{"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__A","search_lanes":{"direct_problem":{"queries":["shallow lake mesocosm eigenvalue leading mode restoration control macrophyte fish removal","early warning signals shallow lake regime shift mesocosm coupled indicators","early warning signals have limited applicability empirical lake data multivariate"],"source_ids":["SRC1","SRC2","SRC5"],"no_result_note":null},"closest_prior_art":{"queries":["dynamic mode decomposition ecology lake ecosystem regime shift","multivariate autoregressive model eigenvalues ecological stability early warning","probabilistic multivariate early warning signals vector autoregression ecology"],"source_ids":["SRC3","SRC4","SRC5","SRC8"],"no_result_note":null},"historical_terminology":{"queries":["biomanipulation shallow lakes fish removal macrophyte restoration","lake restoration critical nutrient loading food web model PCLake","vector autoregressive MAR freshwater plankton community stability"],"source_ids":["SRC3","SRC6","SRC8"],"no_result_note":null},"products_practices_standards":{"queries":["OECD aquatic mesocosm guidance document official","PCLake model official STOWA shallow lakes ecological model","lake restoration biomanipulation fish removal macrophyte planting guidance"],"source_ids":["SRC6","SRC7","SRC8"],"no_result_note":null},"non_english_regional":{"queries":["生态修复 浅水湖泊 鱼类去除 沉水植物 营养盐 联合 mesocosm","浅水湖泊模型 PCLake 应用","PCLake en PCDitch kritische nutriëntenbelasting waterbeheerders"],"source_ids":["SRC6","SRC8"],"no_result_note":null},"composition_subproblems":{"queries":["shallow lake restoration combined nutrient reduction fish removal macrophyte transplantation experiment","mesocosm phosphorus reduction fish removal macrophyte addition shallow lake factorial","regularized transition model randomized control ecological community eigenmode"],"source_ids":["SRC1","SRC3","SRC4","SRC5","SRC6","SRC7"],"no_result_note":null}},"sources":[{"source_id":"SRC1","title":"Early Warnings of Regime Shifts: A Whole-Ecosystem Experiment","url":"https://www.caryinstitute.org/sites/default/files/public/reprints/Carpenter_et_al_2011_Science.pdf","publisher":"Science / American Association for the Advancement of Science","date_or_year":"2011","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["An experimentally manipulated aquatic food web produced warning signals more than a year before its transition was complete.","Fish, zooplankton, phytoplankton, and chlorophyll changed through coupled trophic pathways.","The authors caution that warnings can fail under fast forcing, observation error, or interacting nonlinear processes and did not test reversal by intervention."]},{"source_id":"SRC2","title":"Early warning signals have limited applicability to empirical lake data","url":"https://www.nature.com/articles/s41467-023-43744-8","publisher":"Nature Communications / Nature Portfolio","date_or_year":"2023","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Across empirical multi-trophic lake data, most early-warning indicators did not perform better than chance.","Multivariate indicators were only weakly superior to univariate indicators, and results depended strongly on preprocessing and temporal resolution.","Different trophic levels within one lake can undergo different transition mechanisms, directly supporting concern about relying on separate indicators while challenging generic modal predictability."]},{"source_id":"SRC3","title":"Quantifying effects of abiotic and biotic drivers on community dynamics with multivariate autoregressive (MAR) models","url":"https://eeholmes.github.io/publications/Hamptonetal2013.pdf","publisher":"Ecological Society of America","date_or_year":"2013","source_type":"SECONDARY_RESEARCH","language":"English","claims_supported":["MAR models estimate interaction and environmental-driver matrices from multivariate ecological time series and yield community stability and perturbation-response metrics.","The literature already included extensive applications to freshwater plankton communities and planktivory manipulations.","Observation error, uneven sampling, openness, time-varying interactions, and limited data are recognized implementation problems requiring state-space extensions."]},{"source_id":"SRC4","title":"Probabilistic Multivariate Early Warning Signals","url":"https://research.utu.fi/converis/portal/detail/Publication/176070308?lang=en_GB","publisher":"Springer, via University of Turku Research Portal","date_or_year":"2022","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["A probabilistic time-varying vector-autoregressive approach was proposed to pool information across multivariate ecological time series.","Simulation on a multispecies ecological model showed improved warning sensitivity relative to alternatives.","The work concerns warning detection rather than experimentally identified control-to-mode effects or comparative modal control."]},{"source_id":"SRC5","title":"Eigenvalues of the covariance matrix as early warning signals for critical transitions in ecological systems","url":"https://www.nature.com/articles/s41598-019-38961-5","publisher":"Scientific Reports / Nature Portfolio","date_or_year":"2019","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["The largest covariance eigenvalue and its eigenvector were proposed as multivariate ecological warning signals and indicators of vulnerable regions.","The method was demonstrated theoretically and on simulated spatial ecological models under a codimension-one bifurcation and local-linearization assumption.","This is close directional prior art, but it decomposes covariance rather than a fitted transition operator and does not identify randomized intervention leverage."]},{"source_id":"SRC6","title":"Restoration of Shallow Lakes in Subtropical and Tropical China: Response of Nutrients and Water Clarity to Biomanipulation by Fish Removal and Submerged Plant Transplantation","url":"https://pure.au.dk/portal/en/publications/restoration-of-shallow-lakes-in-subtropical-and-tropical-china-re/","publisher":"Water / MDPI, record hosted by Aarhus University","date_or_year":"2016","source_type":"PRIMARY_RESEARCH","language":"English","claims_supported":["Fish removal followed by submerged-macrophyte transplantation is an implemented shallow-lake restoration package, not merely a conceptual proposal.","Three warm Chinese lakes showed lower nutrients, suspended solids, and chlorophyll and greater transparency after restoration.","Continued fish-biomass control was recommended, showing that intervention persistence and trophic side effects remain practical concerns."]},{"source_id":"SRC7","title":"Guidance Document on Simulated Freshwater Lentic Field Tests (Outdoor Microcosms and Mesocosms)","url":"https://www.oecd.org/en/publications/guidance-document-on-simulated-freshwater-lentic-field-tests-outdoor-microcosms-and-mesocosms_bb366ec4-en.html","publisher":"OECD Publishing","date_or_year":"2006","source_type":"OFFICIAL_GUIDANCE","language":"English","claims_supported":["Freshwater microcosms and mesocosms are recognized tools for testing specific hypotheses and bridging laboratory studies to natural-environment inference.","The guidance supports a bounded mesocosm evidence step but does not itself authorize animal, ecological, or facility manipulations.","Study-specific exposure design, endpoints, replication, and local approvals remain necessary."]},{"source_id":"SRC8","title":"PCLake en PCDitch","url":"https://www.stowa.nl/onderwerpen/waterkwaliteit/realiseren-van-ecologische-waterkwaliteitsdoelen-krw/pclake-en-pcditch","publisher":"STOWA","date_or_year":"2012–2017 project; current product page","source_type":"FIRST_PARTY_PRODUCT","language":"Dutch","claims_supported":["PCLake is an operational ecological model intended to help water managers evaluate water-quality measures in lakes.","It represents interacting lake properties, including hydrology, nutrients, physicochemical processes, water quality, and food webs.","It estimates system-specific critical nutrient loading, establishing an existing model-based alternative for restoration decisions without the proposal's empirically gated modal-control policy."]}],"problem_evidence":{"status":"PARTLY_SUPPORTED","finding":"The general problem is real: aquatic food webs can reorganize through coupled trophic changes, warning signals can precede experimental transitions, and separate empirical lake indicators are often unreliable. Multivariate and eigenvector-based signals already exist, but no retained source demonstrates a reproducible season-local transition eigenmode with useful lead time and controllability in the proposed shallow-lake mesocosm regime.","source_ids":["SRC1","SRC2","SRC4","SRC5"],"uncertainty":"Evidence is mixed and scale-dependent. The strongest positive result used an unusually slow, high-frequency whole-lake manipulation, whereas the broad empirical comparison found weak and preprocessing-sensitive predictability. The exact weekly 8–12-variable mesocosm hypothesis remains untested."},"adopter_evidence":{"status":"SUPPORTED","finding":"Identifiable adopter and authorizer classes exist: mesocosm investigators and risk-assessment facilities can conduct bounded tests, while water managers already use ecosystem models such as PCLake to assess lake measures. For the proposed experiment, the principal investigator, facility authority, permit holder, and applicable animal-care or ecological review bodies are the relevant authorizers.","source_ids":["SRC7","SRC8"],"uncertainty":"No named facility, principal investigator, permit holder, jurisdiction, or committed water-management adopter is supplied, so actual willingness, legal requirements, and capacity remain site-specific."},"implementation_evidence":{"status":"PARTLY_SUPPORTED","finding":"Most components are technically precedented: multivariate transition/state-space models, eigenvalue or eigenvector warning analyses, mesocosm hypothesis tests, and combined fish-removal/macrophyte restoration. The retained sources do not show the complete package of randomized control-to-mode identification, spectral and identity gates, disturbance-matched modal policy selection, and independent comparison against coordinate-wise and nonlinear nonmodal controllers.","source_ids":["SRC1","SRC3","SRC4","SRC5","SRC6","SRC7","SRC8"],"uncertainty":"Weekly data from 72 units may still be inadequate for a stable local operator after blocking, repeated-measure dependence, environmental forcing, nonlinearities, and intervention-induced mode rotation are accounted for. Feasible native species, doses, containment, and facility capacity are unverified."},"prior_art":{"disposition":"ADJACENT_PRIOR_ART","closest_analogues":[{"name":"Ecological multivariate autoregressive and state-space community models","source_ids":["SRC3"],"same_problem":true,"same_causal_lever":false,"overlap":"Fits multivariate community transitions, estimates abiotic and biotic driver effects, and derives eigenvalue-based community stability and perturbation-response metrics, including freshwater-plankton applications.","remaining_difference":"It does not preregister a bootstrap-stable risky transition eigenmode, identify randomized control-to-mode effects, or test a policy selected specifically to damp that mode against disturbance-matched rivals."},{"name":"Probabilistic multivariate vector-autoregressive early warnings","source_ids":["SRC4"],"same_problem":true,"same_causal_lever":false,"overlap":"Uses regularized multivariate autoregression to detect systemic loss of resilience in interacting ecological communities.","remaining_difference":"It is a simulated warning-indicator study, not a randomized intervention map or confirmatory modal-control experiment in lakes."},{"name":"Covariance-eigenvector early-warning analysis","source_ids":["SRC5"],"same_problem":true,"same_causal_lever":false,"overlap":"Extracts a dominant coupled direction and eigenvalue from multivariate ecological data to warn of critical transitions and localize vulnerability.","remaining_difference":"The direction is a covariance or principal-component direction rather than an estimated transition mode; the study assumes a codimension-one local bifurcation and supplies no causal control-to-mode mapping."},{"name":"Fish removal plus submerged-macrophyte transplantation for shallow-lake restoration","source_ids":["SRC6"],"same_problem":true,"same_causal_lever":false,"overlap":"Combines multiple trophic and habitat interventions to restore clear water and measures physicochemical and biological responses.","remaining_difference":"Treatment choice is based on restoration mechanisms and practice rather than randomized modal sensitivity, and it was not compared with coordinate-wise and nonlinear nonmodal policies under a common disturbance budget."},{"name":"PCLake model-assisted water-quality management","source_ids":["SRC8"],"same_problem":true,"same_causal_lever":false,"overlap":"Models coupled lake food-web, nutrient, hydrological, and physicochemical processes to estimate critical loads and assess management measures.","remaining_difference":"It is a process-based decision model, not a season-local empirically estimated transition-mode controller with spectral-gap, residual, drift, and causal-intervention gates."}],"contrastive_claim_remaining":"Within the preregistered season-local mesocosm window and only after all model gates pass, a policy chosen from randomized control-to-state effects to oppose a bootstrap-stable leading transition mode will reduce risky-mode-amplitude AUC by at least 25% and ecological loss by at least 0.5 SD relative to both an intensity-matched coordinate-wise policy and a regularized nonlinear nonmodal controller.","contrastive_claim_falsifier":"Falsify the problem component if independent-variable or exogenous-forcing models are within 0.05 normalized RMSE and no bootstrap-stable coupled direction precedes loss. Falsify the intervention component if any preregistered gate fails or if simultaneous 95% confidence intervals do not establish both specified improvements against both active rivals.","confidence":"MODERATE","search_limitations":"This bounded eight-source search covered all six lanes and opened every retained source, but it was not a systematic review and could miss unpublished work, patents, proprietary restoration systems, non-indexed regional practice, differently named Koopman or dynamic-mode control studies, and post-search publications. It cannot establish world novelty, patentability, freedom to operate, prevalence, market size, or realized impact."},"researchability_gates":{"externally_supported_problem":{"status":"PASS","rationale":"Whole-ecosystem manipulation and multi-lake evidence establish coupled aquatic transitions and material shortcomings of separate or generic warning indicators, while leaving the exact local-mode hypothesis open to falsification.","source_ids":["SRC1","SRC2","SRC5"]},"identifiable_adopter_or_authorizer":{"status":"PASS","rationale":"Mesocosm investigators, facility and permit authorities, animal or ecological review bodies, and water managers are identifiable institutional roles; official guidance and first-party modeling practice show relevant users exist even though no specific adopter has committed.","source_ids":["SRC7","SRC8"]},"distinct_testable_incremental_claim":{"status":"PASS","rationale":"Adjacent work covers multivariate warning directions, MAR stability, process-based lake models, and combined restoration, but not the randomized, gated control-to-transition-mode policy tested against both coordinate-wise and nonlinear nonmodal rivals.","source_ids":["SRC3","SRC4","SRC5","SRC6","SRC8"]},"bounded_next_evidence_step":{"status":"PASS","rationale":"A preregistered enclosed-mesocosm identification stage followed conditionally by a powered four-arm randomized comparison is bounded by specified variables, doses, model gates, endpoints, sample-size rules, and stopping criteria; OECD guidance recognizes hypothesis-specific freshwater mesocosm studies as an appropriate evidence bridge.","source_ids":["SRC7"]},"no_unresolved_safety_or_authority_stop":{"status":"PASS","rationale":"No intrinsic stop was found for enclosed work using site-native organisms and nonlethal, in-range manipulations. Proceeding remains conditional on named facility, permit, animal-care, ecological, biosafety, and capacity approvals; field release, wild removal, non-native or pathogenic additions, and unpermitted actions remain excluded.","source_ids":["SRC6","SRC7"]},"adequate_search_evidence":{"status":"PASS","rationale":"The bounded search covered direct evidence, closest methodological and intervention analogues, older MAR/VAR and biomanipulation terminology, official mesocosm guidance, operational lake-model products, Dutch and Chinese regional terminology, and combinations of the component subproblems using eight opened sources from seven publisher or institutional families.","source_ids":["SRC1","SRC2","SRC3","SRC4","SRC5","SRC6","SRC7","SRC8"]}},"strict_success":true,"screen_survival":true,"remaining_research_value":"MODERATE","recommended_next_step":"Preregister and run Stage A only at a facility that confirms approvals, containment, native-stock sourcing, welfare procedures, and capacity. Lock preprocessing, operator and rival models, randomized control-effect estimation, residual and spectral gates, and simulations before data collection. Advance to the independently randomized Stage B only if every stated model, safety, and power gate passes; otherwise report the modal hypothesis as unsupported without substituting modes or endpoints.","world_novelty_boundary":"The search supports only a contrastive, study-bounded claim: no retained source combined randomized control-to-mode identification with gated season-local ecological transition modes and a disturbance-matched confirmatory comparison against coordinate-wise and nonlinear nonmodal policies. This is not evidence of world novelty, patentability, freedom to operate, routine-practice prevalence, market size, or realized ecological impact."}