{"id":"d1bc198a-c0de-49db-b366-02b173ec7890","arxiv_id":"2606.02005","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":2,"one_line_summary":"CORDEX-NAM12 simulations under SSP5-8.5 show a shift to persistence-dominated heatwave regimes in the continental interior by 2091-2100, with HWMD ~26.66 days/event, HWD ~69 days, HWI ~6.88 K, linked to lower-frequency temperature modes via mrDMD analysis.","lead":"The paper develops an integrated workflow using heatwave metrics and multiresolution dynamic mode decomposition on regional climate simulations to diagnose shifts in heatwave persistence and intensity over southern Canada. A smart generalist might read it to understand why some regions experience amplified heatwave duration under climate change beyond simple frequency increases.","discovery_kind":"unclear","skeptic_critique":{"model":"grok-4.3","headline":"Fixed historical 90th-percentile threshold may artifactually inflate future persistence metrics","rationale":"The reader's weakest_assumption already flags threshold sensitivity; the above test directly isolates whether that assumption is load-bearing for the headline numbers and dynamical interpretation. No other internal inconsistency (e.g., in mrDMD application or spatial-overlap logic) is evident from the given description.","tokens_in":1938,"tokens_out":340,"duration_ms":14667,"concrete_test":"Recompute all heatwave metrics (HWF, HWMD, HWD, HWI) on the identical CORDEX-NAM12 fields using a time-varying 90th-percentile threshold (15-day window centered on each calendar day, recomputed per decade); if the late-century HWMD increase falls below ~15 days/event or loses statistical significance while the mrDMD low-frequency participation ratios remain similar, the persistence-dominance claim is driven primarily by the fixed-threshold definition.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim of a shift to persistence-dominated regimes (HWMD ~26.66 days/event, HWD ~69 days) rests on events defined by a fixed 2001-2010 90th-percentile threshold applied to SSP5-8.5 runs. As the mean temperature rises, exceedances become both more frequent and mechanically longer even if the underlying variability spectrum is unchanged; mrDMD mode participation ratios and spatial alignments are then computed only on these threshold-exceeding days. This conditioning can couple the reported redistribution toward lower-frequency modes and weaker damping directly to the threshold choice rather than to genuine dynamical changes.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper develops an event-dynamical workflow that identifies warm-season heatwaves over southern Canada using a fixed 2001-2010 90th-percentile threshold and minimum-duration criterion on CORDEX-NAM12 output, computes standard metrics (HWF, HWN, HWMD, HWD, HWI, HWM), and links them to multiresolution dynamic mode decomposition (mrDMD) via heatwave-conditioned mode participation ratios and spatial alignment analyses. It reports a shift to persistence-dominated regimes in the continental interior under SSP5-8.5, with late-century values reaching HWMD ≈ 26.66 days/event, HWD ≈ 69 days, and HWI ≈ 6.88 K, accompanied by redistribution toward lower-frequency mrDMD modes and weaker damping.","tokens_in":2089,"tokens_out":570,"duration_ms":16244,"significance":"If the reported dynamical linkages are robust, the work offers a mechanistic bridge between multiscale temperature variability and regional heatwave amplification, moving beyond frequency-only projections; the mrDMD-based participation and alignment diagnostics represent a constructive approach for process-informed attribution in regional climate modeling.","major_comments":[{"comment":"Abstract and Methods (threshold definition): The central persistence claims (HWMD reaching 26.66 days/event and HWD ≈ 69 days) rest on events defined by a fixed 2001-2010 90th-percentile threshold applied to SSP5-8.5 runs. Because mean warming increases the frequency and duration of exceedances even if the variability spectrum is unchanged, the subsequent mrDMD conditioning on those days risks coupling the reported shift to lower-frequency modes and weaker damping directly to the threshold choice rather than to independent dynamical evolution. A sensitivity analysis with time-varying or relative thresholds is required to establish that the persistence amplification is not an artifact of the fixed-threshold definition.","section":"Abstract and Methods"},{"comment":"Results (numerical projections): The reported values (HWMD 26.66 days/event, HWI 6.88 K) are presented without accompanying uncertainty ranges, ensemble spread, or baseline comparisons against the reference period; this weakens the ability to judge whether the interior-coastal contrast exceeds internal variability or model bias.","section":"Results"}],"minor_comments":[{"comment":"The 15-day moving window for the percentile threshold is mentioned but its sensitivity to window length is not quantified.","section":"Methods"},{"comment":"Figure captions should explicitly state the reference period and threshold used for each panel to avoid ambiguity when comparing periods.","section":"Figures"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive and detailed comments. We address each major comment point by point below, indicating where revisions will be incorporated.","responses":[{"response":"We agree that a fixed historical threshold can increase exceedance frequency and duration under mean warming, potentially affecting the conditioning step. The mrDMD decomposition itself is performed on the full temperature field without reference to the threshold, and mode participation ratios are computed conditionally afterward. This design aims to separate dynamical changes from threshold effects. To directly address the concern, we will add a sensitivity analysis using a time-varying (period-specific) 90th-percentile threshold in the revised manuscript and report the resulting changes in HWMD, HWD, and mrDMD participation ratios.","revision_made":"yes","referee_comment":"[Abstract and Methods] Abstract and Methods (threshold definition): The central persistence claims (HWMD reaching 26.66 days/event and HWD ≈ 69 days) rest on events defined by a fixed 2001-2010 90th-percentile threshold applied to SSP5-8.5 runs. Because mean warming increases the frequency and duration of exceedances even if the variability spectrum is unchanged, the subsequent mrDMD conditioning on those days risks coupling the reported shift to lower-frequency modes and weaker damping directly to the threshold choice rather than to independent dynamical evolution. A sensitivity analysis with time-varying or relative thresholds is required to establish that the persistence amplification is not an artifact of the fixed-threshold definition."},{"response":"We acknowledge that the reported point values lack explicit uncertainty quantification. The analysis is based on a single CORDEX-NAM12 simulation (CRCM5 driven by CanESM5), precluding ensemble spread. We will incorporate direct baseline comparisons of all metrics against the 2001-2010 reference period and explicitly discuss the single-model limitation when interpreting the interior-coastal contrasts. However, ensemble-based uncertainty ranges cannot be added without additional simulations.","revision_made":"partial","referee_comment":"[Results] Results (numerical projections): The reported values (HWMD 26.66 days/event, HWI 6.88 K) are presented without accompanying uncertainty ranges, ensemble spread, or baseline comparisons against the reference period; this weakens the ability to judge whether the interior-coastal contrast exceeds internal variability or model bias."}],"tokens_in":1642,"tokens_out":530,"duration_ms":26950,"standing_objections":["Provision of ensemble spread or multi-model uncertainty ranges, as the study employs only one regional climate model simulation."]},"desk_editor":{"model":"grok-4.3","letter":"The paper's core move is to run mrDMD on daily temperature fields, then condition the mode participation ratios and spatial overlaps only on days that exceed a fixed 2001-2010 90th-percentile threshold. That produces the headline numbers: HWMD rising to ~26.66 days per event and HWD to ~69 days by 2091-2100 in the interior, with a shift toward lower-frequency modes and weaker damping.\n\nWhat is actually new is the specific workflow that ties heatwave-conditioned mrDMD participation ratios and hotspot overlap to persistence and intensity metrics on CORDEX-NAM12 output. The regional contrast between interior and coastal responses is cleanly shown and the method is reproducible from the description.\n\nThe soft spot is the threshold itself. Once the mean temperature rises, a fixed percentile from the reference decade will be exceeded on more consecutive days even if the variability spectrum stays the same. Because the mrDMD diagnostics are computed only on those exceedance days, the reported redistribution to low-frequency modes and the alignment with persistence hotspots can be partly an artifact of the conditioning rather than an independent dynamical change. The abstract gives no sign that the authors tested a time-varying threshold or a percentile that follows the mean, nor do they report uncertainty on the 26.66 and 6.88 K figures.\n\nThis is the kind of paper that belongs in a regional climate-dynamics journal. Readers who already work with event-based diagnostics and DMD-type decompositions will get the most out of the workflow details. It is coherent enough on its own terms to deserve referee time, but the threshold sensitivity should be the main point the reviewers press.","headline":"The fixed historical threshold likely drives much of the reported persistence increase mechanically, so the dynamical link to low-frequency modes needs stronger checks.","tokens_in":2592,"tokens_out":408,"would_cite":false,"duration_ms":12487,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Climate change shifts North American heatwaves toward longer, persistence-dominated events in continental interiors through changes in low-frequency temperature modes.","keywords":["heatwave persistence","climate change","multiresolution dynamic mode decomposition","regional climate modeling","temperature variability","North America","event intensity"],"falsifier":"Observational records or independent model ensembles for the late 21st century showing no substantial rise in interior heatwave duration or no spatial alignment between low-frequency modes and persistence hotspots would falsify the central claim.","tokens_in":2839,"feed_emoji":"🌡️","tokens_out":761,"duration_ms":22567,"temperature":0.7,"pith_summary":"The paper develops an integrated workflow that identifies heatwaves using a fixed historical threshold and minimum duration, then connects their statistics to multiscale temperature structures via multiresolution dynamic mode decomposition applied to regional climate model output. It examines three future periods under a high-emissions scenario over southern Canada and finds that interior regions experience a clear transition to longer-lasting events with higher intensity, while coastal areas change less. Dynamical analysis shows dominant activity moving to lower-frequency levels with reduced damping in those interiors, and low-frequency modes align spatially with the locations of strongest persistence and intensity increases. A sympathetic reader would care because this provides a process-based explanation for why event duration and strength amplify beyond simple frequency counts, with direct implications for regional risk assessment.","feed_headline":"Interior heatwaves grow longer and stronger by late century","feed_subtitle":"Simulations link the shift to redistribution of activity into lower-frequency modes with weaker damping in continental regions.","key_machinery":"multiresolution dynamic mode decomposition (mrDMD) of the daily mean temperature field, connected to heatwave metrics through heatwave-conditioned mode participation ratios and spatial alignment analyses.","core_discovery":"Results show a clear shift toward persistence-dominated heatwave regimes in the continental interior. By the late century, increases in seasonal heatwave days are accompanied by much longer events, with regional HWMD reaching about 26.66 days/event and HWD about 69 days, together with stronger above-threshold intensity, with HWI reaching about 6.88 K. Dynamical diagnostics indicate a redistribution of dominant activity toward lower-frequency levels and weaker effective damping in interior regions, while coastal and maritime regions show smaller changes. Heatwave-relevant low-frequency modes remain active during long events and align with persistence and intensity hotspots, supporting a proce","pith_inferences":["Adaptation planning for prolonged heat events may need to focus more on interior continental zones than coastal ones.","The workflow could be reapplied to other variables such as humidity or precipitation to check whether similar multiscale shifts appear.","If low-frequency modes prove robust across models, targeted observational campaigns could test whether those modes strengthen in reality as projected."],"forward_implications":["Seasonal heatwave days increase alongside much longer individual events in continental interiors.","Above-threshold intensity strengthens, reaching regional values around 6.88 K by late century.","Dominant dynamical activity redistributes toward lower-frequency levels with weaker effective damping inland.","Coastal and maritime regions exhibit comparatively small shifts in the same metrics.","Low-frequency modes stay active during extended events and overlap with persistence and intensity hotspots."],"fun_headline_variants":["Interior heatwaves reach average duration of 26 days by 2100","Multiscale modes align with persistence hotspots in interior regions","Late-century heatwaves show HWD of 69 days in continental areas","Weaker damping links to longer heatwave events under SSP5-8.5"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The fixed historical 90th percentile threshold, minimum-duration rule, and mrDMD mode participation on this particular regional model output capture the true causal drivers of persistence changes without major influence from model biases or threshold sensitivity.","fun_headline_variants_meta":{"raw":{"variants":["Interior heatwaves reach average duration of 26 days by 2100","Multiscale modes align with persistence hotspots in interior regions","Late-century heatwaves show HWD of 69 days in continental areas","Weaker damping links to longer heatwave events under SSP5-8.5"]},"model":"grok-4.3","cost_usd":0.007922,"raw_usage":{"total_tokens":3695,"prompt_tokens":837,"num_sources_used":0,"completion_tokens":75,"cost_in_usd_ticks":79224500,"prompt_tokens_details":{"text_tokens":837,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2783,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":837,"tokens_out":75,"duration_ms":17659,"temperature":1.0,"reasoning_tokens":2783,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-28T11:55:19.837041+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Observational records or independent model ensembles for the late 21st century showing no substantial rise in interior heatwave duration or no spatial alignment between low-frequency modes and persistence hotspots would falsify the central claim.","supporting_citations":[],"review_version":1}