Skip to content

GeoGeekLab/nature-reviewer-skills / nature-earth-system-reviewer-skills/skills/nature-atmospheric-science-reviewer-skill

nature-atmospheric-science-reviewer

Verified

Nature-style atmospheric science manuscript reviewer skill for testing observation, reanalysis, model, aerosol-cloud-radiation, extreme-event, circulation-mechanism and attribution claims against their physical evidence chain.

reviewer_db/patterns.csv

View on GitHub
pattern_id,gate,abstract_risk_pattern,revision_direction,severity_default,corpus_unit_count_for_gate,source_type,verbatim_reviewer_text
P001,atmospheric_claim_evidence_calibration,"Broad atmospheric claims are stronger than the demonstrated variable, domain, or diagnostic supports.","Require each headline claim to be mapped to the actual observable, model output, validation target, and inference boundary.",central,264,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P002,atmospheric_claim_evidence_calibration,"A paper treats a proxy, index, or diagnostic as the target atmospheric quantity without quantifying the proxy relationship.",State what is directly observed or simulated and validate the proxy-target link under relevant regimes.,central,264,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P003,atmospheric_claim_evidence_calibration,"The novelty claim is not calibrated against the closest atmospheric dynamics, chemistry, observation, or modelling literature.","Compare against the nearest prior work using the same scale, phenomenon, data family, and diagnostic metric.",major,264,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P004,atmospheric_claim_evidence_calibration,Policy-facing or public-impact statements exceed the physical evidence chain.,"Separate physical attribution, exposure/risk framing, and policy implication; soften claims not directly tested.",major,264,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P005,observational_product_reanalysis_validity,"Reanalysis, satellite, or station products are used as truth without product-specific bias and uncertainty checks.","Add product intercomparison, independent validation, known retrieval limitations, and regime-specific bias tests.",central,210,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P006,observational_product_reanalysis_validity,"The observational product is evaluated at aggregate scale while conclusions concern extremes, vertical structure, clouds, aerosols, or local circulation.",Validate the product at the scale and atmospheric state used in the claim.,central,210,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P007,observational_product_reanalysis_validity,"Retrieval validity under clouds, aerosols, precipitation, boundary-layer conditions, or complex terrain is assumed.","Document retrieval screening, representativeness, missingness, and sensitivity to quality flags.",major,210,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P008,observational_product_reanalysis_validity,A single observational product drives the result although alternative products disagree or have known structural uncertainty.,Use multi-product comparison and show whether the conclusion is robust to product choice.,major,210,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P009,sampling_event_representativeness,"The case, station network, campaign, season, or event sample is too narrow for the claimed atmospheric generalization.","Define the sampling frame and test whether conclusions hold across seasons, regions, regimes, and event types.",central,166,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P010,sampling_event_representativeness,Event selection or compositing may condition on the outcome variable.,Separate discovery and evaluation samples; predefine event thresholds or test alternative definitions.,major,166,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P011,sampling_event_representativeness,"Sparse vertical, temporal, or regional coverage is not propagated into inference uncertainty.",Quantify representativeness error and avoid spatial/temporal extrapolation not supported by coverage.,major,166,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P012,sampling_event_representativeness,The reference period or baseline changes the sign or magnitude of the claimed anomaly.,Test baseline sensitivity and report whether conclusions are reference-period dependent.,major,166,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P013,scale_resolution_boundary_consistency,A process resolved at one scale is interpreted at another scale without closure.,"Align claim scale, model grid, observational footprint, and process scale; discuss unresolved subgrid processes.",central,162,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P014,scale_resolution_boundary_consistency,"Boundary conditions, domain choice, or nesting strategy may impose the atmospheric response.",Add domain and boundary sensitivity tests or limit conclusions to the simulated domain.,major,162,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P015,scale_resolution_boundary_consistency,"Mesoscale, boundary-layer, convective, or orographic processes are inferred from coarse products.",Use scale-aware diagnostics or revise claims to match resolvable dynamics.,major,162,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P016,scale_resolution_boundary_consistency,Spatial aggregation hides regime-dependent or sign-changing effects.,"Report heterogeneity across regimes, seasons, altitudes, and dynamical settings.",major,162,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P017,model_physics_assumption_sensitivity,"A model-based conclusion depends on parameterizations, forcing, initialization, or structural choices not stress-tested.",Run or cite targeted sensitivity tests and identify parameters that control the claim.,central,204,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P018,model_physics_assumption_sensitivity,"The model reproduces headline metrics but not the relevant circulation, flux, cloud, aerosol, or chemistry process.","Evaluate process fidelity, not only outcome fit.",central,204,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P019,model_physics_assumption_sensitivity,Single-model evidence is used for attribution or mechanism claims.,"Use ensembles, alternative models, or structural uncertainty analysis; downgrade if unavailable.",major,204,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P020,model_physics_assumption_sensitivity,Tuning or calibration may use information later treated as independent validation.,"Separate calibration, selection, and validation; disclose tuning targets.",major,204,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P021,ai_weather_climate_model_fidelity,AI weather or climate models are judged mainly by headline skill metrics without physical-fidelity tests.,"Evaluate conservation, spectra, extremes, vertical coupling, regime transitions, and climate statistics.",central,222,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P022,ai_weather_climate_model_fidelity,"Training, validation, or benchmark design may allow temporal leakage, reanalysis memorization, or unfair baselines.","Audit data splits, baselines, lead times, initialization, and operational comparability.",central,222,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P023,ai_weather_climate_model_fidelity,"Short-range skill is extrapolated to climate, extremes, or out-of-distribution regimes.","Test OOD periods, climate-state shifts, rare events, and long-rollout stability.",major,222,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P024,ai_weather_climate_model_fidelity,"Computational advantage is claimed without equivalent accuracy, uncertainty, and deployment constraints.","Compare cost, resolution, latency, reliability, ensembles, and failure modes on matched tasks.",major,222,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P025,aerosol_cloud_radiation_microphysics,Aerosol-cloud or aerosol-radiation effects are inferred from aerosol amount alone.,"Separate amount, composition, vertical distribution, meteorology, cloud regime, and radiative kernel.",central,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P026,aerosol_cloud_radiation_microphysics,"Cloud adjustment, semi-direct effects, or meteorological covariability is not separated from microphysical forcing.","Use process decomposition, controls, or causal design to isolate mechanisms.",central,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P027,aerosol_cloud_radiation_microphysics,"New-particle formation, CCN, or cloud response is sensitive to precursor chemistry not independently constrained.","Add precursor, mechanism, and rate sensitivity; validate against field or chamber constraints where possible.",major,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P028,aerosol_cloud_radiation_microphysics,"Radiative forcing estimates omit uncertainty from optical properties, vertical overlap, cloud masking, or humidity dependence.",Propagate radiative and microphysical uncertainty into the headline forcing result.,major,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P029,chemistry_emissions_budget,Atmospheric chemistry or emissions conclusions depend on inventories or lifetimes that are not independently checked.,"Compare inventory families, source apportionment assumptions, chemical lifetimes, and observation constraints.",central,54,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P030,chemistry_emissions_budget,"Reactive species are aggregated although sign, lifetime, or forcing differs across compounds and pathways.",Disaggregate chemical pathways and report regime-dependent net effects.,major,54,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P031,chemistry_emissions_budget,"Budget closure is claimed without reconciling sources, sinks, transport, deposition, and chemical transformation.",Provide budget closure diagnostics and residual analysis.,major,54,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P032,chemistry_emissions_budget,Emission-reduction implications are inferred without considering co-emitted species or non-atmospheric parts of the life cycle.,Define the atmospheric boundary and state excluded life-cycle terms.,minor_major,54,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P033,extreme_event_attribution_counterfactual,"Extreme-event attribution depends on event definition, spatial extent, duration, or threshold choices.",Show sensitivity to event definitions and justify the physical relevance of thresholds.,central,45,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P034,extreme_event_attribution_counterfactual,"Counterfactual forcing, model evaluation, or internal variability is insufficient for causal attribution.",Evaluate factual/counterfactual model fidelity and quantify internal variability.,central,45,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P035,extreme_event_attribution_counterfactual,Return periods or risk ratios are extrapolated beyond data/model support.,"Report uncertainty, tail sensitivity, and whether extrapolation dominates the conclusion.",major,45,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P036,extreme_event_attribution_counterfactual,"Attribution to emitters, sectors, or regions is presented more precisely than the causal chain supports.",Separate climate-forcing contribution from event-specific impact and allocation uncertainty.,major,45,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P037,trend_internal_variability_signal_detection,"Trend claims are made from short, selected, or autocorrelated time series.","Account for autocorrelation, internal variability, reference period, and multiple testing.",central,69,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P038,trend_internal_variability_signal_detection,Forced signal and natural variability are not separated.,"Use control simulations, ensembles, pacemaker diagnostics, or formal detection-attribution framing.",central,69,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P039,trend_internal_variability_signal_detection,A regional trend is interpreted as global mechanism or forced response without teleconnection validation.,Validate regional-to-global scaling and circulation pathways.,major,69,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P040,trend_internal_variability_signal_detection,"The sign or significance of trend changes under alternative products, seasons, or endpoints.",Report endpoint sensitivity and avoid categorical claims where robustness is weak.,major,69,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P041,mechanism_causality_circulation_dynamics,"Mechanistic explanation is inferred from correlation, composites, or post-hoc diagnostics.","Use process diagnostics, perturbation tests, lag structure, or causal exclusion of alternatives.",central,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P042,mechanism_causality_circulation_dynamics,"Circulation pathway, moisture transport, energy budget, or feedback loop is not closed.",Add budget/pathway diagnostics and quantify competing mechanisms.,central,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P043,mechanism_causality_circulation_dynamics,"Confounding by ENSO, modes of variability, land forcing, aerosols, or background warming is not controlled.",Condition on or remove major confounders and test mechanism persistence.,major,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P044,mechanism_causality_circulation_dynamics,Causal verbs are used where the design supports association only.,Replace causal language or add causal design.,minor_major,77,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P045,uncertainty_statistics_skill_metrics,"Uncertainty intervals omit structural, product, ensemble, or parameter uncertainty.",Propagate all uncertainty sources that materially affect the central claim.,central,46,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P046,uncertainty_statistics_skill_metrics,"Statistical significance is reported without multiple-testing, autocorrelation, spatial dependence, or effective sample size correction.","Use field significance, block bootstrap, hierarchical uncertainty, or equivalent methods.",major,46,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P047,uncertainty_statistics_skill_metrics,Forecast or model skill metrics are insufficiently tied to the atmospheric phenomenon being claimed.,"Use phenomenon-specific metrics for extremes, spectra, circulation, fluxes, and vertical structure.",major,46,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P048,uncertainty_statistics_skill_metrics,Validation is not independent of model/product construction.,"Use withheld data, independent observations, or out-of-sample tests.",central,46,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P049,reproducibility_data_code,"Data, code, configuration, or model version information is insufficient for reproduction.","Provide exact data versions, preprocessing, scripts, configuration, and random seeds where relevant.",major,24,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P050,reproducibility_data_code,Complex workflows lack provenance for intermediate products.,"Document workflow DAG, checksums, generated products, and quality-control flags.",major,24,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P051,reproducibility_data_code,"Large atmospheric datasets are cited generically without temporal, spatial, or processing specification.","Specify product version, access date, subset, processing level, and transformations.",minor_major,24,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P052,detail_audit,"Definitions, units, thresholds, signs, or abbreviations are inconsistent across text, figures, tables, and methods.","Audit terminology, units, and sign conventions throughout.",minor,187,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P053,detail_audit,"Figure captions do not state data source, period, region, uncertainty, or sample size.",Make each figure interpretable without relying on hidden methods.,minor,187,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P054,detail_audit,Supplementary methods contain essential information not visible in the main evidence chain.,Move central validation and uncertainty information into the main text or extended data summary.,minor_major,187,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P055,detail_audit,"Terminology confuses weather forecast skill, climate projection, attribution, and mechanism diagnosis.","Use distinct language for prediction, projection, attribution, and explanation.",minor_major,187,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
P056,reproducibility_data_code,"Reviewers request clearer availability of trained AI weights, inference code, and benchmark scripts.","Provide archived code, model weights where possible, executable benchmarks, and environment files.",major,24,abstracted_from_public_peer_review_files_and_stress_test_anchors,not_stored
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 323kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 323kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 323kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall