Earth and Space Science最新文献

{"title":"The Ceiling Height of Wildland Fire Plumes in Quasi-Steady Stratified and Sheared Boundary Layers","authors":"Jie Sun,&nbsp;Kevin Speer,&nbsp;Bryan Quaife,&nbsp;Ming Cai,&nbsp;David Schvartzman","doi":"10.1029/2025EA004818","DOIUrl":"https://doi.org/10.1029/2025EA004818","url":null,"abstract":"<p>Radar observations from a prescribed fire experiment reveal a large-scale, billow-like vorticity structure associated with the plume head near the onset of plume bending. This bending limits the vertical extent of the plume and defines the characteristic plume ceiling height. This study investigates plume bending under idealized sheared-crossflow and quasi-steady stratified atmospheric boundary layer conditions. Large-eddy simulations using the Cloud Model 1 are conducted under idealized boundary-layer configurations to qualitatively reproduce the observed plume morphology and to examine plume-head evolution under varying fire-generated surface heat-flux, background shear, and stratification. Building on the classic theory, this study presents a scaling framework for estimating the characteristic plume ceiling height based on a modified Byram's convective number that accounts for sheared crossflow. The proposed scaling successfully reproduces the behavior of the idealized large-eddy simulations and highlights the roles of shear and stratification in controlling the characteristic plume ceiling height.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2025EA004818","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring New Instrument deGradation Models and Algorithms (ENIGMA) for the SORCE–Solar Irradiance Monitor","authors":"Stéphane Béland,&nbsp;Steven V. Penton","doi":"10.1029/2026EA005255","DOIUrl":"https://doi.org/10.1029/2026EA005255","url":null,"abstract":"<p>The Spectral Irradiance Monitor (SIM), on board the Solar Radiation and Climate Experiment (SORCE) test satellite operated for 17 years: from February 2003 to February 2020, measuring the Solar Spectral Irradiance from 200 to 2,400 nm. The unavoidable instrument degradation was tracked and corrected based on a measurement equation derived from accessible telemetry, the known instrument refraction geometry, inter-detector comparisons, and inter-spectrometer comparisons. While the previous degradation model captures much of the long-term trending, some of the parameters were adjusted without well-defined physical justifications. The development of enhanced SORCE/SIM measurement equations allowed the evaluation and inclusion of degradation mechanisms not captured in the original model. We present the multiple approaches tested to better understand the degradation mechanisms and the final model that was adopted with the underlying physical implications. An updated SORCE/SIM data product, L3 Enigma Version 1 (L3E V1), covering the lifetime of the mission, starting at 210.0 nm instead of 240.0 nm, and adjusted to the TSIS/SIM absolute irradiances, has been generated and made available to the community.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2026EA005255","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148783446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Localization and Classification of Gravity Wave Events From VIIRS Day/Night Band Satellite Imagery Using Machine Learning Techniques","authors":"Yuta Hozumi,&nbsp;Jia Yue,&nbsp;Seraj Al Mahmud Mostafa,&nbsp;Chenxi Wang,&nbsp;Jianwu Wang,&nbsp;Sanjay Purushotham,&nbsp;Steven D. Miller","doi":"10.1029/2026EA005240","DOIUrl":"https://doi.org/10.1029/2026EA005240","url":null,"abstract":"<p>Gravity waves in the upper mesosphere produce diverse spatial patterns in nighttime airglow, but their global morphology has been difficult to characterize because manual identification is impractical for the massive satellite image archive. We developed a machine learning framework to detect and classify gravity wave events in imagery from the Visible Infrared Imaging Radiometer Suite Day/Night Band (VIIRS DNB) on three satellites. A machine learning based object detector called YOLOv8 (You Only Look Once, version 8) was trained to identify four airglow morphological classes: concentric gravity waves, frontal waves, ripples, and other gravity wave events. On an independent test set, the model achieved a mean average precision of 0.832 at an intersection over union threshold of 0.3. At the confidence threshold adopted for catalog generation, the mean precision and recall were 0.828 and 0.778, respectively. The trained detector was applied to 12 years of moon-free VIIRS DNB observations from Suomi National Polar-orbiting Partnership, NOAA-20, and NOAA-21, resulting in an event catalog containing more than 100,000 detected wave event instances from more than 65,000 images with positive detections, which was archived in a public repository. The catalog captures physically plausible large-scale gravity wave occurrence patterns while also showing that detectability is reduced under elevated sensor noise and strong artificial light contamination. These results demonstrate that machine learning based object detection can systematically extract diverse mesospheric gravity wave events from the long-term VIIRS DNB nightglow record.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2026EA005240","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752877","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Can a Persistence-Based Prediction of GRACE Gravity Field Coefficients Reduce Uncertainties at Finer Temporal Scales?","authors":"D. Kankanige,&nbsp;B. D. Vishwakarma,&nbsp;Y. Y. Liu,&nbsp;A. Sharma","doi":"10.1029/2025EA004554","DOIUrl":"https://doi.org/10.1029/2025EA004554","url":null,"abstract":"<p>Enhancing the applicability of Gravity Recovery and Climate Experiment (GRACE) products poses challenges due to the trade-off between the temporal and the spatial resolution within limits posed by the mission design. If a Kalman filter framework was adopted to characterize the relationship between monthly and finer temporal scales, it could facilitate the prediction of uncertainty associated with rapid mass changes at the higher temporal resolution. This study investigates the following key questions: (a) whether a persistence-based prediction of gravity coefficients effectively reduces the uncertainties of finer temporal scale-gravity coefficients to levels comparable with the known uncertainties of monthly gravity coefficients; and (b) the spectral degrees at which this consistency is most pronounced. The application of the resulting predictive approach to GRACE level-2 data demonstrates its effectiveness in reducing uncertainty at high spherical harmonic degrees. The prediction approach allows for the estimation of sub-monthly and weekly gravity coefficients with uncertainties that are comparable to the monthly resolution-uncertainties for specific ranges of degrees. This is validated by the degree variance of the finer resolution-estimates which largely align with those of the monthly raw gravity coefficients across the higher spherical harmonic degrees for sub-monthly and weekly data. Further, this predictive framework, on a univariate basis, demonstrates its potential to mitigate the correlated errors to a useful extent as observed in the spectral and spatial domains. Reducing uncertainties at finer temporal scales for higher spherical harmonic degrees to levels comparable with monthly data enhances the assessment of rapid mass changes in localized hydrological studies.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2025EA004554","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Supervised Machine Learning Approach to Iron Meteorite Classification Using Random Forest: A Predictive Model to Classify Ungrouped Irons","authors":"A. F. Rogers,&nbsp;L. Doucet,&nbsp;L. V. Forman,&nbsp;K. Rankenburg,&nbsp;G. K. Benedix","doi":"10.1029/2026EA005211","DOIUrl":"https://doi.org/10.1029/2026EA005211","url":null,"abstract":"<p>Iron meteorites provide insight into the formation processes of planetary cores from the early Solar System. Through further classifying individual irons into specific groups, which are assumed to represent a common parent body, greater information can be yielded into the formation processes of those bodies. The current classification system of iron meteorites includes 13 groups distinguished by trace element compositions and mineralogical properties (i.e., bandwidth size of the α-Fe,Ni-alloy, silicate inclusions, etc.). Group sizes vary with only 6 members identified as IIG irons and 391 members in the largest group, the IAB-complex. One-hundred and fifty-nine irons (or &gt;11% of all iron meteorites) that cannot be grouped with the current classification scheme are labeled “ungrouped.” This research explores whether supervised machine learning (ML) algorithms offer insights into classification of the ungrouped irons, despite the heavily imbalanced group data used for algorithm training. A Random Forest classifier is presented in this work that achieved weighted F1-scores of ∼91–97% of testing data sets. This outcome indicated that ML is possible for the iron meteorite data set. The accuracy of the model ranges from ∼90–98%. As a result, at least 59 ungrouped irons could be reclassified to preexisting iron groups, 16 may belong to small grouplets while 12 should remain ungrouped. This work further discusses the implications of these potential reclassifications and how ML may be applicable to other meteorite studies.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2026EA005211","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"On the Generation of Pre-Sunrise Equatorial Plasma Bubbles in Southeast Asia During the Geomagnetic Storm of May 2024","authors":"Sri Ekawati,&nbsp;Prayitno Abadi,&nbsp;Takuya Sori,&nbsp;Rezy Pradipta,&nbsp;Ihsan Naufal Muafiry,&nbsp;Teguh Nugraha Pratama,&nbsp;Agri Faturahman,&nbsp; Noersomadi,&nbsp;Febrylian Fahmi Chabibi,&nbsp;Punyawi Jamjareegulgarn","doi":"10.1029/2025EA004969","DOIUrl":"https://doi.org/10.1029/2025EA004969","url":null,"abstract":"<p>A set of fresh Equatorial Plasma Bubbles (EPBs) formed during pre-sunrise hours in the Southeast Asian sector amid the recovery phase of May 2024 geomagnetic storm. They offered new insights into their formation mechanism. This study used Total Electron Content (TEC) and Rate of TEC Index maps over Indonesia, along with ionosondes in Thailand and Vietnam, magnetometers in South American sectors, and prompt penetration electric field (PPEF) modeling, to reveal these insights. The first insight from our observations was that the pre-sunrise EPBs were initiated by the eastward overshielding PPEF linked to the northward turn of the Interplanetary Magnetic Field Bz. Then, the disturbance dynamo electric field, expected to operate during the storm's recovery phase, might maintain the survival of the EPBs. The second, and more intriguing, insight is that these pre-sunrise EPBs formed at specific longitudes, where TEC values had already decreased since the post-sunset hours. Further investigation showed that these lower TEC regions consist of large plasma density gradients, serving as favorable conditions necessary for EPB formation near sunrise. Finally, this study implied that, even during a strong eastward electric field in an intense storm, preconditioning or favorable conditions, such as larger plasma density gradients, are required for pre-sunrise EPB generation. The storm-driven electric field then causes the pre-sunrise EPBs to develop at distinctive longitudes where these favorable conditions are located.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2025EA004969","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using Machine Learning to Predict Phytoplankton Blooms in the Salish Sea","authors":"Ilias Bougoudis,&nbsp;Karyn Suchy,&nbsp;Susan E. Allen,&nbsp;Matias Salibian-Barrera","doi":"10.1029/2025EA004826","DOIUrl":"https://doi.org/10.1029/2025EA004826","url":null,"abstract":"<p>Strong phytoplankton blooms occur every spring in the Salish Sea, but vary significantly, spatially and temporally. Natural variability and climate change make the prediction of the bloom challenging, as they affect both atmospheric drivers and nutrient interactions that contribute to the growth of phytoplankton. Understanding these drivers is essential for a successful bloom prediction. Here, we used a 25-year data set (18 training, 4 evaluation, and 1 independent year), consisting of 7 atmospheric drivers and 4 phytoplankton variables for the Salish Sea, to build Machine Learning (ML) models that are able to emulate phytoplankton blooms. For phytoplankton productivity, we employed histogram gradient boosting models, whereas for the biomass of phytoplankton, we employed functional regression models, to capture its dynamic nature and dependence on the atmospheric drivers. Feature selection was implemented to identify the optimal set of input features for each model. By using only meteorological drivers (and in some cases spatiotemporal variables) as inputs, the proposed ML models are able to efficiently capture spatial and temporal patterns of phytoplankton blooms over the Salish Sea. Different phytoplankton variables (e.g., biomass, production rates) and different regions require different input features (e.g., precipitation), with shortwave solar radiation being the most frequently used atmospheric driver.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2025EA004826","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Formulation and Implementation of a Dynamic Roughness Length for Improved Surface Wind Speeds in WRF","authors":"S. Srivastava,&nbsp;K. R. Schell","doi":"10.1029/2025EA004980","DOIUrl":"https://doi.org/10.1029/2025EA004980","url":null,"abstract":"<p>Accurate near-surface winds from the Weather Research and Forecasting (WRF) model are essential for wind-energy applications, yet persistent biases remain that depend on terrain, season, and land cover. This study develops and evaluates an observation-informed, time-varying aerodynamic roughness length for WRF. The roughness length is inferred from the neutral log-law using measured 10-m winds and friction velocity from control WRF runs. A random-forest model, trained on local meteorology is then used to produce a continuous 3-hourly dynamic roughness length series. The dynamic roughness length is implemented in WRF by overwriting its default values each cycle while retaining the local land-surface properties. The framework is tested at three Texas environments: complex topography, homogeneous cropland, and a coastal site during January and July 2021. Across sites, the inferred roughness length frequently exceeds WRF's default bounds, especially in summer, indicating under-representation of effective surface drag in default tables. With the dynamic roughness length implemented into WRF, model performance improves across sites: mean-bias magnitude decreases, point-by-point scatter tightens, explained variance increases, and high-wind events are captured with smaller errors. These gains appear alongside frequent exceedance of the static WRF default bounds, indicating that observation-informed roughness more faithfully represents effective surface drag. Overall, allowing roughness length to vary in time, anchored by local measurements—provides a practical and physically motivated lever to reduce near-surface wind errors in WRF and highlights limitations of fixed, class-based roughness values.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2025EA004980","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148647663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Migrating and Nonmigrating Neutral Mass Density Tides From GRACE-FO Observations for Solar Minimum Conditions","authors":"Mukta Neogi,&nbsp;Jens Oberheide","doi":"10.1029/2026EA005322","DOIUrl":"https://doi.org/10.1029/2026EA005322","url":null,"abstract":"<p>This study presents the migrating and nonmigrating thermospheric neutral mass density tides from the Gravity Recovery And Climate Experiment Follow-On (GRACE-FO) observations under solar minimum conditions (year 2020). The observed densities are first mapped to a reference altitude of 500 km using an empirical model (NRLMSIS 2.0) to remove the impact of the elliptical orbit. To address incomplete local solar time sampling due to the slow orbit precession and to achieve full local solar time coverage, 161-day composite windows centered on each analysis day are constructed. A two-dimensional fast Fourier transform is then applied in time and longitude to extract the amplitudes and phases of migrating tides, including DW1, SW2, TW3, and QW4, as well as nonmigrating tides such as DE3, DW2, and SE2. The retrieved tides show distinct seasonal-latitudinal variability. Migrating DW1 and SW2 are the dominant contributors to the density tidal spectrum, while higher-order migrating components such as TW3 and QW4 also produce significant variability at 500 km. A consistency check is also performed using the independent CHAllenging Minisatellite Payload (CHAMP) relative density tidal data set under comparable solar minimum conditions. The GRACE-FO tidal structures show good agreement with CHAMP relative density tides, supporting the reliability of the tidal diagnostics. These results provide new observational constraints on upper-atmospheric tidal dynamics and establish a framework for using GRACE-FO neutral density measurements to study vertical coupling and thermospheric variability, with important implications for satellite orbit-drag prediction.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2026EA005322","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148617114","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI Based Velocity Calibration for Cassini's Cosmic Dust Analyzer (CDA)","authors":"Thomas Albin,&nbsp;Jonas Simolka,&nbsp;Ralf Srama","doi":"10.1029/2026EA005130","DOIUrl":"https://doi.org/10.1029/2026EA005130","url":null,"abstract":"&lt;p&gt;This paper presents a novel AI-based velocity calibration method for the Cosmic Dust Analyzer (CDA) onboard the Cassini spacecraft, in particular for impacts on the large Impact Ionization Detector (IID). The legacy polynomial calibration function, derived from laboratory measurements at a dust accelerator facility, assigns a velocity-independent relative uncertainty of &lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mi&gt;σ&lt;/mi&gt;\u0000 &lt;mo&gt;=&lt;/mo&gt;\u0000 &lt;mn&gt;0.62&lt;/mn&gt;\u0000 &lt;mo&gt;⋅&lt;/mo&gt;\u0000 &lt;mi&gt;v&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt; $sigma =0.62cdot v$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt; and is restricted to signals with a peak amplitude above &lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mn&gt;20&lt;/mn&gt;\u0000 &lt;mspace&gt;&lt;/mspace&gt;\u0000 &lt;mi&gt;f&lt;/mi&gt;\u0000 &lt;mi&gt;C&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt; $20 mathrm{f}mathrm{C}$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt; for the QI channel. We propose a two-stage deep learning architecture that combines a Temporal Convolutional Network Autoencoder for unsupervised feature extraction with a Mixture Density Network for probabilistic velocity prediction. The Autoencoder compresses the 510-sample QI impact signals into a 3-dimensional latent space that captures the essential morphological characteristics of each impact event, while the Mixture Density Network maps this latent representation to a conditional Gaussian distribution, providing both a mean velocity estimate and an individual, signal-dependent uncertainty. Trained and validated on 839 laboratory calibration events from two measurement campaigns, the method achieves a mean relative uncertainty of approximately &lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mn&gt;30&lt;/mn&gt;\u0000 &lt;mi&gt;%&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt; $30%$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt; for velocities above &lt;span&gt;&lt;/span&gt;&lt;math&gt;\u0000 &lt;semantics&gt;\u0000 &lt;mrow&gt;\u0000 &lt;mn&gt;3&lt;/mn&gt;\u0000 &lt;mspace&gt;&lt;/mspace&gt;\u0000 &lt;mi&gt;k&lt;/mi&gt;\u0000 &lt;mi&gt;m&lt;/mi&gt;\u0000 &lt;mo&gt;/&lt;/mo&gt;\u0000 &lt;mi&gt;s&lt;/mi&gt;\u0000 &lt;/mrow&gt;\u0000 &lt;annotation&gt; $3 mathrm{k}mathrm{m}/mathrm{s}$&lt;/annotation&gt;\u0000 &lt;/semantics&gt;&lt;/math&gt;, a substantial improvement over the legacy function. Furthermore, the amplitude threshold for evaluating a signal is reduced from 20 fC to 9.7 fC, significantly expanding the population of calibratable dust impacts in the Cassini CDA archive. The Autoencoder's reconstruction loss additionally serves as an anomaly detector, flagging signals that fall outside the learned calibration space. These improvements enable more accurate d","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":"13 8","pages":""},"PeriodicalIF":3.3,"publicationDate":"2026-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2026EA005130","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148617115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}