# Chandrasekara and Shahidi test adaptive HF radar estimates of significant wave height [Canonical reading](https://escapemeridian.com/readings/chandrasekara-and-shahidi-test-adaptive-hf-radar-estimates-of-ea311b2d) Authors Senal Chandrasekara and Reza Shahidi investigate whether adaptive spatial processing can improve time-domain significant wave height (SWH) estimates from high-frequency (HF) radar data. Their method extends a physics-guided linear ordered-statistics estimator (LOSE): three trainable complex spatial filters process radar voltages, and ordered magnitudes from their outputs are mapped to SWH through least-squares regression. In measurements from an HF radar in Argentia, Canada, paired with directional-wave buoy observations, the selected adaptive configuration reduced root-mean-square error by approximately 14.3% while maintaining a Pearson correlation coefficient of 0.949. The authors present this as evidence of potential improvement over fixed Bartlett-based LOSE, not as a result established across independent tests. They note that feature counts and smoothing windows were selected using the evaluation interval, leaving the generality of the improvement unresolved. Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims. ## Original source arXiv physics.ao-ph · Article Original source date: Oct 26, 2026 (2026-10-26). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-06T11:03:13.178Z. First observed: 2026-10-28T04:04:35.058Z. Retrieved: 2026-10-29T18:46:32.355Z. - [Spatially-Adaptive Neural-Network Based Extraction of Significant Wave Heights from HF Radar Data](https://arxiv.org/abs/2610.36301) ## Limits The reported results use measurements from one radar location, Argentia, Canada; the source does not establish whether the improvement generalizes to other locations or radar setups. Feature counts and smoothing windows were selected using the evaluation interval, so independent testing is needed to establish the generality of the improvement. ## Open question Does the reported improvement hold in independent testing when feature counts and smoothing windows are selected without using the evaluation interval? ## Source-stated Chandrasekara and Shahidi extend a linear ordered-statistics estimator with three trainable complex spatial filters, mapping ordered magnitudes from the spatial outputs to SWH. > The LOSE framework is extended using three trainable complex spatial filters initialized around the buoy look direction. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source-stated The evaluation paired measurements from a 13.385-MHz, 12-receiver element WERA radar at Argentia, Canada, with directional-wave buoy observations. > Evaluation used measurements from a 13.385-MHz, 12-receiver element WERA radar at Argentia, Canada, paired with directional-wave buoy observations. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source-stated Under the study's chronological training-evaluation partition, the selected adaptive configuration reduced root-mean-square error by approximately 14.3% and maintained a Pearson correlation coefficient of 0.949. > Under a common chronological training-evaluation partition, the selected adaptive configuration reduced root-mean-square error by approximately $14.3\\%$, while maintaining the Pearson correlation coefficient at $0.949$. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source-stated The authors say independent testing is needed because feature counts and smoothing windows were selected using the evaluation interval. > Feature counts and smoothing windows were selected using the evaluation interval; independent testing is therefore needed to establish the generality of the improvement. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source and contributor context Chandrasekara, Senal, Shahidi, Reza · Article author These are credits stated by the source; they do not establish a profile match or authorship of every passage. ## Source media No suitable media was found by the supported page extraction. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## arXiv arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## TIME TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services. - [Explore this context](/entities/time) - [Reference](https://time.com/) ## Editorial provenance Version 1. Publication ID: 58d9ed71-e4f5-4de6-85b3-13d60d233aa9. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-06T11:03:13.178Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: 6f933eb476bdf3c898b30a39e0736ccb42b165d366b540596c3d07397461e151. Source text hash: 1caa897fce56b864f83d56efa88525e52d3ef4b7b8ccc1eee872f1d78fa34d65. ## Revision and source-check history ### Version 1 — 2026-11-06T11:03:13.178Z Authors Senal Chandrasekara and Reza Shahidi investigate whether adaptive spatial processing can improve time-domain significant wave height (SWH) estimates from high-frequency (HF) radar data. Their method extends a physics-guided linear ordered-statistics estimator (LOSE): three trainable complex spatial filters process radar voltages, and ordered magnitudes from their outputs are mapped to SWH through least-squares regression. In measurements from an HF radar in Argentia, Canada, paired with directional-wave buoy observations, the selected adaptive configuration reduced root-mean-square error by approximately 14.3% while maintaining a Pearson correlation coefficient of 0.949. The authors present this as evidence of potential improvement over fixed Bartlett-based LOSE, not as a result established across independent tests. They note that feature counts and smoothing windows were selected using the evaluation interval, leaving the generality of the improvement unresolved. ## Original source arXiv physics.ao-ph · Article Original source date: Oct 26, 2026 (2026-10-26). Date basis: Date basis not provided. Added to Escape Meridian: 2026-11-06T11:03:13.178Z. First observed: 2026-10-28T04:04:35.058Z. Retrieved: 2026-10-29T18:46:32.355Z. - [Spatially-Adaptive Neural-Network Based Extraction of Significant Wave Heights from HF Radar Data](https://arxiv.org/abs/2610.36301) ## Limits The reported results use measurements from one radar location, Argentia, Canada; the source does not establish whether the improvement generalizes to other locations or radar setups. Feature counts and smoothing windows were selected using the evaluation interval, so independent testing is needed to establish the generality of the improvement. ## Open question Does the reported improvement hold in independent testing when feature counts and smoothing windows are selected without using the evaluation interval? ## Source-stated Chandrasekara and Shahidi extend a linear ordered-statistics estimator with three trainable complex spatial filters, mapping ordered magnitudes from the spatial outputs to SWH. > The LOSE framework is extended using three trainable complex spatial filters initialized around the buoy look direction. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source-stated The evaluation paired measurements from a 13.385-MHz, 12-receiver element WERA radar at Argentia, Canada, with directional-wave buoy observations. > Evaluation used measurements from a 13.385-MHz, 12-receiver element WERA radar at Argentia, Canada, paired with directional-wave buoy observations. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source-stated Under the study's chronological training-evaluation partition, the selected adaptive configuration reduced root-mean-square error by approximately 14.3% and maintained a Pearson correlation coefficient of 0.949. > Under a common chronological training-evaluation partition, the selected adaptive configuration reduced root-mean-square error by approximately $14.3\\%$, while maintaining the Pearson correlation coefficient at $0.949$. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source-stated The authors say independent testing is needed because feature counts and smoothing windows were selected using the evaluation interval. > Feature counts and smoothing windows were selected using the evaluation interval; independent testing is therefore needed to establish the generality of the improvement. - [Supporting source](https://arxiv.org/abs/2610.36301) ## Source and contributor context Chandrasekara, Senal, Shahidi, Reza · Article author These are credits stated by the source; they do not establish a profile match or authorship of every passage. ## Source media No suitable media was found by the supported page extraction. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## arXiv arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed. - [Explore this context](/entities/arxiv) - [Reference](https://arxiv.org/) - [Reference](https://info.arxiv.org/help/api/index.html) ## TIME TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services. - [Explore this context](/entities/time) - [Reference](https://time.com/) ## Editorial provenance Version 1. Publication ID: 58d9ed71-e4f5-4de6-85b3-13d60d233aa9. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-06T11:03:13.178Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: 6f933eb476bdf3c898b30a39e0736ccb42b165d366b540596c3d07397461e151. Source text hash: 1caa897fce56b864f83d56efa88525e52d3ef4b7b8ccc1eee872f1d78fa34d65. ### uncertain — 2026-11-06T11:03:13.178Z — anthropic/claude-opus-5.5 Publication: 58d9ed71-e4f5-4de6-85b3-13d60d233aa9 attribution — pass: The reading correctly attributes the work to Senal Chandrasekara and Reza Shahidi and frames the method, results and caveats as the authors' claims. > Authors:Senal Chandrasekara, Reza Shahidi summary — uncertain: The reading is faithful and keeps the authors' own caveat, but a reader without specialist background would struggle with it. It never says in plain words what significant wave height measures or why HF radar wave estimates are useful. The baseline, fixed Bartlett-based LOSE, is also left opaque, as are ordered magnitudes, root-mean-square error and Pearson correlation. Smallest repair: keep the terms exactly, add term-library tooltips for SWH, LOSE, Bartlett beamforming, RMSE and Pearson correlation, and add one plain sentence saying the baseline is the authors' earlier fixed single-beam estimator. > In our previous study, a physics-guided linear ordered-statistics estimator (LOSE), motivated by first-order rough-surface scattering assumptions and radar-envelope statistics, mapped the largest rank-ordered magnitudes of Bartlett-beamformed voltages to buoy-observed SWH date — pass: The originalPublished date of 2026-10-26 matches the v1 submission date on arXiv. The 2026-10-28 feed date is later and is correctly not used. > \[Submitted on 26 Oct 2026\] support — pass: All four claims are verbatim-supported. The single-location limitation is phrased as a gap the source leaves open, not as a finding. The question follows directly from the authors' stated caveat about evaluation-interval selection. The 14.3% figure is not inflated into real-world reliability. > Feature counts and smoothing windows were selected using the evaluation interval; independent testing is therefore needed to establish the generality of the improvement. media — pass: The selected media is the verified arXiv abstract page for this paper, which is the direct original source. ## Recorded references Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter. No public references recorded here yet. [Reference data (JSON)](/readings/ea311b2d9d5cd2d180e793b020d4eab07e9928cc6052fffadf2b14ebd011e4ea/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/chandrasekara-and-shahidi-test-adaptive-hf-radar-estimates-of-ea311b2d.json)