{"schema_version":1,"research_id":"eoa_inverse_innovation_exp05_external_evaluation_20260803","source_assessment_id":"invariant_mode_decomposition_design__biology_ecology:P3:v0","cell_id":"invariant_mode_decomposition_design__biology_ecology","search_queries":["site:usgs.gov amphibian disease surveillance chytrid monitoring program ponds","site:woah.org amphibian chytridiomycosis surveillance sampling manual","amphibian pathogen surveillance environmental DNA ponds life stages adaptive sampling study","dynamic mode decomposition disease surveillance ecology pathogen transmission modes","optimal sensor placement dynamic mode decomposition biological systems surveillance observability modes paper","adaptive sampling wildlife disease surveillance risk based sampling official guidance","amphibian disease surveillance sampling protocol Bd USGS swab qPCR animal handling official","Batrachochytrium salamandrivorans strategic plan surveillance USFWS need monitoring","2026 wildlife disease qPCR testing fee amphibian chytrid Bsal Bd laboratory price per sample","amphibian chytrid qPCR test fee laboratory 2025 Bd Bsal","field biology technician salary 2026 USGS cost hourly official","amphibian disease surveillance project budget qPCR ponds sampling","\"Discovering dynamic patterns from infectious disease data\" PLOS One","\"Dynamic sensor selection for biomarker discovery\" publisher","\"Beyond the swab\" ecosystem sampling amphibian pathogen Springer","\"Design- and model-based recommendations\" Ecology and Evolution full text","site:link.springer.com/article/10.1007/s00442-018-4167-6","site:onlinelibrary.wiley.com/doi/10.1002/ece3.3616","Scheele amphibian chytrid decline 501 species Science 2019 official abstract","site:science.org/doi/10.1126/science.aav0379"],"sources":[{"source_id":"S1","title":"Amphibian fungal panzootic causes catastrophic and ongoing loss of biodiversity","publisher":"Science / American Association for the Advancement of Science","url":"https://pubmed.ncbi.nlm.nih.gov/30923224/","source_class":"PRIMARY_RESEARCH","publication_date":"2019-03-29","accessed_at":"2026-08-03","claims_supported":["Chytridiomycosis visibly matters at biodiversity scale: the study associated it with declines in at least 501 amphibian species and 90 presumed extinctions.","Only a minority of affected species showed recovery, supporting continued surveillance relevance.","The estimate concerns global disease impact, not the prevalence of the candidate's proposed hidden-mode surveillance failure."]},{"source_id":"S2","title":"Batrachochytrium salamandrivorans (Bsal) Surveillance","publisher":"U.S. Geological Survey, National Wildlife Health Center","url":"https://www.usgs.gov/centers/nwhc/science/batrachochytrium-salamandrivorans-bsal-surveillance","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"2026-01-22","accessed_at":"2026-08-03","claims_supported":["USGS NWHC is an identifiable surveillance operator working with ARMI and state, federal, and Tribal partners.","The program actively samples high-risk locations and has tested more than 15,000 samples from 39 states and more than 55 species since 2016.","USGS describes Bsal as capable of mass mortality, severe decline, and localized extinction, and continues to provide surveillance services.","Existing targeting is risk-assessment based, establishing both stakeholder activity and a comparator to modal allocation."]},{"source_id":"S3","title":"General guidelines for surveillance of diseases, pathogens and toxic agents in free-ranging wildlife","publisher":"World Organisation for Animal Health","url":"https://www.woah.org/app/uploads/2024/09/2024-final-guidelines-disease-pathogen-toxin-surv-wildlife-v27.06.pdf","source_class":"OFFICIAL_GUIDANCE","publication_date":"2024-09-01","accessed_at":"2026-08-03","claims_supported":["Wildlife surveillance should have explicit objectives, identified authorities and stakeholders, defined uses and limitations, and feasible field and laboratory workflows.","Programs must address resources, cold chain, biosafety, biological-risk management, legal permissions, animal welfare, land access, data ownership, and stakeholder or rights-holder participation.","WOAH recognizes targeted, risk-based, sentinel, environmental, and mixed surveillance strategies, making adaptive allocation institutionally intelligible but not proving this candidate's advantage.","The guidance supplies workflow and safety requirements that remain mandatory around any modal optimization."]},{"source_id":"S4","title":"Aquatic Animal Health Code, Chapter 1.4: Aquatic animal disease surveillance","publisher":"World Organisation for Animal Health","url":"https://www.oie.int/fileadmin/Home/eng/Health_standards/aahc/current/chapitre_aqua_ani_surveillance.pdf","source_class":"STANDARD","publication_date":"2024-07-02","accessed_at":"2026-08-03","claims_supported":["Risk-based sampling is established practice for concentrating effort where infection is most likely and can improve efficiency relative to random sampling for specified surveillance objectives.","Risk factors and their assumptions must be documented and can include water pathways, habitat, environmental conditions, species, and susceptible life stages.","Diagnostic sensitivity and specificity must be incorporated into surveillance interpretation, and positives may require confirmatory testing.","This standard is a close workflow comparator but targets infection risk rather than observability of estimated transition modes."]},{"source_id":"S5","title":"Design- and model-based recommendations for detecting and quantifying an amphibian pathogen in environmental samples","publisher":"Ecology and Evolution / John Wiley & Sons","url":"https://sites.warnercnr.colostate.edu/llbailey/wp-content/uploads/sites/113/2018/01/Mosher-et-al.-2017-EcolEvol.pdf","source_class":"PRIMARY_RESEARCH","publication_date":"2017-11-12","accessed_at":"2026-08-03","claims_supported":["Environmental sampling of free-living Bd can add inference beyond host sampling.","Natural-water qPCR inhibition biased detection and quantification, while repeated spatial or temporal samples mitigated some occurrence bias but not persistent quantification bias.","Pathogen concentration affected detection probability, supporting explicit observation and missingness models rather than treating assay values as direct pathogen-state coordinates.","The study supports dense-reference and assay-quality controls but does not validate transition-matrix estimation or modal allocation."]},{"source_id":"S6","title":"Discovering dynamic patterns from infectious disease data using dynamic mode decomposition","publisher":"International Health / Oxford University Press","url":"https://doi.org/10.1093/inthealth/ihv009","source_class":"PRIMARY_RESEARCH","publication_date":"2015-03-02","accessed_at":"2026-08-03","claims_supported":["Dynamic mode decomposition has already been applied to infectious-disease surveillance data to find coherent spatiotemporal modes and interpret their eigenvalues and location loadings.","The authors explicitly discuss surveillance-team and resource allocation, sparse measurements, minimizing redundant surveillance sites, and dynamically linked locations.","The work demonstrates the broad modal-surveillance concept in human infectious disease, creating substantial prior-art overlap.","Its examples do not allocate amphibian host-stage, environmental-reservoir, and pond assays under welfare, minimum-coverage, and diagnostic-quality constraints."]},{"source_id":"S7","title":"Dynamic Sensor Selection for Biomarker Discovery","publisher":"Proceedings of the National Academy of Sciences","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC11142321/","source_class":"PRIMARY_RESEARCH","publication_date":"2025-10-07","accessed_at":"2026-08-03","claims_supported":["Biological sensor selection using learned dynamics, DMD, observability measures, budgets, experimental constraints, and time-varying sensor reallocation is existing research practice.","The framework recommends learning dynamics from rich data before selecting a reduced sensor set and notes poor conditioning, changing regimes, local state dependence, and diminishing returns.","This is a close analogue to dense identification followed by observability-guided assay allocation.","The demonstrated domains are biomarkers, transcriptomics, and neural measurements rather than wildlife-pathogen field surveillance."]},{"source_id":"S8","title":"Diagnostic Services","publisher":"University of Tennessee Institute of Agriculture Amphibian Disease Laboratory","url":"https://amphibiandisease.tennessee.edu/diagnostic-services/","source_class":"OFFICIAL_ORGANIZATION_DATA","publication_date":"2026-01-01","accessed_at":"2026-08-03","claims_supported":["A first-party amphibian diagnostic laboratory lists extraction plus one-pathogen qPCR at $35 per sample, additional pathogens at $10, and Bd/Bsal/ranavirus testing at $45 per sample under a listed program.","The laboratory lists data-processing or interpretation consultation at $100 per hour.","These unit prices anchor assay and analytical cost assumptions but exclude field labor, travel, permitting, cold chain, duplicates, consumables, and program overhead."]}],"problem_evidence":{"support":"MODERATE","rationale":"Amphibian chytrid disease demonstrably matters, and active surveillance must cope with spatial, environmental, life-stage, detection-probability, and assay-quality heterogeneity. DMD research also shows that coupled disease patterns can differ from surface coordinates. However, no opened source demonstrates the proposal's specific operational failure: that a current amphibian program using positives, abundance, or accessibility has missed a reproducible growing host-stage–reservoir–pond mode. That premise remains a plausible, testable hypothesis rather than a verified prevalence claim.","source_ids":["S1","S2","S5","S6"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"USGS NWHC, ARMI, and their state, federal, Tribal, laboratory, and educational partners are identifiable operators already conducting targeted amphibian-pathogen surveillance. WOAH directly addresses wildlife authorities and expresses needs for efficient, objective-linked, risk-aware surveillance with documented limitations. Neither source expresses demand for eigenmode-based allocation specifically, so adopter existence is strong but pull for this implementation is inferred.","source_ids":["S2","S3","S4"]},"prior_art":{"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Dynamic mode decomposition for infectious-disease surveillance and resource allocation","similarity":"Already decomposes spatiotemporal disease measurements into dynamic modes, maps mode entries to locations, interprets growth or decay, and discusses surveillance-site and resource allocation.","remaining_difference":"It does not demonstrate amphibian pond, life-stage, and environmental-reservoir assay allocation with minimum coverage, welfare caps, assay-quality gates, dense-reference validation, and withdrawal rules.","source_ids":["S6"]},{"name":"Dynamic Sensor Selection for biological monitoring","similarity":"Learns biological dynamics with DMD, optimizes sensor sets using observability under budgets and experimental constraints, reallocates sensors over time, and addresses changing regimes and conditioning.","remaining_difference":"Its sensors are biomarkers in molecular and neural systems, not pathogen assays distributed across ponds, amphibian stages, and environmental reservoirs; field biosecurity and wildlife authority are absent.","source_ids":["S7"]},{"name":"WOAH risk-based aquatic and wildlife surveillance","similarity":"Established practice already directs constrained surveillance toward higher-risk populations, places, pathways, species, and life stages and requires documented assumptions and diagnostic performance.","remaining_difference":"It prioritizes infection risk rather than the uncertainty of a reproducible transition-mode coordinate and does not prescribe eigendecomposition, spectral-gap gates, or modal reconstruction.","source_ids":["S3","S4"]},{"name":"USGS NWHC targeted Bsal surveillance","similarity":"A live amphibian-pathogen program uses geographic risk assessments to select high-risk locations and coordinates field and laboratory partners at national scale.","remaining_difference":"The public description does not estimate a host-stage–reservoir–pond transition operator or allocate discretionary samples by modal observability.","source_ids":["S2"]}],"distinctive_claim_remaining":"At equal assay and handling cost and with identical minimum pond coverage, a preregistered schedule chosen to reduce uncertainty in a reproducible, adequately conditioned host-stage–reservoir–pond transition mode will estimate that held-out modal coordinate and its direction of change more accurately than fixed stratification, recent-positive allocation, habitat/connectivity risk targeting, and a direct spatiotemporal predictor, without increasing missed reference signals, welfare events, or loss of protected-mode observability.","confidence":"HIGH"},"implementation_evidence":{"support":"MODERATE","rationale":"qPCR environmental and swab workflows, targeted wildlife surveillance, DMD, biological observability analysis, and diagnostic services all exist. The main technical weakness is sample complexity: a pond-by-compartment state can have dozens of coordinates, while one breeding season may provide few independent transition pairs. A complete unconstrained transition matrix and eigenbasis may therefore be unidentifiable or unstable without pooling assumptions, regularization, longer time series, or a lower-dimensional state. Detection heterogeneity, inhibition, censoring, non-normality, near-degenerate modes, seasonal drift, and informative missingness further separate assay measurements from pathogen state. Authority is feasible only with land access, wildlife permits, animal-welfare review, laboratory acceptance, biosafety, and data-governance approvals. The observational design and rollback rules bound safety, but those approvals and a site-specific biosecurity plan are not yet evidenced.","source_ids":["S2","S3","S4","S5","S6","S7","S8"]},"scores":{"meaningful_impact":{"score":4,"rationale":"Chytrid pathogens have caused severe documented biodiversity loss, and a surveillance improvement could matter where it changes reliable detection or characterization.","source_ids":["S1","S2"]},"stakeholder_pull":{"score":3,"rationale":"An active USGS surveillance network and formal wildlife-surveillance authorities exist, but no identified operator has requested modal allocation or committed data, staff, or funds.","source_ids":["S2","S3"]},"incremental_advantage":{"score":2,"rationale":"The proposed benefit is unmeasured, and close prior art already combines disease modes, observability, sparse sensors, and resource allocation. Advantage over established risk-based sampling and direct predictors requires a field comparison.","source_ids":["S4","S6","S7"]},"distinctiveness_plausibility":{"score":2,"rationale":"The amphibian-specific safeguards and comparators are distinctive as a protocol package, but the mathematical core substantially overlaps infectious-disease DMD and dynamic biological sensor-selection research.","source_ids":["S6","S7"]},"technical_implementability":{"score":2,"rationale":"All component methods are available, but a high-dimensional complete transition matrix is likely underidentified from a single seasonal time series, and assay values have imperfect, heterogeneous observation processes. A reduced or regularized model may be required before the stated complete-eigenbasis workflow is credible.","source_ids":["S5","S6","S7"]},"adoption_authority_feasibility":{"score":3,"rationale":"USGS NWHC and partner wildlife authorities are credible adopters, and laboratories can run the assays. Site access, jurisdiction-specific wildlife permits, welfare authorization, and laboratory agreements remain unverified.","source_ids":["S2","S3","S8"]},"evidence_readiness":{"score":2,"rationale":"The claim and comparators are preregistrable, but no candidate-specific dataset, power analysis, operator-identifiability result, or prospective validation exists. The decisive evidence requires proprietary or newly collected field and laboratory data.","source_ids":["S5","S7"]},"safety_net_benefit":{"score":4,"rationale":"Minimum geographic coverage, environmental-sampling fallback, welfare caps, diagnostic quality controls, suspension gates, and reversion to fixed stratification provide meaningful downside protection if enforced.","source_ids":["S3","S4","S5"]},"scalability":{"score":3,"rationale":"Computation and rule reuse are inexpensive relative to sampling, but scaling increases assay, travel, permitting, coordination, data-harmonization, and model-dimension burdens. Existing national partnerships show organizational reach, not scalability of the modal method.","source_ids":["S2","S3","S8"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"50K_TO_250K","scope":"One small preregistered breeding-season identification/validation study over roughly 6–10 nearby ponds, with repeated minimum-coverage sampling, blinded duplicates, capped swabs, environmental samples, equal-budget comparator schedules, statistical analysis, and a final report.","confidence":"MODERATE","assumptions":["Approximately 500–1,200 laboratory samples at roughly $35–$45 per pathogen-testing sample before shipping, repeats, controls, and pooling effects.","Two seasonal field staff, a project biologist, limited local travel, consumables, cold chain, permits, welfare review, laboratory coordination, and 150–300 analytical hours.","Existing vehicles, storage, qPCR capacity, basic GIS, and open-source numerical software are available.","The pilot remains geographically compact and does not purchase major laboratory equipment."],"source_ids":["S3","S8"]},"initial_deployment_startup":{"band_2026_usd":"250K_TO_1M","scope":"Prepare a regional authority for operational use: finalize SOPs and biosecurity, obtain permits and agreements, configure data and laboratory pipelines, train crews, collect an overinstrumented identification baseline, implement audit-ready software, and independently review the frozen allocation rule.","confidence":"LOW","assumptions":["A 15–30 pond network and multiple host or environmental compartments are included.","Existing authority and laboratory infrastructure is reused; no new diagnostic laboratory is built.","Costs include one dense identification season, software validation, staff training, stakeholder engagement, and contingency for repeat assays.","Actual cost is highly sensitive to travel distances, protected-species requirements, and the state-vector dimension."],"source_ids":["S2","S3","S7","S8"]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Run the first full regional season after startup, including repeated field rounds, minimum coverage, mode-directed discretionary samples, laboratory controls, reference subsampling, model refreshes, welfare and biosecurity monitoring, and decision reports.","confidence":"LOW","assumptions":["Approximately 1,500–5,000 samples including duplicates and controls.","Two or three field crews operate within one region during one breeding season.","The launch evaluates surveillance only and excludes treatment, closures, animal movement, or habitat intervention.","Laboratory unit pricing is indicative and program agreements may differ."],"source_ids":["S2","S3","S8"]},"annual_recurring":{"band_2026_usd":"250K_TO_1M","scope":"Maintain one regional program annually: seasonal field crews, assays, shipping and cold chain, quality assurance, data management, mode and residual monitoring, retraining, permits, stakeholder coordination, and annual reporting.","confidence":"LOW","assumptions":["The pond network and sampling intensity remain similar to initial launch.","Major equipment and software have already been acquired.","Annual dense-reference sampling is reduced but not eliminated.","Expansion to multiple regions or pathogens would multiply field and laboratory costs rather than remain inside this band."],"source_ids":["S2","S3","S8"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"YES","reason":"The disease burden, active surveillance need, environmental-pathogen sampling problem, and heterogeneous assay detection are externally supported, although the exact modal-miss failure is not yet observed.","source_ids":["S1","S2","S5"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"USGS NWHC, ARMI, and cooperating state, federal, and Tribal wildlife authorities are identifiable surveillance actors; laboratories and welfare or permit authorities would share authorization.","source_ids":["S2","S3","S8"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The equal-budget claim specifies fixed stratification, recent-positive allocation, risk targeting, and direct prediction as comparators and specifies accuracy, missed-signal, welfare, coverage, and protected-mode outcomes.","source_ids":["S4","S6","S7"]},"bounded_next_evidence_step":{"status":"YES","reason":"A geographically compact, one-season identification/held-out-validation pilot can freeze the operator and allocation rule before opening validation assays and can terminate on explicit falsifiers.","source_ids":["S3","S5","S7"]},"no_unresolved_safety_or_authority_stop":{"status":"UNCERTAIN","reason":"The proposal has strong exclusions and rollback rules, but site access, wildlife collection permits, animal-welfare approval, biosafety procedures, sample transport authority, rights-holder participation, and laboratory acceptance must be verified for the selected jurisdiction before fieldwork.","source_ids":["S3"]},"credible_cost_scope_and_range":{"status":"YES","reason":"Assay pricing supplies a direct unit anchor, and the ranges explicitly scope field, laboratory, analytical, regulatory, and coordination work. Overall confidence is only low to moderate because no site-specific staffing, travel, or permit budget is available.","source_ids":["S3","S8"]}},"next_evidence_step":"Run a two-stage partnered feasibility study. First, before animal handling, use one existing dense longitudinal amphibian-pathogen dataset or an environmental-only pre-season series to freeze a reduced state definition and demonstrate that the regularized transition operator is identifiable under blocked resampling: require adequate effective sample size, bounded eigenvector condition number, stable mode matching, and out-of-time residuals. Halt if no reproducible action-relevant mode exists. If that gate passes, conduct one breeding-season pilot across 6–10 nearby ponds with minimum coverage at every interval, blinded duplicate environmental samples, capped approved swabs, and a dense reference retained from allocation decisions. Freeze all rules before validation. Compare equal-budget mode-aligned allocation with fixed stratification, recent-positive allocation, documented habitat/connectivity risk sampling, and a direct regularized spatiotemporal predictor. Primary outcomes are held-out error and uncertainty for the preregistered modal coordinate and its direction of change; secondary outcomes are whole-state reconstruction, missed reference signals, assay failures, welfare events, coverage loss, cross-mode observability, conditioning, spectral separation, and drift. Falsify the intervention if the mode is unstable under resampling, the operator is underidentified, the mode-aligned schedule fails to outperform at least fixed stratification on the primary outcome, any comparator has better reference concordance, a protected signal is missed, or welfare, biosecurity, or minimum-coverage limits are breached.","blocking_evidence":["No candidate-specific evidence shows that current count-led or accessibility-led amphibian surveillance has missed a coupled growing mode.","No longitudinal dataset or power analysis establishes that the proposed pond-by-stage-by-reservoir transition matrix is identifiable within one breeding season.","No evidence yet shows stable eigenvectors, adequate spectral separation, acceptable conditioning, or reproducible mode identity under resampling and temporal holdout.","No prospective comparison establishes incremental performance over fixed stratification, recent-positive allocation, risk-based targeting, or a direct spatiotemporal predictor.","No named wildlife authority has committed to adopt, fund, or authorize the modal protocol.","Jurisdiction-specific land access, wildlife permits, welfare review, sample-transport authority, biosecurity plan, laboratory agreement, and data governance are unresolved.","World novelty, patentability, freedom to operate, market size, and realized impact were not measured."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"World novelty was not assessed. The bounded search found substantial conceptual prior art in infectious-disease DMD, observability-guided dynamic biological sensor selection, and risk-based wildlife surveillance. The remaining amphibian-specific protocol claim may still be useful and testable, but no inference is made about world novelty, patentability, freedom to operate, market size, or realized impact.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":true,"progress_targets":["Obtain an authority and laboratory partner with written willingness to supply data and conditionally authorize the observational pilot.","Predefine a lower-dimensional or regularized state model and demonstrate transition-operator identifiability using longitudinal data, simulation-based calibration, and blocked resampling.","Show a reproducible action-relevant mode with acceptable conditioning, spectral separation, temporal persistence, and held-out residual structure.","Complete the equal-budget prospective comparison against fixed stratification, recent-positive allocation, habitat/connectivity risk targeting, and a direct spatiotemporal predictor.","Verify wildlife permits, landholder or rights-holder participation, animal-welfare approval, biosecurity, sample transport, laboratory acceptance, and data governance.","Produce a site-specific budget and power calculation tied to pond count, intervals, host-handling caps, assay duplicates, missingness, and expected detection probability."],"reason":"Web evidence verifies a consequential disease, credible surveillance institutions, implementable component methods, and substantial adjacent prior art. It cannot establish whether a reproducible amphibian host-stage–reservoir–pond mode exists, whether the proposed high-dimensional operator is identifiable, or whether mode-aligned allocation outperforms equal-budget alternatives safely. Those questions require proprietary longitudinal data and live field/laboratory validation; under the controller rule this is an empirical-research stop and therefore repairable is false."},"proposal_index":3}