The article presents the results of several studies to assess the impact of the configuration of NUMA (Non-Uniform Memory Access) nodes on the performance of GPU-accelerated applications in hybrid computing system with shared memory. Using Crossroads/N9 DGEMM (NVBLAS library) as a model application, the performance in various NUMA modes with one or more GPUs was analyzed, and the throughput of the memory subsystem and data transfer channels between the host memory and graphics processors was also measured. The impact of coprocessor distribution across NUMA nodes on the efficiency of the model application was also examined.Results showed that configuration of NUMA nodes can have a significant impact on the performance of applications that offload calculations to graphics coprocessors in a hybrid computing system with shared memory, and this impact could have an effect in different ways. For example, using one NUMA node for the entire computing system is the least optimal approach in terms of memory bandwidth, but it provided the highest bandwidth for communication between host memory and coprocessors during active data transfer to several accelerators. Thus, this mode achieves maximum performance when performing calculations on multiple GPUs that actively exchange data through host memory. Other modes showed advantages in different situations. Overall, to achieve maximum performance during active data transfer to coprocessors, they should be part of one NUMA node. These results will help to develop approaches to configuration of hybrid computing systems on processors with a chiplet layout, and help to improve the performance of software that offloads calculations to graphics accelerators with the Ampere architecture, such as NVIDIA A800 and NVIDIA A100, which are currently widely represented in the high-performance computing industry.

Many natural and man-made phenomena cause ionospheric responses that can be captured in total electron content data obtained from Global Navigation Satellite System observations. However, the increasing volume of the data poses challenges for its analysis. This study provides an algorithm for detecting covolcanic ionospheric disturbances in total electron content time series based on machine learning. Using the Sarychev Peak eruption (June 11-16, 2009) as an example, we identified and labeled the observational data, proposed data features, and generated datasets. The design and number of features were chosen considering the computational efficiency of the algorithm and its potential applicability in monitoring systems. A machine learning model based on a gradient boosting technique was trained and achieved a Matthews correlation coefficient of 0.86 on the test dataset, indicating the high quality. The proposed algorithm detects 96 % of the disturbances in the datasets with a minimal number of false positives (48 out of 200 test files), exhibiting a significant improvement compared with the conventional STA\LTA algorithm, which detected only 11 % of the disturbances. The validation of the algorithm on data corresponding to the Calbuco (2015) and Hunga Tonga-Hunga Ha'apai (2022) volcanoes eruptions showed 94 % and 81 % of the disturbances detected, respectively. (c) 2024 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Automatic cropland monitoring is becoming increasingly important in the advancement of sustainable agriculture. However, multiannual satellite-based crop mapping across different regions remains challenging due to variations in crop phenology and meteorological conditions. The use of multispectral data from a single satellite can also present difficulties in constructing vegetation index time series, particularly in regions affected by persistent cloud cover. In this study, NDVI time series obtained from Sentinel-2 and Landsat-8/9 imagery were fitted using a Fourier series, and daily NDVI composites from the Meteor-M satellite were obtained for Khabarovsk Krai in the Russian Far East from 2022 to 2024. These data were used to perform random forest (RF) classification for each year for five land cover classes: soybean, grain crops, perennial grasses, buckwheat, and fallow land. The average annual classification accuracies were 87% for Landsat 8/9, 89% for Meteor-M, and 93% for Sentinel-2. Combining data from all three satellites improved classification performance, increasing cross-validation overall accuracy from 92% to 96% in 2022, from 96% to 97% in 2023. These results demonstrate the potential of using both individual satellite data for sufficiently accurate mapping and combined datasets, which provide consistently high accuracy and are a reliable alternative when Sentinel data are limited due to cloud cover.

This paper presents a method for retrieving optical parameters from volcanic sulfate aerosols from the AHI radiometer on board the Himawari-8 satellite. The proposed method is based on optical models for various mixtures of aerosol components from volcanic clouds, including ash particles, ice crystals, water drops, and sulfate aerosol droplets. The application of multi-component optical models of various aerosol compositions allows for the optical thickness and mass loading of sulfate aerosol to be estimated in the sulfuric cloud formed after the Karymsky volcano eruption on 3 November 2021. A comprehensive analysis of the brightness temperatures of the sulfuric cloud in the infrared bands was performed, which revealed that the cloud was composed of a mixture of sulfate aerosol and water droplets. Using models of various aerosol compositions allows for the satellite-based estimation of optical parameters not only for sulfate aerosol but also for the whole aerosol mixture.

Crop identification at the field level using remote sensing data is a very important task. However, the use of multispectral data for the construction of vegetation indices is sometimes impossible or limited. For such situations, solutions based on the use of time series of synthetic aperture radar (SAR) indices are promising, eliminating the problems associated with cloudiness and providing an assessment of crop development characteristics during the growing season. We evaluated the use of time series of synthetic aperture radar (SAR) indices to characterize crop development during the growing season. The use of SAR imagery for crop identification addresses issues related to cloudiness. Therefore, it is important to choose the SAR index that is the most stable and has the lowest spatial variability throughout the growing season while being comparable to the normalized difference vegetation index (NDVI). The presented work is devoted to the study of these issues. In this study, the spatial variabilities of different SAR indices time series were compared for a single region for the first time to identify the most stable index for use in precision agriculture, including the in-field heterogeneity of crop sites, crop rotation control, mapping, and other tasks in various agricultural areas. Seventeen Sentinel-1B images of the southern part of the Khabarovsk Territory in the Russian Far East at a spatial resolution of 20 m and temporal resolution of 12 days for the period between 14 April 2021 and 1 November 2021 were obtained and processed to generate vertical–horizontal/vertical–vertical polarization (VH/VV), radar vegetation index (RVI), and dual polarimetric radar vegetation index (DpRVI) time series. NDVI time series were constructed from multispectral Sentinel-2 images using a cloud cover mask. The characteristics of time series maximums were calculated for different types of crops: soybean, oat, buckwheat, and timothy grass. The DpRVI index exhibited the highest stability, with coefficients of variation of the time series that were significantly lower than those for RVI and VH/VV. The main characteristics of the SAR and NDVI time series—the maximum values, the dates of the maximum values, and the variability of these indices—were compared. The variabilities of the maximum values and dates of maximum values for DpRVI were lower than for RVI and VH/VV, whereas the variabilities of the maximum values and the dates of maximum values were comparable for DpRVI and NDVI. On the basis of the DpRVI index, classifications were carried out using seven machine learning methods (fine tree, quadratic discriminant, Gaussian naïve Bayes, fine k nearest neighbors or KNN, random under-sampling boosting or RUSBoost, random forest, and support vector machine) for experimental sites covering a total area of 1009.8 ha. The quadratic discriminant method yielded the best results, with a pixel classification accuracy of approximately 82% and a kappa value of 0.67. Overall, 90% of soybean, 74.1% of oat, 68.9% of buckwheat, and 57.6% of timothy grass pixels were correctly classified. At the field level, 94% of the fields included in the test dataset were correctly classified. The paper results show that the DpRVI can be used in cases where the NDVI is limited, allowing for the monitoring of phenological development and crop mapping. The research results can be used in the south of Khabarovsk Territory and in neighboring territories.

The powerful explosive eruptions with large volumes of volcanic ash pose a great danger to the population and jet aircraft. Global experience in monitoring volcanoes and observing changes in the parameters of their thermal anomalies is successfully used to analyze the activity of volcanoes and predict their danger to the population. The Kamchatka Peninsula in Russia, with its 30 active volcanoes, is one of the most volcanically active regions in the world. The article considers the thermal activity in 2015–2022 of the Klyuchevskoy, Sheveluch, Bezymianny, and Karymsky volcanoes, whose rock composition varies from basaltic andesite to dacite. This study is based on the analysis of the Value of Temperature Difference between the thermal Anomaly and the Background (the VTDAB), obtained by manual processing of the AVHRR, MODIS, VIIRS, and MSU-MR satellite data in the VolSatView information system. Based on the VTDAB data, the following “background activity of the volcanoes” was determined: 20 °C for Sheveluch and Bezymianny, 12 °C for Klyuchevskoy, and 13–15 °C for Karymsky. This study showed that the highest temperature of the thermal anomaly corresponds to the juvenile magmatic material that arrived on the earth’s surface. The highest VTDAB is different for each volcano; it depends on the composition of the eruptive products produced by the volcano and on the character of an eruption. A joint analysis of the dynamics of the eruption of each volcano and changes in its thermal activity made it possible to determine the range of the VTDAB for different phases of a volcanic eruption.

Provision of flight safety during hypersonic flight is a challenging scientific and technical problem. Waverider concept is based on matching the wing leading edge of the shock formed off the vehicle forebody. A hypersonic vehicle spends the major part of its cruise flight in high temperature flow. Design and optimization of waverider at hypersonic speeds is a challenging problem because a large number of variants are required to be computed to achieve a larger lift-to-drag ratios. Numerical simulation of the flowfield around a hypersonic waverider is performed using a high-temperature air model and a hybrid architecture based on graphics processing units. The mathematical model and computational algorithm are verified and validated against CFD benchmark problems. The results obtained show flowfield around hypersonic waverider and its aerodynamic quality at different angles of attack. The scalability of the developed model is investigated, and the results of the study of the efficiency of calculating hypersonic fluidflows on graphics processors are presented. The computational times achieved with the perfect and real gas models are compared.

One of the most important tasks when studying volcanic activity is to monitor their thermal radiation. To fix and assess the evolution of thermal anomalies in areas of volcanoes, specialized hardware-thermal imagers are usually used, as well as specialized instruments of modern satellite systems. The data obtained with their help contain information that makes it relatively easy to track changes in temperature and the size of a thermal anomaly. At the same time, due to the high cost of such complexes and other limitations, thermal imagers sometimes cannot be used to solve scientific problems related to the study of volcanoes. In the current paper, day/night video cameras with an infrared-cut filter are considered as an alternative to specialized tools for monitoring volcanoes’ thermal activity. In the daytime, a camera operated in the visible range, and at night the filter was removed, increasing the camera’s light sensitivity by allowing near-infrared light to hit the sensor. In that mode, a visible thermal anomaly could be registered on images, as well as other bright glows, flares, and other artifacts. The purpose of this study is to detect thermal anomalies on night images, separate them from other bright areas, and find their characteristics, which could be used for volcano activity monitoring. Using the image archive of the Sheveluch volcano as an example, this article presents the results of developing a computer algorithm that makes it possible to find and classify thermal anomalies on video frames with an accuracy of 98%. The test results are presented, along with their validation based on thermal activity data obtained from satellite systems.

This paper reconstructs, for the first time, the motion dynamics of an eruptive cloud formed during the catastrophic eruption of the Sheveluch volcano in November 1964 (Volcanic Explosivity Index 4+). This became possible due to the public availability of atmospheric reanalysis data from the ERA-40 archive of the European Center for Medium-Range Weather Forecasts (ECMWF) and the development of numerical modeling of volcanic ash cloud propagation. The simulation of the eruptive cloud motion process, which was carried out using the FALL3D and PUFF models, made it possible to clarify the sequence of events of this eruption (destruction of extrusive domes in the crater and the formation of an eruptive column and pyroclastic flows), which lasted only 1 h 12 min. During the eruption, the ash cloud consisted of two parts: the main eruptive cloud that rose up to 15,000 m above sea level (a.s.l.), and the co-ignimbrite cloud that formed above the moving pyroclastic flows. The ashfall in Ust-Kamchatsk (Kamchatka) first occurred out of the eruptive cloud moving at a higher speed, then out of the co-ignimbrite cloud. In Nikolskoye (Bering Island, Commander Islands), ash fell only out of the co-ignimbrite cloud. Under the turbulent diffusion, the forefront of the main eruptive cloud rose slowly in the atmosphere and reached 16,500 m a.s.l. by 04:07 UTC on November 12. Three days after the eruption began, the eruptive cloud stretched for 3000 km over the territories of the countries of Russia, Canada, the USA, Mexico, and over both the Bering Sea and the Pacific Ocean. It is assumed that the well-known long-term decrease in the solar radiation intensity in the northern latitudes from 1963–1966, which was established according to the world remote sensing data, was associated with the spread of aerosol clouds formed not only by the Agung volcano, but those formed during the 1964 Sheveluch volcano catastrophic eruption.

Strong explosive eruptions of volcanoes are the most dangerous for aircraft because they can produce in a few hours or days to the atmosphere and the stratosphere till several cubic kilometers of volcanic ash and aerosols. Ash plumes and the clouds, depending on the power of the eruption, the strength and wind speed, can travel thousands of kilometers from the volcano for several days, remaining hazardous to aircraft, as the melting temperature of small particles of ash below the operating temperature of jet engines. There are 30 active volcanoes in the Kamchatka; scientists of KVERT monitor these volcanoes since 1993. Description of volcanic eruptions is based on video monitoring and various satellite data from the information system "Remote monitoring of the activity of volcanoes of the Kamchatka and the Kuriles" (VolSatView, http://kamchatka.volcanoes.smislab.ru). In 2021, three volcanoes (Sheveluch, Klyuchevskoy, and Karymsky) had eruptions. The eruptive activity of Sheveluch (growth of the lava dome) is continuing since 1980. In 2021, explosions sent ash up to 7.5 km a.s.l. mainly in August and December; ash plumes were extending more 380 km to the different directions of the volcano. A new plastic lava block Dolphin-2 squeezed at the dome from February till July 2021. Resuspended ash was observed on 02-03 April, 06-07 July, 13-14 and 22 August, and 06-07 and 21 October: ash plumes were extending for 400 km to the east and southeast of the volcano. Satellite data by KVERT showed a thermal anomaly over the volcano all year. Activity of the volcano was dangerous to local aviation. The terminal explosive-effusive eruptions of Klyuchevskoy volcano took place from 30 September, 2020 to 08 February, 2021. Explosions sent ash up to 8 km a.s.l., gas-steam plumes containing some amount of ash were extending for 500 km to the different directions of the volcano. The lava flows moved along Apakhonchichsky and Kozyrevsky chutes. Satellite data by KVERT showed a thermal anomaly over the volcano all year. The lateral break on the northwestern slope of Klyuchevskoy at an altitude of 2.8 km a.s.l. lasted from 17 February to 20 March, 2021: lava effused from two cracks, a cinder cone 60 m high was formed. By February 23, lava flows 1.2 km long reached the Erman glacier, mud flows passed about 30 km. Activity of the volcano was dangerous to international and local aviation. Eruptive activity of Karymsky volcano was uneven in 2021. According to satellite data, the strong ash explosions were observed: on 04 April (8.5 km a.s.l.), 10 September (7 km a.s.l.), 03 November (11 km a.s.l.), and 06, 13, and 18 November (8 km a.s.l.); in the other months explosions sent ash up to 6 km a.s.l.; ash plumes and clouds drifted for 2700 km to the different directions from the volcano. The thermal anomaly over the volcano was recorded on satellite images from time to time. Activity of the volcano was dangerous to international and local aviation.

О. А. Гирина , А. Г. Маневич , Д. В. Мельников , А. А. Нуждаев , А. В. Кашницкий , И. А. Уваров , И. М. Романова , А. А. Сорокин , С. И. Мальковский , С. П. Королев , Л. С. Крамарева 4 1 Институт вулканологии и сейсмологии ДВО РАН Петропавловск-Камчатский, 683006, Россия E-mail: girina@kscnet.ru 2 Институт космических исследований РАН, Москва, 117997, Россия 3 Вычислительный центр ДВО РАН, Хабаровск, 680000, Россия 4 Дальневосточный центр НИЦ «Планета», Хабаровск, 680000, Россия

This paper contains the results of research on the general matrix multiplication routine performance on modern heterogeneous computing systems. In addition to the single-threaded and multi-threaded performance of the routine for matrices of double-precision real and complex numbers on the IBM POWER and Intel Xeon CPUs, the possibility of automatic offload calculation to NVIDIA GPUs, which is supported by certain BLAS library implementations, was studied. Special attention was paid to the impact on the performance of the bandwidth of the interconnects, which ensure CPU-to-GPU interaction. The obtained results show that IBM computing systems with a high-speed NVLink interconnect demonstrate the best performance doing matrix multiplication on GPUs. Accordingly, these systems can be used to accelerate the solution of tasks that utilize this routines without the need to significantly alter the existing software. It should be noted that CPUs of Intel computing system and the Intel MKL library show the best efficiency performing operations with small matrices. Research results can be used to develop approaches to improving the performance of software, which utilize the general matrix multiplication routine.

Forecasting crop yields based on remote sensing data is one of the most important tasks in agriculture. Soybean is the main crop in the Russian Far East. It is desirable to forecast soybean yield as early as possible while maintaining high accuracy. This study aimed to investigate seasonal time series of the normalized difference vegetation index (NDVI) to achieve early forecasting of soybean yield. This research used data from the Moderate Resolution Image Spectroradiometer (MODIS), an arable-land mask obtained from the VEGA-Science web service, and soybean yield data for 2008–2017 for the Jewish Autonomous Region (JAR) districts. Four approximating functions were fitted to model the NDVI time series: Gaussian, double logistic (DL), and quadratic and cubic polynomials. In the period from calendar weeks 22–42 (end of May to mid-October), averaged over two districts, the model using the DL function showed the highest accuracy (mean absolute percentage error –4.0%, root mean square error (RMSE) –0.029, P < 0.01). The yield forecast accuracy of prediction in the period of weeks 25–30 in JAR municipalities using the parameters of the Gaussian function was higher (P < 0.05) than that using the other functions. The mean forecast error for the Gaussian function was 14.9% in week 25 (RMSE was 0.21 t ha−1) and 5.1%−12.9% in weeks 26–30 (RMSE varied from 0.06 to 0.15 t ha−1) according to the 2013–2017 data. In weeks 31–32, the error was 5.0%−5.4% (RMSE was 0.07 t ha−1) using the Gaussian parameters and 7.4%−7.7% (RMSE was 0.09–0.11 t ha−1) for the DL function. When the method was applied to municipal districts of other soy-producing regions of the Russian Far East. RMSE was 0.14–0.32 t ha−1 in weeks 25–26 and did not exceed 0.20 t ha−1 in subsequent weeks.

Currently, video observation systems are actively used for volcano activity monitoring. Video cameras allow us to remotely assess the state of a dangerous natural object and to detect thermal anomalies if technical capabilities are available. However, continuous use of visible band cameras instead of special tools (for example, thermal cameras), produces large number of images, that require the application of special algorithms both for preliminary filtering out the images with area of interest hidden due to weather or illumination conditions, and for volcano activity detection. Existing algorithms use preselected regions of interest in the frame for analysis. This region could be changed occasionally to observe events in a specific area of the volcano. It is a problem to set it in advance and keep it up to date, especially for an observation network with multiple cameras. The accumulated perennial archives of images with documented eruptions allow us to use modern deep learning technologies for whole frame analysis to solve the specified task. The article presents the development of algorithms to classify volcano images produced by video observation systems. The focus is on developing the algorithms to create a labelled dataset from an unstructured archive using existing and authors proposed techniques. The developed solution was tested using the archive of the video observation system for the volcanoes of Kamchatka, in particular the observation data for the Klyuchevskoy volcano. The tests show the high efficiency of the use of convolutional neural networks in volcano image classification, and the accuracy of classification achieved 91%. The resulting dataset consisting of 15,000 images and labelled in three classes of scenes is the first dataset of this kind of Kamchatka volcanoes. It can be used to develop systems for monitoring other stratovolcanoes that occupy most of the video frame.

Fast Fourier transform is widely used to solve numerous scientific and engineering problems. In particular, this transform is behind the software dealing with speech and image recognition, signal analysis, modeling of properties of new materials and substances, etc. Newly emerging high-performance hybrid computing systems, as well as systems with alternative architectures, require research on discrete Fourier transform computation efficiency on these new platforms. The results of such research allow assessing the feasibility of certain solutions for building modern computing and data processing centers. This paper presents the results of such research covering modern hybrid computing systems based on the IBM POWER and Intel Xeon processors, as well as on NVIDIA Tesla co-processors. The analysis is carried out, and conclusions are presented on their performance when executing fast Fourier transforms. The impact of the existing architectural aspects of the hardware (CPU simultaneous multithreading mode, GPU data transfer bus, etc.) on the transform performance efficiency is assessed. The obtained results are used to provide recommendations on the optimal operation modes and settings of the considered mathematical libraries.

Strong explosive eruptions of volcanoes are the most dangerous for aircraft because they can produce in a few hours or days to the atmosphere and the stratosphere till several cubic kilometers of volcanic ash and aerosols. Ash plumes and the clouds, depending on the power of the eruption, the strength and wind speed, can travel thousands of kilometers from the volcano for several days, remaining hazardous to aircraft, as the melting temperature of small particles of ash below the operating temperature of jet engines. There are 30 active volcanoes in the Kamchatka, and several of them are continuously active. Scientists of KVERT monitor Kamchatkan volcanoes since 1993. In 2020, four of these volcanoes (Sheveluch, Klyuchevskoy, Bezymianny, and Karymsky) had strong and moderate explosive eruptions. The eruptive activity of Sheveluch volcano began since 1980 (growth of the lava dome) and it is continuing at present. In 2020, strong explosions sent ash up to 7-10 km a.s.l. on 08 April, and 22 and 29 December. Ash from explosions rose up to 5-6 km a.s.l. on 13 June, and 24 December. Ash plumes extended more 625 km mainly to the south-east of the volcano. A form of resuspended ash was observed on 20 April, 28 June, 24 August, and 07-10 October: ash plumes extended for 310 km to the northeast and southeast of the volcano. Activity of Sheveluch was dangerous to international and local aviation. Two moderate explosive-effusive eruptions of Klyuchevskoy volcano occurred in 2020: first from 01 November 2019 till 03 July 2020, and second from 30 September, it is continuing in 2021. Explosions sent ash up to 7 km a.s.l., gas-steam plumes containing some amount of ash extended for 465 km to the different directions of the volcano. The lava flows moved along Apakhonchichsky and Kozyrevsky chutes. Activity of the volcano was dangerous to local aviation. The strong explosive eruption of Bezymianny volcano occurred on 21 October: explosions sent ash up to 11 km a.s.l., the large ash cloud was located over Klyuchevskoy group of volcanoes long time and later drifted up to1200 km to the southeast of the volcano. Activity of the volcano was dangerous to international and local aviation. Eruptive activity of Karymsky volcano was uneven in 2020: ash explosions were observed from one (June) to seven (October) days a month, for five months the volcano was quiet. Explosions rose ash up to 8 km a.s.l. (08 November), ash plumes and clouds drifted for 380 km to the different directions of the volcano. The eruptive volcanic activity was observed in April, May, June, July, October, November, and December. Activity of Karymsky was dangerous to international and local aviation.

Relevance. Video surveillance systems are the most important type of instrumental means of observing the state of volcanoes. The high intensity of data collection contributes to the formation of a huge array of images. At the same time climatic peculiarities of the investigated objects location as well as not always stable operation of data transmission channels cause a large number of spoiled or uninformative images. Additional time is spent on viewing them and excluding them from the archive by specialists, which, in general, may affect the speed of decision-making, related to informing the interested parties about the occurrence of natural hazards. In this regard, the task of developing methods and technologies of image analysis and filtering for creating an effective system of video surveillance of volcanoes is relevant. The aim of the research is to develop the algorithms for image analysis of volcanoes. Methods: Canny boundary detector, discrete contours transformation into parametric ones in the form of broken lines, calculation of reference characteristics and comparison of the obtained data with them, analysis of frequency characteristics of the image using Haar transformation, decomposition of the image into a multiscale pyramid, drawing up a vector of characteristic visual features of an object, detection of the object being searched for using an SVM-classifier. Results: algorithms of photo analysis to assess the visibility of volcanoes and search for brightness anomalies. Conclusion. The considered study resulted in developing volcano image analysis algorithms. The methods and approaches used make it possible to effectively solve the problem of assessing the visibility of an object and searching for brightness anomalies. The system capabilities provide a sufficient set of tools to effectively solve the problem of image filtering and searching for brightness anomalies as a possible sign of a volcano's activity. The results obtained can be used in volcano activity monitoring systems to ensure aviation safety and population protection.

Crop yield modeling at the regional level is one of the most important methods to ensure the profitability of the agro-industrial economy and the solving of the food security problem. Due to a lack of information about crop distribution over large agricultural areas, as well as the crop separation problem (based on remote sensing data) caused by the similarity of phenological cycles, a question arises regarding the relevance of using data obtained from the arable land mask of the region to predict the yield of individual crops. This study aimed to develop a regression model for soybean crop yield monitoring in municipalities and was conducted in the Khabarovsk Territory, located in the Russian Far East. Moderate Resolution Imaging Spectroradiometer (MODIS) data, an arable land mask, the meteorological characteristics obtained using the VEGA-Science web service, and crop yield data for 2010–2019 were used. The structure of crop distribution in the Khabarovsk District was reproduced in experimental fields, and Normalized Difference Vegetation Index (NDVI) seasonal variation approximating functions were constructed (both for total district sown area and different crops). It was found that the approximating function graph for the experimental fields corresponds to a similar graph for arable land. The maximum NDVI forecast error on the 30th week in 2019 using the approximation parameters according to 2014–2018 did not exceed 0.5%. The root-mean-square error (RMSE) was 0.054. The maximum value of the NDVI, as well as the indicators characterizing the temperature regime, soil moisture, and photosynthetically active radiation in the region during the period from the 1st to the 30th calendar weeks of the year, were previously considered as parameters of the regression model for predicting soybean yield. As a result of the experiments, the NDVI and the duration of the growing season were included in the regression model as independent variables. According to 2010–2018, the mean absolute percentage error (MAPE) of the regression model was 6.2%, and the soybean yield prediction absolute percentage error (APE) for 2019 was 6.3%, while RMSE was 0.13 t/ha. This approach was evaluated with a leave-one-year-out cross-validation procedure. When the calculated maximum NDVI value was used in the regression equation for early forecasting, MAPE in the 28th–30th weeks was less than 10%.