The article presents the results of agrobiological and physiological-biochemical studies to ensure the stability of grape plantations in unstable weather conditions and anthropogenic intensification of production. Studies in the agroecological conditions of the Krasnodar region on the southern chernozems revealed the most productive norm of loading of grape bushes with shoots for the technical varieties Kurchanskiy, Dmitriy, Vladimir, Granatovyi, Antaris and Alkor, on leached chernozems – for the table seedless variety Centennial Seedless. According to the results of electrophoretic separation of peroxidases in polyacrylamide gel and an increased content of ascorbic acid, increased frost resistance was established in the varieties Kristall, Vostorg, Krasnostop AZOS. The Vostorg and Zarif varieties proved to be more adaptive to the stresses of the summer period in terms of leaf hydration, bound water content, the ratio of chlorophylls and carotenoids, the amount of proline and water-soluble sugars, membrane permeability. The role of physiologically active substances in increasing the adaptive potential of grape plants has been established. The treatment of Merlot vines with methyl jasmonate and epibrassinolide during deep dormancy period, as well as proline, salicylic acid and epibrassinolide during induced dormancy period had a positive effect on the condition of the vine: there was a decrease in electrolyte yield. Sustainable grape cultivation is achieved with the systemic use of the «Bioconcentrate-Z» effluent by the non-root method. Against the background of nonroot treatments of Merlot and Cabernet Sauvignon varieties, a higher level of plant water supply, an increase in potassium content, green pigments, organic acids and grape yields were established. Physiologically active compounds and the preparation Turinbash have a positive effect in intensive own-root plantations of native and promising introduced grape varieties against the background of phylloxera infection in the southern part of Dagestan. The created yield forecasting method based on the NDVI of the grape plant and the actual data of multispectral images of the phenophases of flowering and growth allows to increase the accuracy of the calculated yield capacity to the actual one with a deviation from 0 to 0.8 kg per bush, the accuracy of the forecast increases by 2 times by the end of the growth phase and by the beginning of the maturation phase. Key words: GRAPES, METHODS, MANAGEMENT, SUSTAINABILITY, PRODUCTIVITY, QUALITY, FORECASTING