The aim of this study was to experimentally determine the effect of different systems of mineral and organic fertilisation on the dynamics of belowground biomass accumulation in giant miscanthus (Miscanthus × giganteus), the intensity of soil respiration, and the carbon sequestration potential under the conditions of the Forest-Steppe of Ukraine. The study was conducted during 2021-2025 in a two-factor field experiment, in which Factor A comprised a control treatment, leonardite application (100 kg/ha), and complete mineral fertiliser N₆₀P₁₆K₈₀, whilst Factor B comprised foliar feeding with potassium humate at the tillering stage, at the stem elongation stage, and at both stages combined. Belowground biomass was determined by the monolith method in the 0-30 cm soil layer, and the biological activity of the rhizosphere was assessed using the closed static chamber method. It was established that the root system mass of miscanthus increased consistently throughout the first five years of vegetation, reaching 10.76 t/ha in 2025 under the optimal combination of N₆₀P₁₆K₈₀ with double foliar feeding with potassium humate, compared with 6.14 t/ha in the control treatment. Carbon input into the soil via root mass in this treatment amounted to 4.52 t C/ha, and the maximum soil respiration intensity (432 mg CO₂/m²·h) was recorded in the leonardite + double potassium humate application treatment. A strong positive correlation was found between root mass and soil respiration intensity (r = 0.74), and the constructed second-degree polynomial regression model (r = 0.96) adequately describes the dynamics of belowground biomass accumulation and enables projection of its further increase. The practical significance of this study lies in the substantiation of the combined use of mineral fertilisers and humic preparations to enhance the productivity of miscanthus plantations, strengthen the carbon sequestration function of the agrocoenosis, and stabilise the biological activity of the soil
carbon sequestration; soil respiration; leonardite; potassium humate; regression model