Olga Ivanovna Manakova

Microbial production and CO2 emission in Albic Retisol (Loamic) and Albic Retisol (Aric, Loamic) of Lomonosov Moscow State University carbon polygon «Chashnikovo»≫ in the Moscow region were studied during the peak of vegetation activity. The object represents monitoring sites in two natural ecosystems (secondary spruce forest and mixed-grass meadow) and in two agroecosystems (perennial grasses and bare fallow). The soils differed in their stocks
of organic matter carbon (Corg), microbial carbon (Cmic), and potentially mineralizable carbon (Cpm). The highest values of microbial indicators and carbon content were observed in the upper 30 cm with a maximum in the upper 10 cm. Corg stocks in the 0–30 cm layer were 1.1–1.3 times higher in agricultural soils (72.7–75.6 t ha-1) compared to natural ones (59.4–65.0 tha-1). The contribution of Cmic stocks to Corg stocks in meadow soils and agricultural soils under perennial grasses (1.8 and 1.4%) was higher than in forest soils and soils under bare fallow. Сpm stocks in natural soils (1.34 and 0.97 tha-1 for a layer of 0–10 cm) were 1.3–2.3 times greater than in agricultural soils, but the intensity of mineralization of organic matter in agricultural soils was 1.4–2.9 times lower than in soils of natural ecosystems. For ecosystems with similar vegetation, the influence of Сpm stocks on the amount of microbial CO2 production was shown; its proportional increase was noted. Maximum potential microbial CO2 production (4.8 gCm-2day-1 for the 0–10 cm layer and 10.5 gCm-2day-1 for the 0–30 cm layer), Cmic stocks (0.50 tha-1), as well as CO2 emission (11.09±0.29 gCm-2day-1) were characteristic of the soils of a dry meadow. In other ecosystems, CO2 production by microorganisms was lower by approximately 2 times. Since CO2 emission, in addition to microbial respiration, is also caused by the respiration of plant roots, its minimum values were observed in agricultural soils under bare fallow (5.01±1.43 gCm-2・day-1). CO2 emission from forest soils and agricultural soils under perennial grasses was statistically not significantly lower compared to the meadow.

The carbon reserves in main components of Moscow region coniferous-broadleaf forest were studied: various fractions of the tree stand, dead wood, mortmass, litter, living ground cover and mineral profile of the soil. To assess the potential intensity of organic matter decomposition, a number of indicators of the functioning of the microbial biomass were determined. An assessment is given of the carbon reserves and their shares in the ecosystem components that differ in the rate of renewal and potential capacity for carbon sequestration, as well as the degree of their spatial variation. The total carbon pool of the studied forest ecosystem is 18.7+0.8 kg·m–2, with almost 90% of the total stock concentrated in the perennial parts of the tree stand, dead wood, mortmass and mineral profile of the soil. These most stable pools are characterized by the least spatial variation within the biogeocenosis. The carbon reserves of the assimilating part of the grass layer and tree leaves are only 0.02 and 0.08 kg·m–2, respectively. The carbon reserves of litter are quite low - 0.21 ± 0.04 kg·m–2, which does not exceed 2% of the total carbon reserves of the ecosystem. The data obtained indicate that even secondary subclimax forest ecosystems are a significant absorber of atmospheric carbon, mainly due to the mass of the tree stand and soil organic matter.

The dependence of soil bulk density on penetration resistance, moisture content, soil organic carbon content and particle size distribution was investigated for Retisols of the Educational and Experimental Soil and Ecological Center of Lomonosov Moscow State University “Chashnikovo”. Penetration resistance was measured with a penetrologger at six key sites with different land-use types, in areas in front of three soil profiles. Soil samples for bulk density were taken from the same profiles using 100 cm³ cutting rings. The samples were subsequently analyzed for moisture content, organic carbon (by the Tyurin method), and particle-size distribution using a laser diffraction analyzer. Within the study, 24 regression models for predicting soil bulk density were developed and analyzed. The quality of these models varied; the coefficients of determination (R²) ranged from 0.23 to 0.89 and the root mean square error (RMSE) from 0.07 to 0.18 g×cm^–3. Based on the modelling results, the greatest contribution to predicting bulk density was made by organic carbon content, depth and penetration resistance, in descending order. Adding specific particle-size fractions to the model is more informative than using the principal components of particle-size fractions as predictors. The use of specific particle-size fractions is more informative than using principal components as predictors. In many models, field soil moisture proved to be an insignificant predictor for bulk density. Given the labour-intensive nature of determining a full set of predictors, models with a reduced set of predictors were proposed: 1) depth and penetration resistance; 2) organic carbon content and penetration resistance; and 3) organic carbon content alone. The data from the first of these models are automatically collected during the use of the penetrometer, making it convenient for monitoring surveys. The method has important limitations relating to the particle size distribution of soils and how this is determined. For example, it is impossible to take a top-down measurement of penetration resistance with a penetrologger in the presence of large boulders in the soil, as the measuring tool hits them and cannot be pushed into the underlying horizons. The use of regression equations in which particle-size distribution is determined by a method other than laser diffraction is incorrect. The choice of which particle-size fraction to use as a predictor should be based on a correlation analysis for each territory in question.