Journal of Atmospheric and Solar-Terrestrial Physics最新文献_第6页

Research on hourly precipitation prediction along railways based on ERA5 reanalysis and post-processing correction 基于ERA5再分析及后处理校正的铁路沿线逐时降水预报研究

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-25 DOI: 10.1016/j.jastp.2025.106666

Xiangshun Meng , Yong Wang , Yunlong Zhang , Chengwu Yang , Chen Chang , Haozhe Chi , Yanping Liu

Global climate change has intensified extreme precipitation events, highlighting the urgent need for high-precision short-term rainfall forecasts to ensure railway transportation safety. However, existing meteorological monitoring remains limited by sparse station distribution, observational blind spots, and data inaccuracies. Global reanalysis datasets are hindered by low spatial resolution and precipitation underestimation, while numerical weather prediction models, typically with spatial resolutions exceeding 10 km, cannot satisfy the kilometer-scale disaster prevention demands along railway corridors. To address these limitations, we propose an “FFT–LSTM + post-processing correction” framework, which combines Fast Fourier Transform (FFT) and Long Short-Term Memory (LSTM) networks to extract nonlinear temporal characteristics of precipitation evolution from multivariate meteorological variables. The model further refines precipitation predictions through post-processing correction methods, including Simple Linear Regression (SLR), enhanced Piecewise Linear (PL), and Quantile Mapping (QM). FFT is initially employed to identify the best common period (143 h) among the inputs, guiding the optimal LSTM input window length. Subsequently, tailored correction strategies are applied according to rainfall intensity levels to improve prediction accuracy. Validation based on Meiyu-season data from four representative stations along the Guangzhou–Zhanjiang railway confirms that the proposed approach significantly enhances prediction skill. In hourly predictions, the Probability of Detection (POD) for moderate, heavy, and torrential rainfall reaches 0.562, 0.625, and 0.500, respectively; the Critical Success Index (CSI) for torrential rainfall peaks at 1.0, and the False Alarm Rate (FAR) is reduced to 0.000—indicating substantial gains over baseline models such as ARIMA and XGBoost (CSI <0.08). This study effectively integrates deep learning and statistical correction techniques to overcome key limitations of reanalysis data, providing high-precision support for short-term precipitation forecasting along railways and thereby supporting meteorological disaster mitigation and transportation safety decision-making.

全球气候变化加剧了极端降水事件，凸显了对高精度短期降水预报的迫切需求，以确保铁路运输安全。然而，现有的气象监测仍然受到站点分布稀疏、观测盲点和数据不准确的限制。全球再分析数据集受到低空间分辨率和降水低估的阻碍，而空间分辨率通常超过10 km的数值天气预报模式无法满足铁路走廊千米尺度的防灾需求。为了解决这些问题，我们提出了一个“FFT - LSTM +后处理校正”框架，该框架结合了快速傅里叶变换（FFT）和长短期记忆（LSTM）网络，从多元气象变量中提取降水演变的非线性时间特征。该模型通过简单线性回归（SLR）、增强分段线性（PL）和分位映射（QM）等后处理校正方法进一步细化降水预测。最初采用FFT识别输入间的最佳共同周期（143 h），指导LSTM输入窗口的最佳长度。随后，根据降雨强度等级，采用有针对性的校正策略，提高预报精度。基于广湛铁路沿线4个代表性站点梅雨季节数据的验证表明，该方法显著提高了预测能力。逐时预报中、强、暴雨的探测概率（POD）分别为0.562、0.625和0.500；暴雨的关键成功指数（CSI）达到峰值1.0，虚警率（FAR）降至0.000，这表明与ARIMA和XGBoost （CSI <0.08）等基线模型相比有了实质性的进步。本研究有效结合深度学习和统计校正技术，克服再分析数据的关键局限性，为铁路沿线短期降水预报提供高精度支持，从而为气象减灾和交通安全决策提供支持。

{"title":"Research on hourly precipitation prediction along railways based on ERA5 reanalysis and post-processing correction","authors":"Xiangshun Meng , Yong Wang , Yunlong Zhang , Chengwu Yang , Chen Chang , Haozhe Chi , Yanping Liu","doi":"10.1016/j.jastp.2025.106666","DOIUrl":"10.1016/j.jastp.2025.106666","url":null,"abstract":"<div><div>Global climate change has intensified extreme precipitation events, highlighting the urgent need for high-precision short-term rainfall forecasts to ensure railway transportation safety. However, existing meteorological monitoring remains limited by sparse station distribution, observational blind spots, and data inaccuracies. Global reanalysis datasets are hindered by low spatial resolution and precipitation underestimation, while numerical weather prediction models, typically with spatial resolutions exceeding 10 km, cannot satisfy the kilometer-scale disaster prevention demands along railway corridors. To address these limitations, we propose an “FFT–LSTM + post-processing correction” framework, which combines Fast Fourier Transform (FFT) and Long Short-Term Memory (LSTM) networks to extract nonlinear temporal characteristics of precipitation evolution from multivariate meteorological variables. The model further refines precipitation predictions through post-processing correction methods, including Simple Linear Regression (SLR), enhanced Piecewise Linear (PL), and Quantile Mapping (QM). FFT is initially employed to identify the best common period (143 h) among the inputs, guiding the optimal LSTM input window length. Subsequently, tailored correction strategies are applied according to rainfall intensity levels to improve prediction accuracy. Validation based on Meiyu-season data from four representative stations along the Guangzhou–Zhanjiang railway confirms that the proposed approach significantly enhances prediction skill. In hourly predictions, the Probability of Detection (POD) for moderate, heavy, and torrential rainfall reaches 0.562, 0.625, and 0.500, respectively; the Critical Success Index (CSI) for torrential rainfall peaks at 1.0, and the False Alarm Rate (FAR) is reduced to 0.000—indicating substantial gains over baseline models such as ARIMA and XGBoost (CSI <0.08). This study effectively integrates deep learning and statistical correction techniques to overcome key limitations of reanalysis data, providing high-precision support for short-term precipitation forecasting along railways and thereby supporting meteorological disaster mitigation and transportation safety decision-making.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106666"},"PeriodicalIF":1.9,"publicationDate":"2025-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145417622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spectral analyses of short-to medium-term gas cycles in Riyadh: Environmental and cosmic drivers 利雅得中短期气体循环的光谱分析：环境和宇宙驱动因素

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-25 DOI: 10.1016/j.jastp.2025.106667

Maghrabi A

This study employs Lomb periodogram analysis to investigate the spectral characteristics of atmospheric gases (NH₃, O₃, NOx, NO₁, NO₂, NMHC), solar radio flux (F10.7), and cosmic rays (CR) in Riyadh from 1999 to 2007. Significant periodicities ranging from 10 days to 1.6 years were identified, with prominent cycles for gases at 454–584 days, 291–293 days, 215 days, 155–171 days, 113 days, 97 days, 39 days, 15–17 days, and 10 days; for F10.7 at 467 days, 255 days, 172 days, and 17 days; and for CR at 489 days, 273 days, 140 days, and 32 days. The 215-day semi-annual and 39-day synoptic cycles, prevalent across multiple gases, alongside the 155–171-day cycle strongly linked to F10.7 and CR, highlight robust seasonal, synoptic, and heliospheric influences. Cross-spectral analyses, along with zero-order and partial correlation analyses, were conducted and confirmed that, while variations in atmospheric gases are subject to terrestrial factors such as meteorological conditions (temperature, humidity, wind), extraterrestrial factors, including solar activity and CR, play a significant role in their variations, with common periodicities validating these influences. Solar activity enhances NOx and NMHC photochemistry, while CR ionization significantly affects O₃ and reduces NOx. These extraterrestrial impacts, which warrant further investigation, are critically modulated by meteorological factors. These findings are vital for addressing Riyadh's environmental challenges, supporting sustainable urban development, and enhancing understanding of extraterrestrial influences on climate, atmospheric processes, and environmental sciences.

利用Lomb周期图分析了1999 ~ 2007年利雅得地区大气气体（NH3、O3、NOx、NO1、NO2、NMHC）、太阳射电通量（F10.7）和宇宙射线（CR）的光谱特征。在10 ~ 1.6年的周期范围内，气体的显著周期为454 ~ 584天、291 ~ 293天、215天、155 ~ 171天、113天、97天、39天、15 ~ 17天和10天；F10.7分别为467天、255天、172天、17天；CR为489天、273天、140天、32天。在多种气体中普遍存在的半年周期215天和天气周期39天，以及与F10.7和CR密切相关的155 - 171天周期，突出了强劲的季节、天气和日球层影响。进行了交叉光谱分析以及零阶和部分相关分析，并证实，虽然大气气体的变化受气象条件（温度、湿度、风）等地球因素的影响，但包括太阳活动和CR在内的地外因素在其变化中起着重要作用，其共同的周期性证实了这些影响。太阳活动增强NOx和NMHC光化学反应，而CR电离作用显著影响O3并降低NOx。这些值得进一步调查的地外撞击受到气象因素的严重调节。这些发现对于应对利雅得的环境挑战、支持可持续城市发展以及增进对气候、大气过程和环境科学的地外影响的理解至关重要。

{"title":"Spectral analyses of short-to medium-term gas cycles in Riyadh: Environmental and cosmic drivers","authors":"Maghrabi A","doi":"10.1016/j.jastp.2025.106667","DOIUrl":"10.1016/j.jastp.2025.106667","url":null,"abstract":"<div><div>This study employs Lomb periodogram analysis to investigate the spectral characteristics of atmospheric gases (NH<sub>3</sub>, O<sub>3</sub>, NOx, NO<sub>1</sub>, NO<sub>2</sub>, NMHC), solar radio flux (F10.7), and cosmic rays (CR) in Riyadh from 1999 to 2007. Significant periodicities ranging from 10 days to 1.6 years were identified, with prominent cycles for gases at 454–584 days, 291–293 days, 215 days, 155–171 days, 113 days, 97 days, 39 days, 15–17 days, and 10 days; for F10.7 at 467 days, 255 days, 172 days, and 17 days; and for CR at 489 days, 273 days, 140 days, and 32 days. The 215-day semi-annual and 39-day synoptic cycles, prevalent across multiple gases, alongside the 155–171-day cycle strongly linked to F10.7 and CR, highlight robust seasonal, synoptic, and heliospheric influences. Cross-spectral analyses, along with zero-order and partial correlation analyses, were conducted and confirmed that, while variations in atmospheric gases are subject to terrestrial factors such as meteorological conditions (temperature, humidity, wind), extraterrestrial factors, including solar activity and CR, play a significant role in their variations, with common periodicities validating these influences. Solar activity enhances NOx and NMHC photochemistry, while CR ionization significantly affects O<sub>3</sub> and reduces NOx. These extraterrestrial impacts, which warrant further investigation, are critically modulated by meteorological factors. These findings are vital for addressing Riyadh's environmental challenges, supporting sustainable urban development, and enhancing understanding of extraterrestrial influences on climate, atmospheric processes, and environmental sciences.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106667"},"PeriodicalIF":1.9,"publicationDate":"2025-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145417618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An analysis of the impacts of meteorological factors on ozone concentration using generalized additive model in Tianjin, China 利用广义加性模式分析气象因子对天津地区臭氧浓度的影响

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-24 DOI: 10.1016/j.jastp.2025.106669

Xu Zhang , Chang Liu , Yixin Liu , Xumei Yuan

Ozone (O₃) pollution in China is an increasingly serious problem and Tianjin experiences O₃ pollution as its predominant air quality challenge. An analysis of data from 2018 to 2022 indicated that the level of O₃ pollution in Tianjin showed a trend of first decreasing and then increasing. The monthly distribution of O₃ concentration and the number of O₃ exceeding the standard day in Tianjin showed a single peak trend in June. Based on the ground meteorological data and air quality data in Tianjin from 2018 to 2022, the study developed a time-phased generalized additive model (GAM) of meteorological factors (temperature, relative humidity, sunshine duration, pressure, precipitation and wind speed) in Tianjin to analyze their specific effects on O₃ concentration. The results showed that the GAM had a high quality and effectively captured the complex nonlinear relationship between O₃ and meteorological factors. Seasonal differences were identified in the relationship between O₃ concentration and different meteorological factors in Tianjin. Notably, temperature was the dominant meteorological factor affecting O₃ concentration change in Tianjin. The interaction of high temperature and medium relative humidity was highly correlated with O₃ concentration in Tianjin in summer. The research results are helpful to clarify the influence of meteorological conditions in different seasons on O₃ concentration change in Tianjin. It is of great significance for the accurate prediction of O₃ pollution and the formulation of pollution prevention and control policies according to local conditions.

臭氧（O3）污染在中国日益严重，天津的主要空气质量挑战是O3污染。2018 - 2022年数据分析表明，天津市O3污染水平呈现先下降后上升的趋势。6月天津市O3浓度和O3超标日数的月分布呈单峰趋势。基于天津市2018 - 2022年地面气象资料和空气质量数据，建立了天津市气温、相对湿度、日照时数、气压、降水、风速等气象因子的时间阶段广义加性模型（GAM），分析其对O3浓度的具体影响。结果表明，GAM具有较高的质量，能有效地捕捉到O3与气象因子之间复杂的非线性关系。天津市不同气象因子对O3浓度的影响存在季节差异。气温是影响天津市O3浓度变化的主要气象因子。天津夏季高温和中等相对湿度的交互作用与O3浓度高度相关。研究结果有助于阐明不同季节气象条件对天津市O3浓度变化的影响。对准确预测O3污染，因地制宜制定污染防治政策具有重要意义。

{"title":"An analysis of the impacts of meteorological factors on ozone concentration using generalized additive model in Tianjin, China","authors":"Xu Zhang , Chang Liu , Yixin Liu , Xumei Yuan","doi":"10.1016/j.jastp.2025.106669","DOIUrl":"10.1016/j.jastp.2025.106669","url":null,"abstract":"<div><div>Ozone (O<sub>3</sub>) pollution in China is an increasingly serious problem and Tianjin experiences O<sub>3</sub> pollution as its predominant air quality challenge. An analysis of data from 2018 to 2022 indicated that the level of O<sub>3</sub> pollution in Tianjin showed a trend of first decreasing and then increasing. The monthly distribution of O<sub>3</sub> concentration and the number of O<sub>3</sub> exceeding the standard day in Tianjin showed a single peak trend in June. Based on the ground meteorological data and air quality data in Tianjin from 2018 to 2022, the study developed a time-phased generalized additive model (GAM) of meteorological factors (temperature, relative humidity, sunshine duration, pressure, precipitation and wind speed) in Tianjin to analyze their specific effects on O<sub>3</sub> concentration. The results showed that the GAM had a high quality and effectively captured the complex nonlinear relationship between O<sub>3</sub> and meteorological factors. Seasonal differences were identified in the relationship between O<sub>3</sub> concentration and different meteorological factors in Tianjin. Notably, temperature was the dominant meteorological factor affecting O<sub>3</sub> concentration change in Tianjin. The interaction of high temperature and medium relative humidity was highly correlated with O<sub>3</sub> concentration in Tianjin in summer. The research results are helpful to clarify the influence of meteorological conditions in different seasons on O<sub>3</sub> concentration change in Tianjin. It is of great significance for the accurate prediction of O<sub>3</sub> pollution and the formulation of pollution prevention and control policies according to local conditions.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106669"},"PeriodicalIF":1.9,"publicationDate":"2025-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145417621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Analysis of ozone pollution characteristics and meteorological factors in Yichang City, Hubei 湖北宜昌市臭氧污染特征及气象因子分析

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-24 DOI: 10.1016/j.jastp.2025.106668

Ting Zhou , Hui Hu , Pan Wang , Mi Zhang , Haoqi Wen , Dan Liu , Wei Liu

Grasping the local ozone pollution characteristics and associated meteorological factors is of great significance for scientific support for ozone (O₃) pollution prevention. O₃ concentrations and related meteorological data from 2020 to 2022 Yichang City, Hubei were analyzed for exploring the spatiotemporal distribution of O₃ concentrations and the impact of associated meteorological factors. The results revealed that O₃ pollution in Yichang from 2020 to 2022 exhibited a clear trend of the increasing pollution severity and frequency. High incidences of ozone pollution was between May and September (mainly in summer), but high concentrations of precursors (NO₂ and TVOCs) appeared from November to January (mainly in winter), highlighting the decisive role of meteorological conditions in the ozone formation. Moreover, the diurnal variation of O₃ concentrations displayed a typical single peak distribution pattern, with hourly O₃ concentration (ρ(O₃)) peaking at 15:00–16:00, suggesting that ρ(O₃) in Yichang was primarily influenced by local emissions. In addition, the sustained high ρ(O₃) on exceedance days were related to the local topography and ρ(O₃) along the river were strongly influenced by river-land breezes. The difference of meteorological factors between O₃ exceedance and non-exceedance days indicated that temperature and relative humidity had more pronounced impacts on ρ(O₃). Specifically, 89.6 % of ρ(O₃) exceedances occurred under meteorological conditions with temperature ≥25 °C and relative humidity between 25 % and 75 %, with wind speeds consistently below 5 m/s. It was recommended that environmental management departments should deploy related emission control measures in advance when the above meteorological conditions were forecasted, while fully considering the influence of the topography and river-land breeze on pollutant transports to optimize control strategies and coordinated prevention measures.

掌握局地臭氧污染特征及相关气象因子，对臭氧污染防治具有重要的科学支撑意义。以湖北省宜昌市2020 - 2022年O3浓度及相关气象资料为研究对象，探讨了宜昌市O3浓度的时空分布特征及相关气象因子的影响。结果表明：2020 - 2022年宜昌市O3污染呈现出明显的加重趋势和频次增加趋势；臭氧污染高发期为5 - 9月（以夏季为主），而前体NO2和TVOCs浓度较高出现在11 - 1月（以冬季为主），凸显了气象条件对臭氧形成的决定性作用。此外，O3浓度的日变化呈典型的单峰分布，每小时O3浓度（ρ(O3)）在15:00 ~ 16:00达到峰值，表明宜昌地区的ρ(O3)主要受局地排放的影响。此外，超限日ρ(O3)持续高值与局地地形有关，沿江ρ(O3)受河陆风影响较大。气象因子在O3超标日与非超标日之间的差异表明，温度和相对湿度对ρ(O3)的影响更为显著。其中，89.6%的ρ(O3)异常发生在温度≥25℃、相对湿度在25% ~ 75%之间、风速始终低于5 m/s的气象条件下。建议环境管理部门在预测到上述气象条件时，提前部署相关排放控制措施，同时充分考虑地形、河陆风对污染物运移的影响，优化控制策略，协调预防措施。

{"title":"Analysis of ozone pollution characteristics and meteorological factors in Yichang City, Hubei","authors":"Ting Zhou , Hui Hu , Pan Wang , Mi Zhang , Haoqi Wen , Dan Liu , Wei Liu","doi":"10.1016/j.jastp.2025.106668","DOIUrl":"10.1016/j.jastp.2025.106668","url":null,"abstract":"<div><div>Grasping the local ozone pollution characteristics and associated meteorological factors is of great significance for scientific support for ozone (O<sub>3</sub>) pollution prevention. O<sub>3</sub> concentrations and related meteorological data from 2020 to 2022 Yichang City, Hubei were analyzed for exploring the spatiotemporal distribution of O<sub>3</sub> concentrations and the impact of associated meteorological factors. The results revealed that O<sub>3</sub> pollution in Yichang from 2020 to 2022 exhibited a clear trend of the increasing pollution severity and frequency. High incidences of ozone pollution was between May and September (mainly in summer), but high concentrations of precursors (NO<sub>2</sub> and TVOCs) appeared from November to January (mainly in winter), highlighting the decisive role of meteorological conditions in the ozone formation. Moreover, the diurnal variation of O<sub>3</sub> concentrations displayed a typical single peak distribution pattern, with hourly O<sub>3</sub> concentration (ρ(O<sub>3</sub>)) peaking at 15:00–16:00, suggesting that ρ(O<sub>3</sub>) in Yichang was primarily influenced by local emissions. In addition, the sustained high ρ(O<sub>3</sub>) on exceedance days were related to the local topography and ρ(O<sub>3</sub>) along the river were strongly influenced by river-land breezes. The difference of meteorological factors between O<sub>3</sub> exceedance and non-exceedance days indicated that temperature and relative humidity had more pronounced impacts on ρ(O<sub>3</sub>). Specifically, 89.6 % of ρ(O<sub>3</sub>) exceedances occurred under meteorological conditions with temperature ≥25 °C and relative humidity between 25 % and 75 %, with wind speeds consistently below 5 m/s. It was recommended that environmental management departments should deploy related emission control measures in advance when the above meteorological conditions were forecasted, while fully considering the influence of the topography and river-land breeze on pollutant transports to optimize control strategies and coordinated prevention measures.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106668"},"PeriodicalIF":1.9,"publicationDate":"2025-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145516884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Scale-dependent coupling between galactic cosmic rays and trace gases revealed by multifractal analysis 多重分形分析揭示星系宇宙射线与微量气体之间的尺度依赖性耦合

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-21 DOI: 10.1016/j.jastp.2025.106661

D. Sierra-Porta

Galactic cosmic rays (GCR) modulate atmospheric ionisation and may influence reactive greenhouse gases, yet linear correlations have proved inconclusive. We analyse 74 328 hourly observations (2016–2024) of pressure-corrected neutron-monitor counts and co-located CH

_{4}

and O

_{3}

mixing ratios from the high-alpine Jungfraujoch station using Multifractal Detrended Fluctuation Analysis (MFDFA) and its bivariate extension (MFDCCA). Cosmic rays exhibit a narrow, quasi-monofractal singularity spectrum (

Δ α_{full} = 0.43

), consistent with heliospheric modulation as a single dominant driver, whereas O

_{3}

and CH

_{4}

display progressively broader spectra (

Δ α = 0.78

and 0.84). Cross-Hurst exponents exceed unity for small-to-moderate fluctuations, indicating super-persistent joint variability on 1–7 d (O

_{3}

) and 7–30 d (CH

_{4}

) horizons—time-scales compatible with HO

_{x}

/NO

_{x}

chemistry triggered by GCR ionisation. Quadratic fits to the cross-singularity spectra yield half-maximum widths of 0.39 (CH

_{4}

–GCR) and

< 0.33

(O

_{3}

–GCR), quantifying a broader amplitude hierarchy for methane. Extreme gas anomalies, by contrast, show weak cross persistence, implicating dynamical intrusions rather than ionisation. Multifractal metrics thus expose a scale-selective GCR imprint masked in Pearson and Spearman statistics and suggest that incorporating GCR flux as a multiscale covariate could improve sub-monthly O

_{3}

/CH

_{4}

predictions. The approach provides a transferable framework for disentangling cosmic-ray forcing from chemical and dynamical controls in other trace-gas records.

银河宇宙射线（GCR）调节大气电离并可能影响活性温室气体，但线性相关性已被证明是不确定的。利用多重分形去趋势波动分析（MFDFA）及其二元扩展（MFDCCA）方法，对2016-2024年高高山少女峰站压力校正中子监测计数和同地CH4和O3混合比的74328 h观测数据进行了分析。宇宙射线表现出狭窄的准单分形奇异光谱（Δαfull=0.43），与日球层调制作为单一主导驱动因素相一致，而O3和CH4则显示出逐渐扩大的光谱（Δα=0.78和0.84）。Cross-Hurst指数在小到中等波动中超过一，表明在1-7 d （O3）和7-30 d （CH4）水平时间尺度上的超持久联合变异与GCR电离引发的HOx/NOx化学反应相容。交叉奇异谱的二次拟合得到半最大宽度为0.39 （CH4-GCR）和0.33 (O3-GCR)，量化了甲烷更广泛的振幅层次。相比之下，极端气体异常表现出微弱的交叉持久性，暗示动力学侵入而不是电离。因此，多重分形指标揭示了皮尔逊和斯皮尔曼统计数据中掩盖的尺度选择性GCR印记，并表明将GCR通量作为多尺度协变量可以改善亚月O3/CH4预测。该方法为将宇宙射线强迫与其他痕量气体记录中的化学和动力学控制分离开来提供了一个可转移的框架。

{"title":"Scale-dependent coupling between galactic cosmic rays and trace gases revealed by multifractal analysis","authors":"D. Sierra-Porta","doi":"10.1016/j.jastp.2025.106661","DOIUrl":"10.1016/j.jastp.2025.106661","url":null,"abstract":"<div><div>Galactic cosmic rays (GCR) modulate atmospheric ionisation and may influence reactive greenhouse gases, yet linear correlations have proved inconclusive. We analyse 74328 hourly observations (2016–2024) of pressure-corrected neutron-monitor counts and co-located CH<span><math><msub><mrow></mrow><mrow><mn>4</mn></mrow></msub></math></span> and O<span><math><msub><mrow></mrow><mrow><mn>3</mn></mrow></msub></math></span> mixing ratios from the high-alpine Jungfraujoch station using Multifractal Detrended Fluctuation Analysis (MFDFA) and its bivariate extension (MFDCCA). Cosmic rays exhibit a narrow, quasi-monofractal singularity spectrum (<span><math><mrow><mi>Δ</mi><msub><mrow><mi>α</mi></mrow><mrow><mtext>full</mtext></mrow></msub><mo>=</mo><mn>0</mn><mo>.</mo><mn>43</mn></mrow></math></span>), consistent with heliospheric modulation as a single dominant driver, whereas O<span><math><msub><mrow></mrow><mrow><mn>3</mn></mrow></msub></math></span> and CH<span><math><msub><mrow></mrow><mrow><mn>4</mn></mrow></msub></math></span> display progressively broader spectra (<span><math><mrow><mi>Δ</mi><mi>α</mi><mo>=</mo><mn>0</mn><mo>.</mo><mn>78</mn></mrow></math></span> and 0.84). Cross-Hurst exponents exceed unity for small-to-moderate fluctuations, indicating super-persistent joint variability on 1–7 d (O<span><math><msub><mrow></mrow><mrow><mn>3</mn></mrow></msub></math></span>) and 7–30 d (CH<span><math><msub><mrow></mrow><mrow><mn>4</mn></mrow></msub></math></span>) horizons—time-scales compatible with HO<span><math><msub><mrow></mrow><mrow><mi>x</mi></mrow></msub></math></span>/NO<span><math><msub><mrow></mrow><mrow><mi>x</mi></mrow></msub></math></span> chemistry triggered by GCR ionisation. Quadratic fits to the cross-singularity spectra yield half-maximum widths of 0.39 (CH<span><math><msub><mrow></mrow><mrow><mn>4</mn></mrow></msub></math></span>–GCR) and <span><math><mrow><mo><</mo><mn>0</mn><mo>.</mo><mn>33</mn></mrow></math></span> (O<span><math><msub><mrow></mrow><mrow><mn>3</mn></mrow></msub></math></span>–GCR), quantifying a broader amplitude hierarchy for methane. Extreme gas anomalies, by contrast, show weak cross persistence, implicating dynamical intrusions rather than ionisation. Multifractal metrics thus expose a scale-selective GCR imprint masked in Pearson and Spearman statistics and suggest that incorporating GCR flux as a multiscale covariate could improve sub-monthly O<span><math><msub><mrow></mrow><mrow><mn>3</mn></mrow></msub></math></span>/CH<span><math><msub><mrow></mrow><mrow><mn>4</mn></mrow></msub></math></span> predictions. The approach provides a transferable framework for disentangling cosmic-ray forcing from chemical and dynamical controls in other trace-gas records.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106661"},"PeriodicalIF":1.9,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145359170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Directional characteristics of low-level jet over the Arabian Sea and its impact on monsoon precipitation over the Western Ghats 阿拉伯海低空急流的方向特征及其对西高止山脉季风降水的影响

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-19 DOI: 10.1016/j.jastp.2025.106665

Xinming Zhang , M.V. Subrahmanyam , Zhou Le , Dongxiao Wang

Previous studies have conducted a comprehensive examination of the impact of wind speed on monsoon precipitation; however, the significance of wind direction remains largely unexplored. This analysis utilizes ERA-5 reanalysis data to quantitatively investigate the directional characteristics of the Arabian Sea low-level jet (LLJ) over the last four decades. It applies wind angle metrics to evaluate the LLJ's direction and its influence on monsoon precipitation over the Western Ghats (WG), with a larger angle signifying that the LLJ is more aligned with the southerly wind. At 925 hPa, the LLJ exhibits a greater wind angle than at 850 hPa, showing an upward trend. This finding suggests a northerly movement of the LLJ, potentially associated with the Pacific Decadal Oscillation. Statistical analysis reveals that a larger LLJ angle is associated with heightened precipitation on the northern windward slopes of the WG, and precipitation decreases in the southern regions, suggesting the influence of the LLJ's directional changes on precipitation patterns. The results offer valuable perspectives for enhancing monsoon predictions and deepening our scientific comprehension of the relationship between the LLJ and precipitation.

以前的研究对风速对季风降水的影响进行了全面的研究；然而，风向的重要性在很大程度上仍未被探索。该分析利用ERA-5再分析数据定量研究了过去40年阿拉伯海低空急流（LLJ）的方向特征。应用风角度量来评价LLJ的方向及其对西高止山脉季风降水的影响，风角越大表明LLJ更偏向于南风。在925 hPa处，LLJ风角大于850 hPa，呈上升趋势。这一发现表明LLJ向北移动，可能与太平洋年代际振荡有关。统计分析表明，LLJ角的增大与北风坡降水增加有关，而南风坡降水减少有关，表明LLJ的方向变化对降水模式有影响。这些结果为加强季风预测和加深我们对LLJ与降水关系的科学理解提供了有价值的观点。

{"title":"Directional characteristics of low-level jet over the Arabian Sea and its impact on monsoon precipitation over the Western Ghats","authors":"Xinming Zhang , M.V. Subrahmanyam , Zhou Le , Dongxiao Wang","doi":"10.1016/j.jastp.2025.106665","DOIUrl":"10.1016/j.jastp.2025.106665","url":null,"abstract":"<div><div>Previous studies have conducted a comprehensive examination of the impact of wind speed on monsoon precipitation; however, the significance of wind direction remains largely unexplored. This analysis utilizes ERA-5 reanalysis data to quantitatively investigate the directional characteristics of the Arabian Sea low-level jet (LLJ) over the last four decades. It applies wind angle metrics to evaluate the LLJ's direction and its influence on monsoon precipitation over the Western Ghats (WG), with a larger angle signifying that the LLJ is more aligned with the southerly wind. At 925 hPa, the LLJ exhibits a greater wind angle than at 850 hPa, showing an upward trend. This finding suggests a northerly movement of the LLJ, potentially associated with the Pacific Decadal Oscillation. Statistical analysis reveals that a larger LLJ angle is associated with heightened precipitation on the northern windward slopes of the WG, and precipitation decreases in the southern regions, suggesting the influence of the LLJ's directional changes on precipitation patterns. The results offer valuable perspectives for enhancing monsoon predictions and deepening our scientific comprehension of the relationship between the LLJ and precipitation.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106665"},"PeriodicalIF":1.9,"publicationDate":"2025-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145359095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Centennial solar EUV irradiance from ionospheric currents: Varying sunspot-EUV irradiance relation and modified spot-facula ratio 电离层电流的百年太阳EUV辐照度：变化的太阳黑子-EUV辐照度关系和修正的黑子-光斑比

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-17 DOI: 10.1016/j.jastp.2025.106653

K. Mursula

Sunspots offer a uniquely long view of solar magnetic activity, and depict large variability during the last 100 years, a period known as the Modern Maximum (MM). However, if our view of solar variability was only based on the strongest magnetic fields, it would be incomplete. Therefore, other variables are needed to study the long-term evolution, e.g., of weaker magnetic fields and the different radiative emissions. Recently, the relation between sunspots and several other solar activity proxies like the F10.7 and F30 radio fluxes and the MgII index (all proxies of EUV irradiance) was found to vary during the last 70 years so that a relative sunspot dominance over EUV in the 1950s–1960s changed to EUV dominance in the 2000s (Mursula et al., 2024). Here we use data from eight long-operating observatories to calculate the yearly range of daily variation of the geomagnetic Y-component, the rY index, for the last 137 years. We show that the rY index correlates extremely well with the MgII index and the solar F30 radio flux. These three indices have no trend relative to each other over the respective intervals. On the other hand, the F10.7 flux has a significant trend with respect to the three co-varying EUV indices (MgII, F30, rY). Therefore, the rY index replaces F10.7 as the best long-term EUV proxy, and extends the MgII index by 90 years. We verify that all the four EUV proxies (rY, MgII, F30, F10.7) have an increasing trend with respect to sunspots during the last 50–70 years. This is valid both for sunspot numbers and group numbers. Extending this earlier with the rY index, we find that the relation between EUV irradiance and sunspots has a nonlinear, quadratic evolution over the MM. This implies that the Sun has more sunspots relative to EUV irradiance during the growth and maximum of the MM, while the opposite is true during its decay and minimum. We estimate that the MgII index (solar EUV irradiance) increases by 24% of its solar cycle variation with respect to the sunspot number during the last 70 years. Our results indicate a systematic difference in the evolution between sunspots (generally: photosphere) and plages (generally: chromosphere) with long-term solar activity. The implied varying spot-facula ratio has consequences to the stellar evolution of the Sun and Sun-like stars.

太阳黑子提供了一个独特的太阳磁活动的长期视图，并描绘了过去100年的大变化，这一时期被称为现代极大期（MM）。然而，如果我们对太阳变化的看法仅仅基于最强的磁场，那将是不完整的。因此，需要其他变量来研究长期演变，例如，较弱的磁场和不同的辐射发射。最近，太阳黑子与其他几个太阳活动指标(如F10.7和F30射电通量以及MgII指数（所有EUV辐照度的指标）之间的关系在过去70年中发生了变化，因此，20世纪50年代至60年代太阳黑子对EUV的相对优势转变为21世纪初的EUV优势（Mursula et al., 2024）。在这里，我们使用八个长期运行的观测站的数据来计算过去137年来地磁y分量（y指数）的日变化的年范围。我们发现rY指数与MgII指数和太阳F30射电通量的相关性非常好。这三个指数在各自的区间内没有相互的趋势。另一方面，F10.7通量相对于三个共变EUV指数（MgII、F30、rY）有显著的变化趋势。因此，rY指数取代F10.7成为最佳的长期EUV指标，并将MgII指数延长90年。我们验证了在过去50-70年间，所有四个EUV代理（rY, MgII, F30, F10.7）在太阳黑子方面都有增加的趋势。这对太阳黑子数和群数都是有效的。用rY指数扩展这一点，我们发现EUV辐照度和太阳黑子之间的关系在MM上具有非线性的二次演化。这意味着在MM的增长和最大值期间，太阳相对于EUV辐照度有更多的太阳黑子，而在其衰减和最小值期间则相反。我们估计，在过去70年中，MgII指数（太阳EUV辐照度）相对于太阳黑子数的太阳周期变化增加了24%。我们的研究结果表明，太阳黑子（通常是光球层）和太阳黑子（通常是色球层）在长期太阳活动中的演化存在系统差异。暗变的斑光斑比对太阳和类太阳恒星的演化有影响。

{"title":"Centennial solar EUV irradiance from ionospheric currents: Varying sunspot-EUV irradiance relation and modified spot-facula ratio","authors":"K. Mursula","doi":"10.1016/j.jastp.2025.106653","DOIUrl":"10.1016/j.jastp.2025.106653","url":null,"abstract":"<div><div>Sunspots offer a uniquely long view of solar magnetic activity, and depict large variability during the last 100 years, a period known as the Modern Maximum (MM). However, if our view of solar variability was only based on the strongest magnetic fields, it would be incomplete. Therefore, other variables are needed to study the long-term evolution, e.g., of weaker magnetic fields and the different radiative emissions. Recently, the relation between sunspots and several other solar activity proxies like the F10.7 and F30 radio fluxes and the MgII index (all proxies of EUV irradiance) was found to vary during the last 70 years so that a relative sunspot dominance over EUV in the 1950s–1960s changed to EUV dominance in the 2000s (Mursula et al., 2024). Here we use data from eight long-operating observatories to calculate the yearly range of daily variation of the geomagnetic Y-component, the rY index, for the last 137 years. We show that the rY index correlates extremely well with the MgII index and the solar F30 radio flux. These three indices have no trend relative to each other over the respective intervals. On the other hand, the F10.7 flux has a significant trend with respect to the three co-varying EUV indices (MgII, F30, rY). Therefore, the rY index replaces F10.7 as the best long-term EUV proxy, and extends the MgII index by 90 years. We verify that all the four EUV proxies (rY, MgII, F30, F10.7) have an increasing trend with respect to sunspots during the last 50–70 years. This is valid both for sunspot numbers and group numbers. Extending this earlier with the rY index, we find that the relation between EUV irradiance and sunspots has a nonlinear, quadratic evolution over the MM. This implies that the Sun has more sunspots relative to EUV irradiance during the growth and maximum of the MM, while the opposite is true during its decay and minimum. We estimate that the MgII index (solar EUV irradiance) increases by 24% of its solar cycle variation with respect to the sunspot number during the last 70 years. Our results indicate a systematic difference in the evolution between sunspots (generally: photosphere) and plages (generally: chromosphere) with long-term solar activity. The implied varying spot-facula ratio has consequences to the stellar evolution of the Sun and Sun-like stars.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106653"},"PeriodicalIF":1.9,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145359171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Supervised land cover classification of the Indian Sundarbans mangrove forest using random forest and Landsat imagery 利用随机森林和Landsat图像对印度孙德尔本斯红树林进行监督土地覆盖分类

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-17 DOI: 10.1016/j.jastp.2025.106663

M. Kasiselvanathan, P. Sridevi

Rapid population enhancement and economic development have significantly changed land use and land cover (LULC) in ecologically sensitive Sundarbans, the world's most fragile ecosystem. Accurate classification of LULC is important for environmental monitoring, resource management, and policy-making. In the present paper, the authors have employed a machine learning (ML) technique, Random Forest (RF), to classify land cover changes in the Sundarbans over two time periods: 2003, and 2023. Landsat-7 and Landsat-8 satellite imagery have been used for land cover (LC) analysis, followed by data pre-processing techniques including layer stacking, mosaicking, and spectral signature collection. Around 640 and 840 training samples have been used for classification, for the year 2003 and 2023 respectively with 30 % samples used for validation. The RF model utilized 500 trees (n-tree) and four variables at each split (m-try). The accuracy has been assessed with confusion matrices, and Kappa statistics showing an accuracy of 0.84 and 0.85 for the year 2003 and 2023 respectively. The result reveals an increase in sparse forest, open land, and sand bar, and decline in water bodies and dense forest. The reason may be due to anthropogenic activities and climate-induced factors such as rising sea levels and storm surges. The findings indicate Sundarbans ecosystem's susceptibility and the necessity of sustainable management techniques. This study indicates that future research should focus on longer time-series analysis, improved scalability, and cloud platform integration to improve. The research demonstrates that ML-based land cover classification can be a useful monitoring tool with high accuracy for environmental conservation.

孙德尔本斯是世界上最脆弱的生态系统，其土地利用和土地覆盖发生了显著变化。土地利用价值的准确分类对环境监测、资源管理和政策制定具有重要意义。在本文中，作者采用了一种机器学习（ML）技术——随机森林（RF），对孙德尔本斯地区2003年和2023年两个时间段的土地覆盖变化进行了分类。Landsat-7和Landsat-8卫星图像用于土地覆盖（LC）分析，然后进行数据预处理技术，包括层堆叠、马赛克和光谱特征收集。2003年和2023年分别使用了大约640和840个训练样本进行分类，其中30%的样本用于验证。RF模型利用500棵树（n-tree）和4个变量在每次分割（m-try）。准确度用混淆矩阵进行了评估，Kappa统计显示2003年和2023年的准确度分别为0.84和0.85。结果表明：疏林、开阔地和沙坝增加，水体和茂密林减少；原因可能是人为活动和气候引起的因素，如海平面上升和风暴潮。研究结果表明孙德尔本斯生态系统的易感性和可持续管理技术的必要性。本研究表明，未来的研究应侧重于更长的时间序列分析、改进的可扩展性和云平台集成来改进。研究表明，基于机器学习的土地覆被分类可以作为一种有用的、高精度的环境保护监测工具。

{"title":"Supervised land cover classification of the Indian Sundarbans mangrove forest using random forest and Landsat imagery","authors":"M. Kasiselvanathan, P. Sridevi","doi":"10.1016/j.jastp.2025.106663","DOIUrl":"10.1016/j.jastp.2025.106663","url":null,"abstract":"<div><div>Rapid population enhancement and economic development have significantly changed land use and land cover (LULC) in ecologically sensitive Sundarbans, the world's most fragile ecosystem. Accurate classification of LULC is important for environmental monitoring, resource management, and policy-making. In the present paper, the authors have employed a machine learning (ML) technique, Random Forest (RF), to classify land cover changes in the Sundarbans over two time periods: 2003, and 2023. Landsat-7 and Landsat-8 satellite imagery have been used for land cover (LC) analysis, followed by data pre-processing techniques including layer stacking, mosaicking, and spectral signature collection. Around 640 and 840 training samples have been used for classification, for the year 2003 and 2023 respectively with 30 % samples used for validation. The RF model utilized 500 trees (n-tree) and four variables at each split (m-try). The accuracy has been assessed with confusion matrices, and Kappa statistics showing an accuracy of 0.84 and 0.85 for the year 2003 and 2023 respectively. The result reveals an increase in sparse forest, open land, and sand bar, and decline in water bodies and dense forest. The reason may be due to anthropogenic activities and climate-induced factors such as rising sea levels and storm surges. The findings indicate Sundarbans ecosystem's susceptibility and the necessity of sustainable management techniques. This study indicates that future research should focus on longer time-series analysis, improved scalability, and cloud platform integration to improve. The research demonstrates that ML-based land cover classification can be a useful monitoring tool with high accuracy for environmental conservation.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106663"},"PeriodicalIF":1.9,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Wind meets solar: crafting the perfect duck curve for future power systems 风能与太阳能的结合：为未来的电力系统打造完美的鸭子曲线

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-17 DOI: 10.1016/j.jastp.2025.106664

Yunxiao Chen, Jinfu Liu, Daren Yu

With the increasing penetration rate of wind and solar energy, the difference between load demand and renewable energy has gradually become the main regulating object (net load) of the power system. The daily net load curve is called the duck curve due to its shape. Duck curves pose new requirements for the overall flexibility of power system resources in terms of power output and power ramp. In order to reduce the cost and difficulty of balancing the duck curve in the power system, this paper proposes a method for constructing a friendly duck curve. Firstly, a friendliness evaluation framework consisting of 8 indicators is proposed to assess the quality of the duck curve. Secondly, through short-term optimization, long-term optimization and real-time adjustment for the wind-solar energy capacity layout of 26 grids in Belgium, the friendly duck curves are achieved. Meanwhile, four conventional wind-solar energy layouts are used as comparison groups. Finally, as the typical representative of flexibility resources, thermal power units are used to measure the net load and difficulty and cost of climbing the balance duck curves. The results indicate that the friendly duck curve requires a smaller scale of flexibility resources compared to conventional duck curves.

随着风能和太阳能渗透率的不断提高，负荷需求与可再生能源之间的差异逐渐成为电力系统的主要调节对象（净负荷）。日净负荷曲线因其形状而被称为鸭曲线。鸭曲线从功率输出和功率斜坡两个方面对电力系统资源的整体灵活性提出了新的要求。为了降低电力系统中平衡鸭子曲线的成本和难度，本文提出了一种构建友好鸭子曲线的方法。首先，提出了由8个指标组成的友好度评价框架，对鸭曲线的质量进行评价。其次，通过对比利时26个电网的风能-太阳能容量布局进行短期优化、长期优化和实时调整，实现友好鸭子曲线。同时，以四种传统的风能-太阳能布局作为对照组。最后，利用火电机组作为柔性资源的典型代表，测量了其净负荷和攀爬平衡曲线的难度和成本。结果表明，与传统鸭子曲线相比，友好鸭子曲线所需的柔性资源规模更小。

{"title":"Wind meets solar: crafting the perfect duck curve for future power systems","authors":"Yunxiao Chen, Jinfu Liu, Daren Yu","doi":"10.1016/j.jastp.2025.106664","DOIUrl":"10.1016/j.jastp.2025.106664","url":null,"abstract":"<div><div>With the increasing penetration rate of wind and solar energy, the difference between load demand and renewable energy has gradually become the main regulating object (net load) of the power system. The daily net load curve is called the duck curve due to its shape. Duck curves pose new requirements for the overall flexibility of power system resources in terms of power output and power ramp. In order to reduce the cost and difficulty of balancing the duck curve in the power system, this paper proposes a method for constructing a friendly duck curve. Firstly, a friendliness evaluation framework consisting of 8 indicators is proposed to assess the quality of the duck curve. Secondly, through short-term optimization, long-term optimization and real-time adjustment for the wind-solar energy capacity layout of 26 grids in Belgium, the friendly duck curves are achieved. Meanwhile, four conventional wind-solar energy layouts are used as comparison groups. Finally, as the typical representative of flexibility resources, thermal power units are used to measure the net load and difficulty and cost of climbing the balance duck curves. The results indicate that the friendly duck curve requires a smaller scale of flexibility resources compared to conventional duck curves.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106664"},"PeriodicalIF":1.9,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145359096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Performance analysis of machine learning techniques and spectral indices of determination water surface areas using Sentinel-2B satellite imagery 利用Sentinel-2B卫星图像确定水面面积的机器学习技术和光谱指标的性能分析

IF 1.9 4区地球科学 Q3 GEOCHEMISTRY & GEOPHYSICS

Journal of Atmospheric and Solar-Terrestrial Physics

Pub Date : 2025-10-17 DOI: 10.1016/j.jastp.2025.106662

Sevim Yasemin Oturanc

With recent advances in satellite technologies, remote sensing techniques have become more frequently used in monitoring water surfaces. The Catalan Reservoir, located on the Seyhan River, is an important source of drinking water, agricultural irrigation, and electricity generation for the city of Adana in southern Türkiye. Therefore, continuous monitoring of the water surface is necessary. In this study, Sentinel-2B MSI images dated December 24, 2023 were used to determine the reservoir water surface area using three spectral indices: Normalized Difference Water Index (NDWI), Water Ratio Index (WRI), and Modified Normalized Difference Water Index (MNDWI), and two supervised pixel-based classification techniques: Support Vector Machines (SVM) and Artificial Neural Networks (ANN). The reservoir water surface area was determined to be 69.3 km² with 98.6 % accuracy using NDWI, 67.7 km² with 98.3 % accuracy using WRI, and 69.9 km² with 97.6 % accuracy using MNDWI. Supervised classifications provided 68.3 km² with 98.6 % accuracy for SVM and 66.7 km² with 99 % accuracy for ANN. These findings demonstrate that spectral indices perform similarly to classification methods and can be used as practical alternatives. The novelty of this study is that the highest-accuracy ANN pixel-based classification algorithm is used for the first time in conjunction with WRI. This integration provides more accurate results for remote sensing of water resources.

随着最近卫星技术的进步，遥感技术已越来越频繁地用于监测水面。位于塞汉河上的加泰罗尼亚水库是基耶省南部阿达纳市重要的饮用水、农业灌溉和发电来源。因此，对水面进行连续监测是必要的。利用2023年12月24日的Sentinel-2B MSI图像，利用归一化差水指数（NDWI）、水比指数（WRI）和修正归一化差水指数（MNDWI） 3个光谱指标，以及支持向量机（SVM）和人工神经网络（ANN）两种基于监督像素的分类技术，确定了水库水面面积。利用NDWI、WRI和MNDWI分别确定了水库水面面积为69.3 km2、98.6%、67.7 km2和97.6%。监督分类SVM提供68.3 km2，准确率为98.6%；ANN提供66.7 km2，准确率为99%。这些发现表明，光谱指数的表现与分类方法相似，可以作为实用的替代方法。本研究的新颖之处在于，首次将最高精度的基于像素的人工神经网络分类算法与WRI结合使用。这种整合为水资源遥感提供了更准确的结果。

{"title":"Performance analysis of machine learning techniques and spectral indices of determination water surface areas using Sentinel-2B satellite imagery","authors":"Sevim Yasemin Oturanc","doi":"10.1016/j.jastp.2025.106662","DOIUrl":"10.1016/j.jastp.2025.106662","url":null,"abstract":"<div><div>With recent advances in satellite technologies, remote sensing techniques have become more frequently used in monitoring water surfaces. The Catalan Reservoir, located on the Seyhan River, is an important source of drinking water, agricultural irrigation, and electricity generation for the city of Adana in southern Türkiye. Therefore, continuous monitoring of the water surface is necessary. In this study, Sentinel-2B MSI images dated December 24, 2023 were used to determine the reservoir water surface area using three spectral indices: Normalized Difference Water Index (NDWI), Water Ratio Index (WRI), and Modified Normalized Difference Water Index (MNDWI), and two supervised pixel-based classification techniques: Support Vector Machines (SVM) and Artificial Neural Networks (ANN). The reservoir water surface area was determined to be 69.3 km<sup>2</sup> with 98.6 % accuracy using NDWI, 67.7 km<sup>2</sup> with 98.3 % accuracy using WRI, and 69.9 km<sup>2</sup> with 97.6 % accuracy using MNDWI. Supervised classifications provided 68.3 km<sup>2</sup> with 98.6 % accuracy for SVM and 66.7 km<sup>2</sup> with 99 % accuracy for ANN. These findings demonstrate that spectral indices perform similarly to classification methods and can be used as practical alternatives. The novelty of this study is that the highest-accuracy ANN pixel-based classification algorithm is used for the first time in conjunction with WRI. This integration provides more accurate results for remote sensing of water resources.</div></div>","PeriodicalId":15096,"journal":{"name":"Journal of Atmospheric and Solar-Terrestrial Physics","volume":"277 ","pages":"Article 106662"},"PeriodicalIF":1.9,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145359230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0