{"schema_version":1,"assessment_id":"eoa_inverse_innovation_exp03_opportunity320_20260801","source_experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"negative_space_design__astronomy_astrophysics","archetype_slug":"negative_space_design","domain_slug":"astronomy_astrophysics","title":"Protected line-free spectral windows for baseline estimation","opportunity_summary":"Predeclare and independently validate bracketing reference windows, preserve all raw channels, and constrain spectral baselines from those windows. A retrospective injection test would compare measurement bias and held-out residuals against ordinary post-inspection baseline fitting and a flexible full-band joint model. The candidate is coherent and testable, but prevalence, superiority, and prior-art distinctiveness are unestablished.","adopter_authorizer":"The observing-program principal investigator, with approval or participation from the designated calibration or instrument scientist and subject to observatory configuration and data-quality rules.","scores":{"meaningful_impact":{"score":4,"rationale":"The proposal targets biased line detections, fluxes, widths, and profiles, which are directly consequential scientific measurements. The magnitude and prevalence of this failure mode are unsupported in the sealed record."},"stakeholder_pull":{"score":3,"rationale":"Observers, analysts, operators, and downstream archive users have identifiable interests in stable measurements, but the candidate provides no evidence of expressed demand, recurring incidents, or willingness to trade bandwidth for protected windows."},"incremental_advantage":{"score":3,"rationale":"Predeclared protected windows, held-out residuals, injection recovery, and mask provenance provide a plausible advantage over adjustable post-inspection fitting. No sealed evidence shows improvement over either the ordinary baseline or the flexible joint-model rival."},"distinctiveness_plausibility":{"score":2,"rationale":"The complete bundle is specifically differentiated inside the proposal, but prior-art status is explicitly unsearched and protected line-free channels may overlap established observing or reduction practice."},"technical_implementability":{"score":4,"rationale":"The first test uses an existing dataset, reversible masks, synthetic-line injections, and standard comparative outputs. Implementation becomes less favorable where clean intervals do not exist or the baseline is nonlocal or discontinuous."},"adoption_authority_feasibility":{"score":4,"rationale":"The principal investigator and calibration or instrument scientist are explicitly identified as able to define pilot windows within observatory rules. Broader operational adoption could require additional observatory scheduling and configuration authority."},"evidence_readiness":{"score":4,"rationale":"The candidate supplies observable instability criteria, two comparators, injection recovery, held-out residuals, problem and intervention falsifiers, and explicit halt conditions. Representative datasets and acceptance thresholds remain unspecified."},"safety_net_benefit":{"score":4,"rationale":"Raw channels are preserved, exclusions are reversible, provenance is retained, and the pilot halts for structured residuals, broad-line occupancy, excessive data loss, or comparator underperformance. Misclassifying weak or unexpected emission remains a material risk."},"scalability":{"score":3,"rationale":"The workflow could be repeated across channelized spectra, but window validity depends on instrument, redshift, source class, RFI environment, and line regime, requiring repeated local validation rather than a universal window standard."}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"10K_TO_50K","scope":"Retrospective analysis of one existing spectral dataset using two preregistered window schemes, a bounded synthetic-line grid, two comparators, held-out residual analysis, and a written decision report.","confidence":"MODERATE","assumptions":["The dataset and required calibration products are already accessible.","Existing analysis software can implement masks, injections, and comparator fits with limited modification.","No new observing time, hardware, or formal compliance review is required."]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Prepare one instrument or observing mode for controlled use through window-selection criteria, contamination checks, provenance capture, analyst tooling, validation cases, documentation, and operator training.","confidence":"LOW","assumptions":["Deployment is limited to one instrument or observing mode.","No major spectrometer, archive, or scheduling-system redesign is needed.","Instrument-scientist and analyst coordination constitutes most of the resource burden."]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Launch the protocol across a bounded observatory program portfolio, including configuration review, multiple-regime validation, pipeline and archive integration, quality controls, training, and evaluation.","confidence":"LOW","assumptions":["Launch does not require new hardware.","Only limited additional bandwidth or observing time is consumed.","Existing observatory governance and data infrastructure can incorporate masks and provenance."]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Maintain contamination maps and validation tests, review exceptions, support analysts, monitor recovery and residual metrics, and preserve window and mask provenance for one bounded deployment.","confidence":"LOW","assumptions":["Protected-window checks are largely integrated into normal observing and reduction operations.","No substantial recurring telescope-time allocation is added.","Instrument changes and new line regimes require periodic rather than continuous revalidation."]}},"research_burden":"HIGH","earliest_credible_horizon":"0_TO_3_MONTHS","pipeline_gates":{"recognizable_externally_supportable_problem":{"status":"YES","reason":"The candidate defines an observable problem state: no independently checked bracketing reference windows and material variation in baseline estimates under plausible masks or polynomial orders. External work is still needed to establish prevalence."},"identifiable_adopter_or_authorizer":{"status":"YES","reason":"The observing-program principal investigator and designated calibration or instrument scientist are explicitly identified as authorities for the bounded pilot, subject to observatory rules."},"distinct_testable_incremental_claim":{"status":"YES","reason":"The protected, predeclared, independently diagnosed windows are explicitly compared with ordinary adjustable baseline fitting and a flexible full-band joint model using recovery bias and held-out residuals. Distinctiveness from world prior art remains unresolved."},"bounded_next_evidence_step":{"status":"YES","reason":"The sealed candidate authorizes a reversible test on one existing dataset with two window schemes, a bounded injection grid, preserved raw channels, explicit comparators, and halt conditions."},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The pilot preserves raw data and provenance, excludes irreversible deletion and concealed failures, identifies authorized decision-makers, and specifies rollback triggers. It does not authorize live operational deployment."},"implementation_cost_scope_and_range":{"status":"UNCERTAIN","reason":"The single-dataset test is bounded, but the candidate does not specify dataset size, software changes, staffing, acceptance thresholds, added bandwidth, telescope-time opportunity cost, or the scale of any operational rollout."}},"blocking_evidence":["No prior-art review establishes whether preregistered, protected, empirically validated reference windows differ materially from established observing and reduction practice.","No representative evidence establishes how often baseline instability is attributable to inadequate reference channels rather than other error sources.","No comparative result shows improved injected-line recovery or held-out residuals against both specified comparators.","No evidence quantifies the loss of velocity coverage, sensitivity, or unexpected-feature discovery caused by reserving windows.","No validated rule establishes when a candidate window is sufficiently clean and locally informative for the target baseline."],"next_evidence_step":"On one existing spectrum exhibiting the stated mask or polynomial-order instability, preregister two bracketing protected-window schemes and a bounded synthetic-line grid; preserve every raw channel and original mask; compare line-parameter bias, total recovery error, and held-out residual structure against both ordinary adjustable baseline fitting and the full-band joint-model rival. Reject the intervention if neither scheme improves recovery or residuals, or if either causes excessive coverage loss, structured residuals, or plausible broad-line exclusion.","research_questions":["Is the complete protected-window and provenance bundle already standard practice in relevant instruments or pipelines?","Across representative spectra, how often does inadequate reference space materially affect line estimates relative to other uncertainties?","What contamination and residual tests reliably distinguish usable reference windows from weak emission, RFI, missing data, or calibration failure?","Under which baseline structures and line regimes do protected windows outperform both ordinary fitting and flexible joint modeling?","How much bandwidth, sensitivity, velocity coverage, and observing-time opportunity cost does the intervention impose?","Can decision thresholds and window rules transfer across instruments, redshifts, source classes, and line regimes without suppressing unexpected discoveries?"],"recommendation":"PRIOR_ART_RESEARCH","uncertainty_constraints":["Closed-book assessment provides no evidence of novelty, prevalence, market size, realized impact, or empirical superiority.","Cost bands are resource-equivalent planning ranges based only on the described scope and may rise materially if additional observing time, bandwidth, hardware, or observatory integration is required.","The first-evidence horizon assumes immediate access to a suitable calibrated dataset and existing analysis capability.","Scores concern the candidate as sealed and do not generalize beyond spectral regimes where reference-window cleanliness can be tested."],"closed_book_prior_art_boundary":"Prior-art status is explicitly UNSEARCHED. The sealed record supports a coherent, reversible, falsifiable comparison but cannot establish that protected or preregistered line-free windows, injection testing, or provenance-bearing masks are novel, uncommon, or incrementally distinct from established astronomy practice."}