show that existing univariate and multivariate warnings can be unreliable. However, no retained source directly establishes that a persistent nutrient–plankton–respiration–oxygen transition-operator mode exists in shallow freshwater mesocosms or that separate thresholds specifically miss it. Moreover, an operational Chinese platform already forecasts chlorophyll and dissolved oxygen jointly with hydrodynamic and weather variables, so the proposal's baseline is too narrow if presented as universal practice.","source_ids":["SRC1","SRC2","SRC3","SRC8"],"uncertainty":"Evidence spans whole lakes, long-term lake records, and operational forecasting rather than the exact replicated shallow-mesocosm setting. Bloom occurrence does not always imply hypoxia, and high-frequency diel oxygen minima may be aliased by sparse sampling."},"adopter_evidence":{"status":"SUPPORTED","finding":"Identifiable adopters and authorizers exist. EPA explicitly targets waterbody managers with monitoring and response tools. The Taihu platform routinely delivered forecasts to management departments and identifies the Nanjing Institute of Geography and Limnology as the technical operator. For the bounded experiment, the responsible mesocosm investigator is the immediate authorizer, while waterbody managers are plausible downstream adopters if comparative performance is demonstrated.","source_ids":["SRC6","SRC8"],"uncertainty":"The sources establish user roles and an operational comparator, not a named organization committed to adopting this particular modal method."},"implementation_evidence":{"status":"PARTLY_SUPPORTED","finding":"The component capabilities are feasible: standardized continuous dissolved-oxygen sensing exists; high-frequency nutrient-addition experiments with warning-triggered cessation have been conducted; DMD with environmental control inputs has been applied to ecosystem data; multivariate dynamical-eigenvalue warnings have been tested on ecological time series; and operational systems already integrate chlorophyll, oxygen, weather, and ecological dynamics. The unvalidated step is the proposed combination of a separable local transition mode, preregistered spectral-gap and residual criteria, and intervention selection by negative projection onto that mode.","source_ids":["SRC2","SRC4","SRC5","SRC7","SRC8"],"uncertainty":"No retained source demonstrates that the required sampling density and replication yield stable mode estimates in shallow mesocosms, or that modal projection supplies causal intervention guidance rather than descriptive association."},"prior_art":{"disposition":"ADJACENT_PRIOR_ART","closest_analogues":[{"name":"Taihu cyanobacterial bloom and black-bloom monitoring and early-warning platform","source_ids":["SRC8"],"same_problem":true,"same_causal_lever":false,"overlap":"Operationally combines ecological dynamics, weather, remote sensing, chlorophyll, and dissolved oxygen to forecast bloom-related low-oxygen events and support management response.","remaining_difference":"It uses coupled process and probability models rather than estimating a local transition operator, selecting an amplifying eigenmode, checking a spectral gap and reconstruction residual, or choosing treatment by modal projection."},{"name":"Manifold-projected Multivariate Dynamical Eigenvalue (MM-DEV)","source_ids":["SRC4"],"same_problem":false,"same_causal_lever":true,"overlap":"Uses high-dimensional ecological time series, local dynamical-eigenvalue behavior, and a dominant stability quantity to anticipate ecosystem transitions; includes experimental and lake-plankton applications.","remaining_difference":"It does not target nutrient-pulse hypoxia, reconstruct a nutrient–plankton–respiration–oxygen mode in mesocosms, or experimentally test interventions selected to reduce that mode's gain."},{"name":"Whole-lake fertilization with autocorrelation-triggered nutrient cessation","source_ids":["SRC2"],"same_problem":true,"same_causal_lever":false,"overlap":"Manipulates nutrients, monitors an algal transition at high frequency, detects a prospective dynamical warning, and stops fertilization in response.","remaining_difference":"The warning is primarily single-variable autocorrelation rather than a multivariate transition mode, and the endpoint is bloom onset rather than demonstrated hypoxia prevention."},{"name":"Dynamic mode decomposition with control input for ecosystem respiration","source_ids":["SRC5"],"same_problem":false,"same_causal_lever":true,"overlap":"Applies a best-fit linear dynamical representation with environmental control inputs to ecosystem time series.","remaining_difference":"It concerns terrestrial ecosystem respiration prediction, not coupled aquatic bloom–hypoxia warning, separability thresholds, or randomized intervention outcomes."},{"name":"Empirical multivariate lake early-warning and stability methods","source_ids":["SRC3"],"same_problem":false,"same_causal_lever":true,"overlap":"Evaluates multivariate dimension-reduction indicators, dominant-eigenvalue-related indicators, and Jacobian or autoregressive stability concepts on empirical lake plankton records.","remaining_difference":"It does not test the proposal's synchronized state vector, prospective hypoxia endpoint, mode-specific treatment map, or replicated nutrient-pulse intervention."}],"contrastive_claim_remaining":"In held-out replicated shallow freshwater mesocosms receiving nutrient pulses, a preregistered local transition-operator mode built from synchronized nutrients, chlorophyll, zooplankton, microbial respiration, temperature, and dissolved oxygen provides earlier and more reliable warning of severe oxygen depletion than the best single-variable thresholds and competitive multivariate or process-model baselines; when a bounded treatment is selected for negative projection onto that mode, it reduces out-of-sample modal gain and improves subsequent oxygen trajectories relative to matched controls.","contrastive_claim_falsifier":"The claim is falsified if no reproducible mode meets the preregistered spectral-gap and reconstruction-residual criteria; if held-out warning lead time, sensitivity, specificity, or calibration is no better than strong oxygen, chlorophyll, autocorrelation, multivariate autoregressive, and process-model baselines; or if mode-projected nutrient reduction or aeration does not reduce modal gain and improve oxygen outcomes relative to controls.","confidence":"MODERATE","search_limitations":"The search was deliberately bounded to eight retained direct sources. It covered English and regional Chinese, German, Spanish, and Japanese terminology; dominant-eigenvalue, Jacobian, critical-slowing-down, Koopman, and DMD terminology; standards; operational products; and component combinations. Relevant patents, paywalled literature, unpublished deployments, and additional regional terminology may remain. The MM-DEV item is a preprint, and the retained operational platform description reports first-party performance rather than an independent audit."},"researchability_gates":{"externally_supported_problem":{"status":"PASS","rationale":"Bloom-driven hypoxia and organism harm are officially documented, nutrient-driven lake transitions have been experimentally manipulated, and empirical evidence shows substantial warning uncertainty. The exact modal explanation remains appropriately testable rather than being assumed.","source_ids":["SRC1","SRC2","SRC3","SRC8"]},"identifiable_adopter_or_authorizer":{"status":"PASS","rationale":"Waterbody managers are explicit users of bloom-monitoring and response tools, management departments already receive operational coupled forecasts, and a responsible investigator can authorize the bounded mesocosm study.","source_ids":["SRC6","SRC8"]},"distinct_testable_incremental_claim":{"status":"PASS","rationale":"Close prior art covers operational multivariate forecasting, ecological dynamical eigenvalues, DMD with controls, and warning-triggered nutrient cessation, but not the combined held-out claim that a separable nutrient–plankton–respiration–oxygen transition mode improves hypoxia warning and selects an intervention that reduces mode gain and oxygen harm.","source_ids":["SRC2","SRC3","SRC4","SRC5","SRC8"]},"bounded_next_evidence_step":{"status":"PASS","rationale":"A preregistered retrospective comparison followed by a small randomized, replicated mesocosm experiment is bounded and uses demonstrated nutrient-manipulation, automated-sensing, and ecosystem-dynamics methods. Prespecified residual, spectral-gap, warning, and oxygen endpoints permit a clear stop or null result.","source_ids":["SRC2","SRC5","SRC7"]},"no_unresolved_safety_or_authority_stop":{"status":"PASS","rationale":"No unresolved stop emerged for an approved, contained mesocosm study using bounded nutrient pulses, no environmental release, standardized continuous oxygen monitoring, predefined welfare or water-quality limits, and rescue aeration. The responsible investigator remains accountable for institutional approvals and disposal requirements.","source_ids":["SRC2","SRC7"]},"adequate_search_evidence":{"status":"PASS","rationale":"All six required lanes were searched adversarially. The eight retained sources include multiple independent publishers and seven primary, official, standards, or first-party sources, with both positive evidence and strong counterevidence to generic early-warning performance.","source_ids":["SRC1","SRC2","SRC3","SRC4","SRC5","SRC6","SRC7","SRC8"]}},"strict_success":true,"screen_survival":true,"remaining_research_value":"MODERATE","recommended_next_step":"Preregister a two-stage test. First, use existing high-frequency mesocosm or whole-lake data to freeze state variables, sampling interval, local-window length, spectral-gap and residual cutoffs, hypoxia definition, and comparison models: dissolved-oxygen and chlorophyll thresholds, autocorrelation, multivariate autoregression or Jacobian stability, and a coupled process or statistical forecast. Require genuinely held-out mesocosms and report lead time, sensitivity, specificity, calibration, residual oxygen error, and mode stability. Only if that stage passes, run a small randomized nutrient-pulse mesocosm experiment with control, threshold-triggered treatment, and mode-projected treatment arms; test whether the proposed arm reduces estimated gain and improves minimum and cumulative low-oxygen exposure, with predefined rescue aeration and termination rules.","world_novelty_boundary":"This bounded search supports only an adjacent-prior-art research claim. It cannot establish world novelty, patentability, freedom to operate, market size, routine generalizability beyond the sampled systems, or realized ecological impact."}