Pub Date : 2024-06-27DOI: 10.3103/s1068373924040095
A. E. Sumachev, L. S. Banshchikova, S. A. Griga
Abstract
The paper examines the implementation of neural network methods for predicting peak water levels during the period of spring ice drift by the example of the Sukhona, Northern Dvina, and Pechora rivers. All considered neural network methods have shown high efficiency according to the criteria recommended by the Hydrometcenter of Russia and surpassed regression dependencies in the skill of forecasts. When using the method of training artificial neural networks, the standard error of prediction is reduced by approximately 10–20% as compared with regression dependencies.
{"title":"Using the Methods of Neural Network Learning for Peak Water Level Prediction: A Case Study for the Rivers in the Dvina-Pechora Basin","authors":"A. E. Sumachev, L. S. Banshchikova, S. A. Griga","doi":"10.3103/s1068373924040095","DOIUrl":"https://doi.org/10.3103/s1068373924040095","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The paper examines the implementation of neural network methods for predicting peak water levels during the period of spring ice drift by the example of the Sukhona, Northern Dvina, and Pechora rivers. All considered neural network methods have shown high efficiency according to the criteria recommended by the Hydrometcenter of Russia and surpassed regression dependencies in the skill of forecasts. When using the method of training artificial neural networks, the standard error of prediction is reduced by approximately 10–20% as compared with regression dependencies.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"30 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523067","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040083
N. Yu. Zakhvatkina, I. A. Bychkova, V. G. Smirnov
Abstract
The paper describes an algorithm to differentiate leads from sea ice using the dual polarization synthetic aperture radar (SAR) data from the Sentinel-1 satellite in an extrawide swath mode. The algorithm uses the polarimetric features of the sea surface signal obtained in the SAR images: the ratio between co- and cross-polarization. A technique is proposed for classifying the SAR images to identify discontinuities (cracks, leads) in drifting sea ice using the ratio and difference of polarizations together with texture features and the neural network implementation. The method was tested using the satellite data obtained over the Arctic seas in the Russian Federation.
{"title":"Using the Neural Network Technique for Lead Detection in Radar Images of Arctic Sea Ice","authors":"N. Yu. Zakhvatkina, I. A. Bychkova, V. G. Smirnov","doi":"10.3103/s1068373924040083","DOIUrl":"https://doi.org/10.3103/s1068373924040083","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The paper describes an algorithm to differentiate leads from sea ice using the dual polarization synthetic aperture radar (SAR) data from the Sentinel-1 satellite in an extrawide swath mode. The algorithm uses the polarimetric features of the sea surface signal obtained in the SAR images: the ratio between co- and cross-polarization. A technique is proposed for classifying the SAR images to identify discontinuities (cracks, leads) in drifting sea ice using the ratio and difference of polarizations together with texture features and the neural network implementation. The method was tested using the satellite data obtained over the Arctic seas in the Russian Federation.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"206 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523065","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040113
A. M. Smetanina, S. A. Gromov, V. A. Obolkin, T. V. Khodzher, O. I. Khuriganova
Abstract
The machine learning model used to predict ozone concentrations at the Listvyanka monitoring station in the Baikal region is described. The model was trained and verified using automatic ground-based gas analyzer ozone measurements. Random forest and boosting machine learning models were used. According to the ERA5 reanalysis, the mean absolute error of ozone values exceeds 16 ppb, and the mean percentage error is 80%. The respective errors in the ozone values calculated using machine learning models are 6.7 ppb and 29%. The results of forecasting are the most sensitive to the season, air temperature, and vegetation. The ozone values for 2017–2022 were simulated and analyzed using the trained model and reanalysis data.
{"title":"Assessment of Atmospheric Ozone from Reanalysis and Ground-based Measurements in the Baikal Region","authors":"A. M. Smetanina, S. A. Gromov, V. A. Obolkin, T. V. Khodzher, O. I. Khuriganova","doi":"10.3103/s1068373924040113","DOIUrl":"https://doi.org/10.3103/s1068373924040113","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The machine learning model used to predict ozone concentrations at the Listvyanka monitoring station in the Baikal region is described. The model was trained and verified using automatic ground-based gas analyzer ozone measurements. Random forest and boosting machine learning models were used. According to the ERA5 reanalysis, the mean absolute error of ozone values exceeds 16 ppb, and the mean percentage error is 80%. The respective errors in the ozone values calculated using machine learning models are 6.7 ppb and 29%. The results of forecasting are the most sensitive to the season, air temperature, and vegetation. The ozone values for 2017–2022 were simulated and analyzed using the trained model and reanalysis data.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"128 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523069","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}
Pub Date : 2024-06-27DOI: 10.3103/s106837392404006x
M. O. Kuchma
Abstract
The paper presents some methods of satellite data preprocessing for the elimination of atmospheric effects on the electromagnetic radiation detected by the target equipment of a satellite and subsequent detection of floods in the Amur River basin. The atmospheric correction algorithm that has been used for the preprocessing is based on the use of a lookup table obtained by applying the Second Simulation of a Satellite Signal in the Solar Spectrum, which is a model of atmosphere radiative transfer. The subsequent flood detection in the Amur River basin water bodies builds on a neural network algorithm, the core of which is the upgraded U-Net. The developed algorithms for atmospheric correction and subsequent flood detection make it possible to receive information in an automatic near-real-time mode for monitoring flood conditions. Some groundwork has been made for applying the algorithm to the data of the Russian satellite instruments for spacecraft planned for launch.
{"title":"Application of Physical and Neural Network Methods in Operational Water Surface Detection","authors":"M. O. Kuchma","doi":"10.3103/s106837392404006x","DOIUrl":"https://doi.org/10.3103/s106837392404006x","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The paper presents some methods of satellite data preprocessing for the elimination of atmospheric effects on the electromagnetic radiation detected by the target equipment of a satellite and subsequent detection of floods in the Amur River basin. The atmospheric correction algorithm that has been used for the preprocessing is based on the use of a lookup table obtained by applying the Second Simulation of a Satellite Signal in the Solar Spectrum, which is a model of atmosphere radiative transfer. The subsequent flood detection in the Amur River basin water bodies builds on a neural network algorithm, the core of which is the upgraded U-Net. The developed algorithms for atmospheric correction and subsequent flood detection make it possible to receive information in an automatic near-real-time mode for monitoring flood conditions. Some groundwork has been made for applying the algorithm to the data of the Russian satellite instruments for spacecraft planned for launch.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"14 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523063","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040101
O. V. Kalmykova
Abstract
Every year about 50 waterspouts occur over the sea off the Black Sea coast of Russia. Over the past few years, the cases of waterspouts have occurred in the immediate vicinity of the coast with their subsequent destruction. The vortex destruction is often accompanied by short-term wind strengthening up to storm levels. The present study solves the problem of nowcasting the Black Sea waterspouts (building a detailed forecast of their formation for the next 2–6 hours) using machine learning methods. Learning by precedents is considered based on the labeled dataset of the radar characteristics of convective systems with and without waterspouts, models for classifying systems in terms of the risk of waterspout occurrence are constructed. The testing of the models showed that it is fundamentally possible to use them to diagnose systems with already formed waterspouts, as well as to identify the risk of waterspouts in advance (within two hours).
{"title":"Using Machine Learning Methods to Develop an Algorithm for Recognizing a Risk of Waterspout Occurrence off the Black Sea Coast of Russia","authors":"O. V. Kalmykova","doi":"10.3103/s1068373924040101","DOIUrl":"https://doi.org/10.3103/s1068373924040101","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Every year about 50 waterspouts occur over the sea off the Black Sea coast of Russia. Over the past few years, the cases of waterspouts have occurred in the immediate vicinity of the coast with their subsequent destruction. The vortex destruction is often accompanied by short-term wind strengthening up to storm levels. The present study solves the problem of nowcasting the Black Sea waterspouts (building a detailed forecast of their formation for the next 2–6 hours) using machine learning methods. Learning by precedents is considered based on the labeled dataset of the radar characteristics of convective systems with and without waterspouts, models for classifying systems in terms of the risk of waterspout occurrence are constructed. The testing of the models showed that it is fundamentally possible to use them to diagnose systems with already formed waterspouts, as well as to identify the risk of waterspouts in advance (within two hours).</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"2012 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523066","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040010
S. A. Soldatenko
Abstract
Artificial intelligence is one of the most popular, frequently discussed, and, meanwhile, ambiguous and controversial metaphorical concepts, which defines a scientific direction in computer science that studies the techniques for gaining knowledge, their computer representation, transformation, and application. Presently, it is intensively penetrating into many areas of human activities, including hydrometeorological ones. The concept of artificial intelligence, the history of its origin, and its methods and technologies are considered. The author analyzes the studies related to the use of artificial intelligence in short- and medium-range weather forecasting, including the collection and quality control of meteorological information, assimilation of data in order to generate initial conditions for numerical weather prediction models, development of forecast models and parameterization schemes for physical processes, postprocessing and physical-statistical interpretation of the output data of numerical weather prediction models.
{"title":"Artificial Intelligence and Its Application in Numerical Weather Prediction","authors":"S. A. Soldatenko","doi":"10.3103/s1068373924040010","DOIUrl":"https://doi.org/10.3103/s1068373924040010","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Artificial intelligence is one of the most popular, frequently discussed, and, meanwhile, ambiguous and controversial metaphorical concepts, which defines a scientific direction in computer science that studies the techniques for gaining knowledge, their computer representation, transformation, and application. Presently, it is intensively penetrating into many areas of human activities, including hydrometeorological ones. The concept of artificial intelligence, the history of its origin, and its methods and technologies are considered. The author analyzes the studies related to the use of artificial intelligence in short- and medium-range weather forecasting, including the collection and quality control of meteorological information, assimilation of data in order to generate initial conditions for numerical weather prediction models, development of forecast models and parameterization schemes for physical processes, postprocessing and physical-statistical interpretation of the output data of numerical weather prediction models.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"39 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141547223","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040058
V. V. Chursin, A. A. Kostornaya
Abstract
A method for the probabilistic identification of the precipitation and thunderstorm zones using artificial neural networks (ANNs), in particular, deep neural networks is described. The vertical profiles of temperature and humidity retrieved from satellite data are used as initial data. The ANN calculations have been validated using the ground-based observations in the Siberian region.
{"title":"Application of Deep Neural Networks for Detecting Probable Areas of Precipitation and Thunderstorms","authors":"V. V. Chursin, A. A. Kostornaya","doi":"10.3103/s1068373924040058","DOIUrl":"https://doi.org/10.3103/s1068373924040058","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>A method for the probabilistic identification of the precipitation and thunderstorm zones using artificial neural networks (ANNs), in particular, deep neural networks is described. The vertical profiles of temperature and humidity retrieved from satellite data are used as initial data. The ANN calculations have been validated using the ground-based observations in the Siberian region.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"39 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523062","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040022
V. V. Asmus, V. D. Bloshchinskiy, L. S. Kramareva, M. O. Kuchma, A. A. Filei
Abstract
The paper presents the research work aimed at improving the quality characteristics of information products based on the MSU-GS/VE radiometer aboard the Arktika-M No. 1 satellite, as well as at obtaining data preprocessing products. All described methods are based on using machine learning algorithms, namely, neural networks of various architectures. The results of developing a technology for minimizing the interference that occurs in the channels of the satellite device are provided. The work on detecting cloud formations based on processing the channel data in the visible and infrared ranges is presented. It is shown that the use of neural networks makes it possible to implement automatic algorithms for obtaining thematic products that take into account various factors and have an accuracy that is commensurate with statistical and physical approaches and reduces the time of satellite data processing.
{"title":"Preliminary Data Processing of the MSU-GS/VE Device aboard the Arktika-M No. 1 Highly Elliptical Satellite Using Machine Learning Methods","authors":"V. V. Asmus, V. D. Bloshchinskiy, L. S. Kramareva, M. O. Kuchma, A. A. Filei","doi":"10.3103/s1068373924040022","DOIUrl":"https://doi.org/10.3103/s1068373924040022","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The paper presents the research work aimed at improving the quality characteristics of information products based on the MSU-GS/VE radiometer aboard the Arktika-M No. 1 satellite, as well as at obtaining data preprocessing products. All described methods are based on using machine learning algorithms, namely, neural networks of various architectures. The results of developing a technology for minimizing the interference that occurs in the channels of the satellite device are provided. The work on detecting cloud formations based on processing the channel data in the visible and infrared ranges is presented. It is shown that the use of neural networks makes it possible to implement automatic algorithms for obtaining thematic products that take into account various factors and have an accuracy that is commensurate with statistical and physical approaches and reduces the time of satellite data processing.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"8 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141547181","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040046
K. G. Kortikova, I. A. Bychkova
Abstract
A method for detecting leads in the ice of the Arctic seas from satellite images of the visible range is presented. It is shown that sea ice leads are formed under the influence of dynamic processes in the ice cover, such as convergence, drift, and deformation of sea ice, as well as during the interaction of drifting ice with icebergs that have gone aground. The method for identifying sea ice leads is based on the use of artificial intelligence. To analyze the Landsat-8 satellite imagery, a convolutional neural network (U-Net architecture) was used. The method was tested using the satellite images of the visible spectral range that were obtained for the Laptev Sea. The results showed that the lead detection accuracy was above 80%. The method of the minimum rotated rectangle surrounding the polygon was used to determine the geometric parameters of the leads (length, width, inflection points).
{"title":"Application of Convolutional Neural Networks for Detecting Sea Ice Leads in the Laptev Sea with Landsat-8 Satellite Imagery","authors":"K. G. Kortikova, I. A. Bychkova","doi":"10.3103/s1068373924040046","DOIUrl":"https://doi.org/10.3103/s1068373924040046","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>A method for detecting leads in the ice of the Arctic seas from satellite images of the visible range is presented. It is shown that sea ice leads are formed under the influence of dynamic processes in the ice cover, such as convergence, drift, and deformation of sea ice, as well as during the interaction of drifting ice with icebergs that have gone aground. The method for identifying sea ice leads is based on the use of artificial intelligence. To analyze the Landsat-8 satellite imagery, a convolutional neural network (U-Net architecture) was used. The method was tested using the satellite images of the visible spectral range that were obtained for the Laptev Sea. The results showed that the lead detection accuracy was above 80%. The method of the minimum rotated rectangle surrounding the polygon was used to determine the geometric parameters of the leads (length, width, inflection points).</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"43 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523061","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}
Pub Date : 2024-06-27DOI: 10.3103/s1068373924040125
P. V. Kulizhskaya
Abstract
Fogs have a serious impact on human activity, in particular, on aviation, since they significantly impair visibility and therefore make aircraft landing difficult. In most cases, fogs cause irregularity of flights and sometimes lead to disasters, so timely and accurate forecasting of the onset of fog and its type is very important. At present, numerical methods greatly facilitate the forecasters’ work, but the problem of predicting visibility and fog remains relevant. Artificial intelligence technologies, in particular, deep learning algorithms using various kinds of neural networks are currently becoming more widespread in hydrometeorological activities. In the present study, the main objective is to develop a method for predicting the appearance of fog and to identify its type based on neural networks. The results of testing the method have showed its practical usefulness.
{"title":"A Method for Predicting Fog and Identifying Its Type Based on Neural Networks for the Saint Petersburg (Pulkovo) Airfield","authors":"P. V. Kulizhskaya","doi":"10.3103/s1068373924040125","DOIUrl":"https://doi.org/10.3103/s1068373924040125","url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>Fogs have a serious impact on human activity, in particular, on aviation, since they significantly impair visibility and therefore make aircraft landing difficult. In most cases, fogs cause irregularity of flights and sometimes lead to disasters, so timely and accurate forecasting of the onset of fog and its type is very important. At present, numerical methods greatly facilitate the forecasters’ work, but the problem of predicting visibility and fog remains relevant. Artificial intelligence technologies, in particular, deep learning algorithms using various kinds of neural networks are currently becoming more widespread in hydrometeorological activities. In the present study, the main objective is to develop a method for predicting the appearance of fog and to identify its type based on neural networks. The results of testing the method have showed its practical usefulness.</p>","PeriodicalId":49581,"journal":{"name":"Russian Meteorology and Hydrology","volume":"3 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141523068","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号