Methodological features of the main clonal micropropagation stages of plum PK SK 1 clonal rootstock (of NCFSCHVW selection) are described. The rootstock is characterized by a high adaptability to soil and climatic conditions of the region. In particular, it exhibits good winter hardiness, resistance to overwatered soils, tolerance to drought, and the capacity to grow on heavy soils. Plum seedlings on this rootstock develop a powerful root system. These advantages render the development and improvement of methods for clonal micropropagation of plum PK SK 1 clonal rootstock highly relevant. The research was carried out in 2019-2022. The period of active shoot growth, i.e., the first and second decade of may, was established to be the most optimal period for introduction of explants into in vitro culture. In this period, regeneration reaches 78.1–93.8 %. Eff ective shoot multiplication proceeds on MS medium with Fe-EDTA or MS medium with Fe-EDDHA and 6-BAP in the amount of 1 mg/L, with the highest number of shoots formed from one explant following the fourth passage (on average, 11.7 and 12.3 shoots per explant). A signifi cant eff ect of the 6-BAP concentration on the number of shoots formed was observed. The combined eff ect of the form of chelated iron and 6-BAP concentration on the number of shoots formed was established only after 3-5 passages on MS medium with Fe-EDDHA. The rooting stage of PK SK 1 rootstock can also be carried out on a hormone-free MS medium with Fe-EDDHA. The rooting effi ciency reaches 57.1–61.8 %. On average, 2.4 roots are formed per shoot with a length from 1.2 to 5.1 cm, with the root quality being high (not cartilaginous and not brittle). For the adaptation stage, rooted shoots with a height of at least 2 cm having two or more roots with a length of at least 2-3 cm are selected. A mixture of the Agrobalt plant nutrient soil, vermiculite, and perlite in a ratio of 3:1:1 can be used as a substrate. Following four weeks of adaptation, the number of adapted plants averaged 87.5 %. 

Physiological parameters of drought tolerance of fi ve raspberry varieties (‘Meteor’, ‘Gusar’, ‘Lavina’, ‘Patricija’, and ‘Ulyabka’) and seven selected raspberry forms (4-122-2, 6-125-3, 1-76-1, 6-125-4, 2-83-21, 1-76-25, 2-90-3) were investigated. Experiments were conducted in 2022–2023 at the Kokinsky experimental station of the Federal Horticultural Center for Breeding, Agrotechnology and Nursery (Bryansk Oblast) and at the Center for Collective Use of Bryansk State Agrarian University. Drought tolerance was assessed based on the indices of total leaf water content, water defi cit, and water-retaining capacity in the phenophases of budding (May) and fruiting (July). The analysis showed that within one vegetation period, the total leaf water content and water-retaining capacity decreased from the phase of budding to that of fruiting, compared to water defi cit, which showed an inverse trend. The water-retaining capacity of the studied varieties was found to be higher in a drier season than in a moderatelyhumid year. A signifi cant variability among genotypes was observed in terms of water regime. During the fruiting phase, the total leaf water content decreased to an average level in all the studied genotypes (less than 70.0 %), except for the ‘Gusar’ and ‘Lavina’ varieties which demonstrated lower levels (less than 60.0 %). Forms outperforming in a number of leaf water regime parameters were identifi ed. Thus, the ‘Meteor’ variety, although showing high values of water content in the budding phenophase, was characterized by low water losses after 6-hour wilting (less than 30.0 %) and a leaf water defi cit (less than 10.0 %). Water defi cit in the ‘Ulyabka’ variety, as well as 6-125-4 and 2-90-3 selected forms, exceeded 10.0 % in the fruiting phenophase, corresponding to the average level of drought tolerance. The 2-83-21 form showed a reduced water-retaining capacity in the fruiting phenophase (water losses exceeded 30.0 %). 

This article describes modification changes in the phenotypic traits of grape varieties, including the number of normally developed shoots and inflorescences, the average bunch weight, and grape yield under the influence of unstable weather conditions. The research was carried out in the moderate continental climate of the Lower Don region in the period from 2013 to 2022. The average annual air temperature ranged from +10.2 to +11.6 °C, with the climatic norm being +10.4 °C. In the period of active vegetation (May–September), the temperature ranged from +21.2 to +22.7 °C, with the norm being 18.1 °C. The maximum temperature ranged from +37.2 to +40.0 °C, with the norm being +40.0 °C. In winter, during the plant dormancy period (December–February), the average air temperature varied from 0.6 to minus 2.6 °С, with the norm being about minus 2.4 °С. The annual precipitation varied from 292 to 693 mm, with the norm being 548 mm. During the vegetation period of grapes, atmospheric precipitation by the years of observation varied from 99 to 359 mm, with the norm being 177 mm. The aim was to establish the limits of modification variability of phenotypic traits under the influence of the unstable moderate continental climate of the Lower Don Region. The research objects were autochthonous grape varieties, including ‘Varyushkin’, ‘Krasnostop’ ‘Zolotovsky’, ‘Kumshatsky Belyi’, ‘Plechistik’, ‘Sibirkovy’, and ‘Tsimlyansky Chernyi’. The reaction norm and modification changes of the phenotypic traits of grape plants under the influence of natural conditions in their habitats were studied. The autochthonous grape varieties in contrasting weather conditions exhibited individual and varietal specificity in terms of reaction norm and adaptability. According to the variability of phenotypic traits and adaptive potential, the varieties were ranked in the following descending order: ‘Plechistik’, ‘Krasnostop Zolotovsky’, ‘Varyushkin’, ‘Kumshatsky Belyi’, ‘Sibirkovy’, and ‘Tsimlyansky Chernyi’. The genotypes ‘Plechistik’, ‘Krasnostop Zolotovsky’, ‘Varyushkin’, and ‘Kumshatsky Belyi’ with their high adaptive potential for formation of stable vine plantations are recommended for developing stable grape plantations in the conditions of the Lower Don region. 

Application of intensive technologies in horticulture with a high mechanical and chemical load on soils and fruit plants leads to deterioration of soil properties, environmental pollution, and development of resistance among pathogens and phytophages. In this study, we aim to developing a biologized technology for apple tree cultivation using a sod and humus system of inter-row maintenance and biopreparations based on active strains of microorganisms, entomoacariphages, and pheromones. The research was conducted in 2019-2023 in intensive apple orchards of the central plain-steppe region of Crimea by the methods conventionally used in soil science, agrochemistry, horticulture, and plant protection. The use of a mixture of cereal-legume grasses in combination with microbiological fertilizers based on nitrogen-fixing microorganisms was found to increase apple yields by 11–12 t/ha, or 64 % compared to the control, by increasing the number of fruits and the average weight of one fruit. A positive balance of nutrition elements in soil with an intensity of 135-270 % was established. The efficiency of controlling the number of phytophages and pathogen harmfulness through the integrated use of biological and biotechnical methods was shown. Thus, the “attract-and-kill” method for Cydia pomonella L. pheromones demonstrated a 98 % efficiency along with a reduction in pesticide load to 7.7 kg/ha season by the preparation. Release of phitoseiid mites for control of tetranychus mites by the method of seasonal colonization provides for a reduction and retention of phytophagous population at the economic threshold of harmfulness during the entire vegetation period. The use of fungicides based on pathogenic bacteria of the Bacillus genus and antagonist fungi of the Trichoderma genus against scab is effective only in the years unfavorable for the development of the disease. The results obtained form the basis for a biologized resource-saving technology for exploitation of intensive apple orchards. This technology contributes to increased soil fertility, reduced chemical and pesticide load, and increased productivity of horticultural agrocenosis under the conditions of the Crimea and southern Russia. 

As part of monitoring works carried out by specialists of the All-Russian Plant Quarantine Center, samples of stone fruit crops collected in 20 Russian Federation subjects were tested by enzyme immunoassay for the presence of the following viruses: ACLSV, ApMV, APLPV, ArMV, CLRV, CRLV, PBNSPaV, PDV, PNRSV, PPV, PRMV, RpRSV, SLRSV, TBRV, ToRSV, and TRSV. The distribution of plum pox potyvirus (PPV), prunus necrotic ring spot ilarvirus (PNRSV), and prune dwarf ilarvirus (PDV) was determined across the area of 12, 12, and 8 Russian Federation subjects in 14.9 %, 11.9 %, and 1.7 % of test samples, respectively. The presence of these viruses was confirmed by polymerase chain reaction with subsequent sequencing of the amplification products obtained. The occurrence of these viruses was analyzed depending on the type of plantation, plant species, and region of distribution. The envelope protein gene region of 18 PNRSV isolates from five Russian Federation subjects was analyzed, which showed a significant genetic variability of this virus among the Russian population. 

This work substantiates the transition to biorational systems of plantation protection using low-toxic preparations to ensure high-quality yields. Regularities in the development of codling moth in Krasnodar Krai are revealed. During the period of 2021–2023, a 30% increase in the expenditures on protective measures in horticultural organizations of Krasnodar Krai was established. As a result, the production costs grew by an average of 17%, decreasing the entire efficiency of fruit production. This explains the need to develop modern protection systems for perennial plantations. Differentiated approaches to apple-tree protection from pests (on the example of codling moth), which are based on application of low-toxic chemical and biorational preparations according to phytophage development stages, are reviewed. An integrated system for protection of industrial fruit plantations in the Krasnodar Krai against coddling moth is proposed. The advantages of the proposed system over the existing approaches that apply highly toxic chemicals are described. These relate to the possibility of introducing flexible changes in protective measures in accordance with phytosanitary characteristics of orchards, including activation of local apple varieties and use of agrotechnical and biological methods based on toxicological monitoring. An ecological and economic evaluation of the integrated system of fruit plantation protection from codling moth is provided. This system enables a reduction in the background pesticide content in food chains (a 2-fold and 5-fold decrease in soils and fruits, respectively), a 1.3-fold reduction in pesticide load, the preservation of yield by more than 98% while complying with the regulations of ecological agrocenosis safety regulations, a 1.5-fold reduction in production costs, and a 10.5% increase in the product profitability. 

This article reports the results of research studies conducted in 2023–2024 on transfer learning of Segmentation Convolutional Neural Networks (Seg-CNN) models for classification, recognition, and segmentation of branches with apple fruits and stems in images. State-of-the-art convolutional neural network architectures, i.e., YOLOv8(n,s,m,l,x)-seg, were used for a detailed segmentation of biological objects in images of varying complexity and scale at the pixel level. An image dataset collected in the field using a GoPro HERO 11 camera was marked up for transfer model training. Data augmentation was performed, producing a total of 2500 images. Image markup was performed using the polygon annotation tool. As a result, polygonal contours around objects were created, outlines of branches, apple tree fruits, and stems were outlined, and segments of objects in the images were indicated. The objects were assigned the following classes: Apple branch, Apple fruit, and Apple stem. Binary classification metrics, such as Precision and Recall, as well as Mean Average Precision (mAP), were used to evaluate the performance of the trained models in recognizing branches with apple fruits and stems in images. The YOLOv8x-seg (mAP50 0.758) and YOLOv8l-seg (mAP50 0.74) models showed high performance in terms of all metrics in recognizing branches, apple fruit, and fruit stems in images, outperforming the YOLOv8n-seg (mAP50 0.7) model due to their more complex architecture. The YOLOv8n-seg model has a faster frame processing speed (11.39 frames/s), rendering it a preferred choice for computing systems with limited resources. The results obtained confirm the prospects of using machine learning algorithms and convolutional neural networks for segmentation and pixel-by-pixel classification of branches with apple fruits and stems on RGB images for monitoring the condition of plants and determining their geometric characteristics.