Advances in Space Research最新文献

{"title":"Extreme temporal clustering of solar flares in AR 13663/13664: magnetic topology and recurrence time in May 2024","authors":"Bruno Fernandes Garcia,&nbsp;Joaquim Eduardo Rezende Costa,&nbsp;Luis Eduardo Antunes Vieira,&nbsp;Alisson Dal Lago,&nbsp;Christofher do Vale Pena,&nbsp;José Matheus da Silva Rocha,&nbsp;Idowu Raji","doi":"10.1016/j.asr.2026.04.125","DOIUrl":"10.1016/j.asr.2026.04.125","url":null,"abstract":"<div><div>Solar Cycle 25’s most extreme episode occurred in early May 2024 when two magnetically complex active regions (AR 13663, AR 13664) generated 17 X-class flares in two weeks, driving G5 geomagnetic storms unseen since 2003. Using SDO/HMI SHARP vector magnetic indices and GOES/XRS soft X-ray data, we characterize flare recurrence times, magnetic evolution, and geomagnetic response. AR 13663 displayed concentrated activity (peak: 18 flares on May 4, median recurrence time 1.2 h), while AR 13664 exhibited sustained multi-peaked productivity (13 flares on May 10, median recurrence time 1.6 h). Critically, peak X-ray flux and recurrence time are decoupled (<span><math><mrow><mi>ρ</mi></mrow></math></span> &lt; 0.25), indicating that flare waiting times are controlled by polarity inversion line (PIL) topology and spatial distribution of magnetic shear rather than individual event energetics. A threshold of the SHARP parameter <em>SHRGT45</em> (fraction of pixels with magnetic shear angle exceeding <span><math><mrow><msup><mrow><mn>45</mn></mrow><mrow><mo>∘</mo></mrow></msup></mrow></math></span>) &gt; <span><math><mrow><mn>40</mn><mo>%</mo></mrow></math></span> marks transitions into hyper-productive regimes (&gt;10 flares/day). Comparison with 2003 (AR 10486) and 2017 (AR 12673) extremes shows that May 2024 represents a new extremal type: record regions (4253 MH) sustaining dense flare sequences. These findings emphasize PIL complexity as a primary predictor of flare cadence, with implications for space weather forecasting.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 1684-1701"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148528955","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":"Enhancing short-term Zenith Wet Delay prediction using a combined data assimilation and conditional GAN approach","authors":"Masoud Dehvari,&nbsp;Saeed Farzaneh,&nbsp;Ehsan Forootan","doi":"10.1016/j.asr.2026.04.077","DOIUrl":"10.1016/j.asr.2026.04.077","url":null,"abstract":"<div><div>Accurate short-term prediction of the Zenith Wet Delay (ZWD) is essential for enhancing the precision of GNSS positioning, numerical weather prediction, and tropospheric water vapor monitoring. Recently, combined data assimilation (C/DA) approaches have shown promising results in improving short-term ZWD estimates by integrating GNSS-derived observations with numerical model outputs. However, their prediction accuracy tends to decrease rapidly at longer forecast horizons due to the limited temporal resolution of background models and the highly variable nature of atmospheric water vapor. To address this limitation, this study proposes an enhanced predictive framework that integrates a Conditional Generative Adversarial Network (cGAN) with the C/DA-derived ZWD maps. The cGAN model is trained using sequences of 24-h C/DA ZWD maps as inputs and the corresponding 24-h-ahead C/DA analysis maps (0-h predictions) as targets, enabling the model to capture the nonlinear spatiotemporal evolution of tropospheric delay with higher fidelity. The proposed framework was evaluated over the year 2022 and compared with C/DA, GTrop, and Global Forecast System (GFS) models using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Correlation Coefficient (CC) metrics. Results demonstrate that the cGAN method consistently outperforms all reference models across all prediction horizons, achieving RMSE values between 0.95–4.5 cm and CC values between 0.99 and 0.86 for prediction from 1 to 24 h horizons. On average, the cGAN model improves RMSE by 33% and 38% compared to the C/DA and GFS models, respectively, and by approximately 50% compared to the GTrop model. These findings confirm that combining deep learning with data assimilation can significantly enhance the accuracy and stability of short-term ZWD forecasts, offering a robust and efficient solution for real-time atmospheric delay modeling and GNSS-based meteorological applications.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 728-745"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148529118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the impacts of long-term weather variability and urban development on crop production to analyze food security in Iran","authors":"Mohammad Kazemi Garajeh,&nbsp;Khalil Valizadeh Kamran,&nbsp;Bakhtiar Feizizadeh,&nbsp;Behnam Khorrami","doi":"10.1016/j.asr.2026.02.088","DOIUrl":"10.1016/j.asr.2026.02.088","url":null,"abstract":"<div><div>This study employed an integrated remote-sensing approach on the cloud-free Google Earth Engine (GEE) platform to assess the impacts of long-term weather trends and human activities, in particular, urban development on crop production to analyze food security in Iran from 2000 to 2022. A range variables such as air temperature (AT), actual evapotranspiration (AET), precipitation, snow cover, built-up area, surface water, and groundwater storage (GWS) were analyzed using pre-processed products from different resources (e.g., ERA5-Land) available within the GEE environment. Relationships between these variables and croplands were then examined using the Mann-Kendall (MK) non-parametric trend test to evaluate their effects on crop production. Our findings revealed correlation coefficients of −0.14, 0.20, 0.35, and 0.30 between croplands and AT, AET, precipitation, and snow cover, respectively. Furthermore, strong correlations between croplands and built-up areas (−0.80), surface water (0.78), and GWS (0.75) were found. The results suggest that precipitation and snow cover had moderate corelations with croplands, indicating that their decline between 2000 and 2022 affected crop production in Iran. In contrast, AT and AET showed weaker correlations with croplands. Urban expansion, however, had the most significant negative impact on crop production, while the reduction in crop production was also linked to a decrease in surface and groundwater resources over the same period. These factors contributed to a reduction in crop area by 7892.26 km², with a marked decline starting in 2007. Before this pivotal year, the average crop area was around 78,568.72 km², which dropped to 71,755.45 km² afterward. In conclusion, the findings underscore those variables (built-up areas, surface water, and GWS) had the negative influence on crop production, posing a direct threat to food security in the country.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 512-526"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148528778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Probabilistic principal component analysis of time variable gravity data considering measurement errors’ covariance matrix","authors":"Yong Feng,&nbsp;Guobin Chang,&nbsp;Nijia Qian,&nbsp;Yueyang Huan,&nbsp;Yu Cao,&nbsp;Yinxiao Sun","doi":"10.1016/j.asr.2026.04.008","DOIUrl":"10.1016/j.asr.2026.04.008","url":null,"abstract":"<div><div>Time-variable gravity data are often analyzed with non-parametric decompositions such as principal component analysis. However, for upstream products such as GRACE Level-2 solutions, the observations can contain heterogeneous and correlated measurement errors, whose correlations are reflected by a non-diagonal error covariance matrix. Fully accounting for this covariance information poses challenges to conventional PCA-type processing. To address this issue, we propose a new probabilistic PCA method that explicitly incorporates the measurement-error covariance structure. The proposed model decomposes the data into three parts: signals, noises, and measurement errors. The signal term is further represented by a small number of components that are orthogonal in both the temporal and spatial domains, consistent with the interpretability of many PCA-based approaches. The method is formulated in a measurement-equation-based probabilistic framework, and model parameters are estimated optimally in the maximum-likelihood sense. An efficient expectation–maximization (EM) algorithm is developed for iterative parameter estimation, and a suboptimal variant is also tested experimentally. Using GRACE Level-2 data as an example, the proposed approach successfully suppresses stripe errors for signal reconstruction and provides a compact set of orthogonal components for spatiotemporal signal analysis. Experimental results demonstrate that the proposed methods perform well in both signal reconstruction and spatiotemporal characterization.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 527-546"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148528851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Electrodynamics and ionospheric irregularities during intense geomagnetic storm that occurred in March 2024 over American and Antarctic sectors","authors":"A.J. de Abreu,&nbsp;E. Correia,&nbsp;E.P. Macho,&nbsp;K. Venkatesh,&nbsp;R. de Jesus,&nbsp;A. Pignalberi,&nbsp;M. Pezzopane,&nbsp;V.G. Pillat,&nbsp;P.R. Fagundes,&nbsp;M. Gende","doi":"10.1016/j.asr.2026.04.052","DOIUrl":"10.1016/j.asr.2026.04.052","url":null,"abstract":"<div><div>In this investigation, we present and discuss the effects of the intense geomagnetic storm that occurred in March 2024 on ionospheric electrodynamics and the ionospheric irregularities using observations from SuperDARN radar, magnetometers, equatorial electric field (EEF) model, GPS receivers, Swarm satellites, and ionosondes over American and Antarctic sectors. During the main phase of the storm, there is an increase in cross-polar cap convection reaching ∼58 and ∼68 kV in the Northern and Southern Hemispheres, respectively. EEJ observations show a strong and prolonged interval of negative values, while the EEF measurements indicate eastward/westward directions of the prompt penetration electric field (PPEF). The recovery phase of the storm is dominated by disturbance dynamo electric field (DDEF), which developed ∼1.5 h earlier in the main phase. VTEC values show a positive phase in mid- and low-latitudes in both hemispheres, reaching up to ∼150% at mid- and near high-latitudes and ∼30–60% at equatorial and low-latitudes. The eastern and western sectors show similar pattern. A combination of PPEF and disturbed thermospheric winds generated this positive phase in both sectors and hemispheres. During the recovery phase, the eastern and western sectors exhibit a weak positive and negative ionospheric storms in both Hemispheres, respectively, associated with DDEF. Ionospheric irregularities are observed at high-latitudes only in the Northern Hemisphere, which is characteristic signatures of storm enhanced density (SED) and polar cap patches (PCPs). S4 remained weak, below 0.2–0.3, at mid- and near high-latitudes in the Northern Hemisphere and moderate to strong, reaching 0.5–0.8, at equatorial and low-latitudes in the Southern Hemisphere, with suppression of irregularities during the night of 03–04 March.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 1427-1451"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148528990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Attention for GNSS height time series prediction: a comparative evaluation of modern transformer-based deep learning architectures","authors":"Mert Bezcioglu","doi":"10.1016/j.asr.2026.04.095","DOIUrl":"10.1016/j.asr.2026.04.095","url":null,"abstract":"<div><div>This study presents a comprehensive comparative evaluation of modern Transformer-based architectures for daily GNSS height time series prediction. Eight models, namely Transformer, Reformer, Autoformer, Informer, Crossformer, PatchTST, iTransformer, and TimeXer, were assessed using 14 years of continuous observations from 20 globally distributed GNSS stations. The dataset was partitioned chronologically into training (2012–2021), validation (2022–2023), and independent test (2024–2025) periods. A sliding window configuration with 365-day input length and one-day forecasting horizon was adopted. Performance was evaluated using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). Results show that architecture-specific design choices significantly influence predictive accuracy. PatchTST and iTransformer achieved the best overall performance, both reaching an aggregated RMSE of 5.1 mm and MAE of 3.7 mm across the test period. TimeXer and Crossformer followed with RMSE values of 5.3 mm and 5.5 mm, respectively. In contrast, Informer showed comparatively lower performance with an RMSE of 7.6 mm and MAE of 5.2 mm. Error distribution analysis revealed near-zero mean bias for all models, while dispersion differences clearly separated the architectures. PatchTST and iTransformer exhibited the lowest standard deviation of 5.1 mm, whereas Informer reached 7.5 mm. Station-level and monthly analyses confirmed that patch-based and variate-wise formulations provide superior robustness under heterogeneous seasonal and noise conditions. The findings demonstrate that modern time-series-oriented Transformer architectures outperform both the original Transformer and sparsity-based adaptations for millimeter-level GNSS height forecasting.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 957-972"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148529086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leaders and associated space debris around the Classical Laplace Plane","authors":"Elvis Lacruz,&nbsp;Daniel Casanova","doi":"10.1016/j.asr.2026.04.060","DOIUrl":"10.1016/j.asr.2026.04.060","url":null,"abstract":"<div><div>The geostationary orbit has been widely used by space agencies, public, and private companies following the launch of the inaugural communications satellite, Syncom-3, on August 19, 1964, positioned at 180 degrees longitude over the Pacific Ocean. The increase in the number of objects in the GEO zone is not solely due to the launch of new satellites into orbit, energetic breakups, or collisions. Rather, a variety of factors have contributed to generating new debris, including surface degradation, material deterioration, and the end-of-line retirement of satellites. The conflict arises from the layout of the GEO zone, as it is challenging to prevent collisions among uncontrolled objects orbiting there. Since the breakup events in the GEO zone of some orbiters for which data is available, such as Ekran 2, Breeze-M, Titan 3C Transtage, and Atlas Centaur, all distributed in the GEO region and around the Classical Laplace Plane. Even though they are located in this particular region, which presents certain stability, some of the objects exhibit different dynamics. The aim of this work is to better understand the dynamics of these objects, by analysing the correlation between the right ascension of the ascending node and the inclination, or by characterizing the space debris in terms of their orbital elements. Additionally, the identification of leaders and associated objects in different break-up events helps to understand which objects govern the dynamics of the debris cloud and which objects are associated with the primary object. These associated objects may eventually depart from the sub-cloud due to intrinsic physical properties, such as the area-to-mass ratio, <span><math><mrow><mi>A</mi><mo>/</mo><mi>m</mi></mrow></math></span>, as all objects in the region are subject to similar perturbative effects. To achieve this, we use a methodology that computes the maximum of the minimum distances between pairs of discretized orbits, propagating them within a perturbed model. This study provides insights into the behavior and evolution of space debris in the GEO region, which is vital for improving satellite collision avoidance strategies and managing the long-term sustainability of space activities. By identifying and characterizing the dynamics of debris from different breakup events, we contribute to a better understanding of the risk factors involved and provide valuable data for future debris mitigation measures.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 1274-1296"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148529055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantifying the thermal impact of road infrastructure on seasonal urban heat using graph theory and machine learning for sustainable planning in Makkah","authors":"Saeed Alqadhi","doi":"10.1016/j.asr.2026.04.108","DOIUrl":"10.1016/j.asr.2026.04.108","url":null,"abstract":"<div><div>Road network structure, often overlooked in urban climate studies, can significantly modulate land surface thermal behavior in hyper-arid cities. This study investigates the seasonal relationship between road network morphology and Land Surface Temperature (LST) in Makkah using an integrated framework combining ensemble machine learning, explainable artificial intelligence (XAI), and graph-theoretic analysis. Seven road network indices i.e. Alpha (α), Beta (β), Gamma (γ), Cyclomatic Number, Detour Index (DI), Eta (η), and Nearest Neighbour Index (NNI) were derived from OpenStreetMap within 1 km² fishnet grids and linked with seasonal daytime LST retrieved from cloud-free Landsat-8 observations for 2024. To isolate structural effects, vegetation, built-up intensity, and surface moisture were explicitly controlled using NDVI, NDBI, and NDWI. Among all evaluated models, the Random Forest ensemble consistently outperformed Gradient Boosting, XGBoost, SVR, Ridge, and MLP models across all seasons, achieving R² values of 0.91–0.93 and RMSE values of 0.33–0.36 °C. Feature importance and SHAP analyses indicate that built-up intensity (NDBI) is the dominant predictor of elevated LST across all road functional classes, while NDVI and NDWI exert seasonally varying cooling effects, particularly along residential and secondary roads. Road network structure further modulates localized heat accumulation: lower NNI values (denser road clustering) and higher DI values (greater route circuitousness) are associated with systematically higher LST, especially during summer and autumn. Model-derived thresholds suggest that DI &gt; 1.8 and Beta &gt; 0.6 are associated with LST amplification of approximately 1.5–2.0 °C in dense urban corridors under stable atmospheric conditions. Overall, the results demonstrate that urban heat intensification in Makkah emerges from the combined influence of impervious surface intensity, biophysical conditions, and road network configuration. These findings support climate-sensitive transport and urban planning strategies—such as compactness-aware street design, road-type zoning, and integration of green–blue infrastructure aligned with Saudi Vision 2030 and Sustainable Development Goals 11 and 13.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 1106-1131"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148529043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MDCT: multi-depth convolutional transformer for polarimetric SAR image classification","authors":"Liang Huang,&nbsp;Xin Ding,&nbsp;You Yang","doi":"10.1016/j.asr.2026.04.080","DOIUrl":"10.1016/j.asr.2026.04.080","url":null,"abstract":"<div><div>Polarimetric Synthetic Aperture Radar (PolSAR) image classification is a critical task in its applications. Although the advancement of deep learning techniques has significantly improved image classification, the complex-valued nature of PolSAR data poses challenges for real-number-based deep learning networks. Specifically, the complex input is difficult to integrate with mainstream deep learning frameworks that primarily operate on real numbers, while the complexity of the data also complicates preprocessing and network training. To address this issue, we propose a Multi-Depth Convolutional Transformer (MDCT), which transforms complex-valued inputs into real-valued representations through concatenation. This approach simplifies the data preprocessing pipeline and enhances the compatibility of deep learning networks with PolSAR data. In response to the increased feature dimensionality caused by the real-number transformation, the MDCT introduces an optimized multi-depth CNN architecture that integrates deep, shallow, and medium CNN layers. This design fully exploits the complementary advantages of input features, significantly improving the compactness and discriminability of feature representation. Furthermore, the MDCT combines a Transformer module to capture global relationships while leveraging a multi-scale feature extraction mechanism to obtain rich detail information from PolSAR images, effectively preserving key characteristics of the original data. Experimental results demonstrate that the MDCT achieves classification accuracies of 99.58%, 98.62%, and 98.52% on the Flevoland, San Francisco, and Oberpfaffenhofen datasets, respectively, showcasing superior performance compared to several traditional models.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 792-808"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148557720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Meteorological effects on the Istok riometer cosmic noise measurements","authors":"Alexey Oinats,&nbsp;Vladimir Ivonin,&nbsp;Artem Setov,&nbsp;Nozomu Nishitani,&nbsp;Kazuo Shiokawa","doi":"10.1016/j.asr.2026.04.113","DOIUrl":"10.1016/j.asr.2026.04.113","url":null,"abstract":"<div><div>We analyzed cosmic radio noise data observed at the Istok station in the framework of the international project “Study of dynamical variation of Particles and Waves in the INner magnetosphere using Ground-based network observations” (PWING) for the period from October 2017 to February 2021. A quiet day curve (QDC) was constructed, and the observed daily-seasonal variations in the level of cosmic radio noise at a frequency of 30 MHz were compared with a sky noise model, taking into account the location of the station and characteristics of the riometer antenna pattern. We observed significant positive deviations in the measured riometer voltage during winter periods that cannot be attributed to standard space weather effects, including solar flares. Comparison of the hourly and daily mean riometer voltage with meteorological ERA5-Land reanalysis data for the Istok station location revealed a significant anti-correlation with air temperature. Snow depth calculated using the INESOSIM model (Improved NASA Eulerian Snow on Sea Ice Model) and ERA5-Land reanalysis data also correlates well with hourly and daily mean riometer voltage. Based on the maximum correlation condition, probable variations in the snow cover depth at the Istok station were determined. Calculations performed in the GAL-ANA program showed that snow cover should lead to an increase in the gain of the riometer antenna compared to the value corresponding to wet soil. Multivariate regression fitting allowed evaluating the maximum additive voltage due to temperature and snow depth variations of about 0.7 V which might reduce/increase measured ionospheric absorption by 4 dB. The correlation coefficient between the daily mean voltage and meteorological variations is 0.75 for the entire period of observation. The combined meteorological contribution to the overall daily mean voltage variability is about 56%.</div></div>","PeriodicalId":50850,"journal":{"name":"Advances in Space Research","volume":"78 2","pages":"Pages 1540-1559"},"PeriodicalIF":3.2,"publicationDate":"2026-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148528943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}