INTRODUCTION

Soils are the largest reservoir of carbon in the biosphere. They contain on average 1500–2400 Gt С (1 Gt = 19 t) [31]. Russian soils contain approximately one-fifth of the world pool of soil carbon [14]. Soils can be both the source and sink of carbon, depending on the environmental conditions and anthropogenic impact. Therefore, assessment of soil carbon stocks and potential CO2 production by soils relative to the problem of global climate change became especially actual, because these data are necessary for optimum decisions on land use and choice of the strategies to mitigate negative consequences of climatic change.

Carbon accumulation in soil is characterized by the high spatial variability, because it is determined by combined influence of bioclimatic, geomorphic, and anthropogenic factors. According to modern estimates, the stock of organic carbon in soils of Russia reaches 285–364 Gt [26]. To refine this estimate, the improvement of research methodology [1, 9, 14] and a larger number of studies of soil organic matter stocks and the factors controlling them in different bioclimatic regions of Russia are necessary [2, 15, 17, 28].

The value of potential CO2 production by soils is of great importance for forecasting the response of terrestrial ecosystems to global climatic change. It can be obtained on the basis of data on litter stock and typology, which characterize the ratio between the intensity of the processes of plant falloff accumulation and decomposition [23] and the data on the stocks of biologically active organic matter in soils [20].

Our work is aimed at the assessment of carbon stocks and the potential for CO2 production by soils of the subzone of coniferous–broadleaved forests in the of European part of Russia using the Chashnikovo Agrobiological Educational and Experimental Station of Lomonosov Moscow State University (Moscow oblast), the territory of which typical for this region landscapes, where forests alternate with meadows, agricultural lands and fallows.

OBJECTS AND METHODS

The study was carried out in the territory of the Chashnikovo Agrobiological Educational and Experimental Station of Lomonosov Moscow State University in Solnechnogorsk district of Moscow oblast. Field practical trainings (soil mapping, landscape study, and soil science) are carried out here for many years and magister and dissertation works are performed here by students and postgraduates of the Soil Science Faculty, Moscow State University. This paper presents a generalization of data obtained in the past decade.

The Chashnikovo Station is located in the subzone of coniferous–broadleaved forests and is characterized by relatively diverse combinations of soil-formation factors: vegetation, parent rocks, and landforms. A significant role is played by the anthropogenic factor, mostly agricultural land use. The climate is moderately continental. Long-time mean annual air temperature is +3.7°С, mean January temperature is ‒10.5°С, mean July temperature is +17°С; frost-free period lasts for 120–130 days [22]. The mean annual precipitation is about 600–650 mm; a larger part of precipitation falls in the second half of summer (late July–August). The territory of the station is allocated to the Klin–Dmitrov Moraine Ridge and is characterized by the undulating topography. Parent rocks are dominated by moraine deposits of Moscow (Riss) glaciation covered by mantle loams of different thickness [12]. The floodplain of the Klyazma River is composed of alluvium; peat deposits covered by alluvium are found in some places. Zonal vegetation is represented by pure spruce forests or spruce–broadleaved forests on the interfluves and river terraces. Swampy deciduous (birch, aspen) forests of semihydromorphic landscapes develop in the lower parts of slopes and in depressions on the interfluves. Thus, in the area of the Krasnyi Voin Ravine (Kirpichnoe Pole catena), the following sequence of ecosystems with increasing degree of hydromorphism has been described: (1) sedge (Carex pilosa) birch forest (tree stand 10B) in the marginal part of the watershed depression with soddy-podzolic gleyed soil, (2) sedge (C. pilosa) birch–aspen forest (8A2B) on the slope of the depression with soddy-podzolic gleyed soil, (3) tufted hairgrass (Deschampsia cespitosa) birch forest (10B) in the lower part of the depression with peat-mucky podzolic-gley soil [22]. Diverse meadow plant communities predominate on the floodplain. A significant part of the land is plowed or is in the state of postagrogenic secondary succession (unmanaged fallow state). The diversity and mosaic character of the soil cover are determined mostly by the position in mesorelief [6] and the character of vegetation, which can be transformed to different degrees by present or past anthropogenic impact. Agrosoddy-podzolic soils (Albic Retisols (Aric, Loamic)) are widespread on the interfluves and terraces on plowed lands, and regraded agrosoddy-podzolic soils (Albic Retisols (Aric, Loamic)) are found on overgrowing fallow lands [4]. Soddy-podzolic soils (Albic Retisols (Loamic)) are developed under undisturbed forests of automorphic landscapes. Soil names are given according to [13]. Soddy-podzolic gleyed soils (Albic Gleyic Retisols (Loamic)) and peat-podzolic gleyed soils (Albic Gleyic Histic Retisols (Loamic)) correspond to semihydromorphic landscapes. Depending on the groundwater level and anthropogenic impact, various alluvial soils have been described on the Klyazma River floodplain: gray-humus (Umbric Fluvisols), gray-humus gleyed (Umbric Fluvisols (Oxyaquic)), and agrohumus alluvial soils (Umbric Fluvisols (Aric)).

To assess carbon stocks in soils of the studied territory, soil sampling was carried out in the following landscapes: undisturbed or slightly disturbed coniferous forests, swampy coniferous–broadleaved and deciduous forests, plowed interfluve areas, meadows and fallows on formerly plowed lands, and floodplain meadows with different degrees of hydromorphism t (Table 1).

Table 1.   Objects of research in the territory of Chashnikovo station
Full size table

Determination of typology, fractional composition, and carbon stock of litter. Numerous studies indicate that trees in forest ecosystems create powerful and long-acting biogeocenotic fields, under the influence of which the stocks and characteristics of litter and some soil properties regularly change within tessera in the direction from tree trunk [7, 8]. By tessera, the area of influence of the edificator tree is understood; within this area, certain patterns of the ground cover, forest litter, and some soil properties are created under the influence of the tree [10].

The degree of manifestation of tessera patterns may be different depending on the forest type, stand density, effects of uprooting, and other factors [11].

The sampling of litter in the spruce forests, where tesserae are well pronounced, was carried out with due respect to the spatial structure of phytocenosis: near tree trunks (within a radius of 50 cm), under the crowns (crown projection), and in forest openings (open space between the crowns). Sampling in tesserae was carried out in fivefold. The samples were taken randomly in ninefold in swampy birch and aspen forests, where tesserae were not pronounced. Litter was sampled from subhorizons using a 25 × 25 cm frame.

Simultaneously with litter sampling, the thickness of every subhorizon was measured. Litter typology was determined according to the classification by Bogatyrev [3]. The stocks of litters and their subhorizons were calculated per absolutely dry mass. Samples from the upper (L) subhorizon were separated into the following fractions: needles, leaves, herb litter, cones, bark, twigs. The fraction, the origin of which could not be determined because of strong transformation, was classified as detritus. The ash content in subhorizons F and Н and in fractions of subhorizon L was determined with dry combustion at 450°С. Recalculation of data on the loss on ignition into the carbon content of litter and its fractions was carried out according to the following formula based on the experimental data:

$$y = 0.3704x + 1.2186,$$

where у is Сorg, %; and х is ignition loss, %.

Additionally to typology, thickness, and stock of litter layers, we determined some indicators directly related to the intensity of biological cycle and, hence, carbon sequestration in litter:

— stock and fraction of detritus in subhorizon L;

— stock and fraction of easily decomposed components (leaves and herb litter) in subhorizon L;

— the ratio between thickness of subhorizon L and total thickness of the lower subhorizons;

— the ratio between the organic matter stock in subhorizon L and total stock in the lower subhorizons;

— the stock of ash in easily decomposable fraction and the ratio between it and the total ash stock in subhorizon L;

— stocks of organic carbon in litter (total and by subhorizons);

— stocks of organic carbon in detritus (as the fraction most resistant to decomposition) and in easily decomposable components (as mostly available for microorganisms potential sources of atmospheric CO2).

Determination of carbon stock in soil. At present, the organic carbon content in soil is usually determined either by dry combustion or by wet combustion (in particular, by classical Tyurin’s method). The application of dry combustion method, despite its high accuracy, is restricted by the insufficient equipment of laboratories with CHN analyzers. The second method can be realized in every laboratory, so 97% of data on the carbon content in soils have been obtained by the classical wet combustion method [1]. The humus content in most soil samples from the Chashnikovo area was determined by Tyurin’s method.

Estimates of the stocks of organic carbon in soil horizons were carried out on the basis of data on their thickness, Сorg content, and bulk density. Data on the horizon thickness were obtained from field descriptions of soil profiles. The conversion factor 0.58 was used, when determining the carbon content from data on the humus content in mineral soil horizons [16].

In the cases, when data on the density were absent, pedotransfer functions suggested by Chestnykh and Zamolodchikov [29] were used; these functions allow obtaining the most exact estimates of the bulk density of mineral horizons in forest soils of the European territory of Russia [27].

In our work, carbon stocks are presented for soil layers of 0–30 and 0–100 cm without litter. Such selection of layer thickness is determined by the fact that the one-meter layer accumulates most of the organic matter actively participating in the modern carbon cycle [16], and the layer of 0–30 cm is suggested as target layer for implementing the FAO UN Global Soil Organic Carbon Map (GSOC-17) project [30].

Data on the carbon content in soils were obtained from the reports of field training on soil cartography and dissertation of Bondarenko [5]. For soils of natural ecosystems, estimates of carbon stocks were obtained from statistical samples (n = 54, and n = 14) for the layers of 0–30 and 0–100 cm. For plowed and fallow soils, the corresponding samples consisted of 137 and 4 entries, respectively.

Determination the capacity of soils to produce CO2. Semenov et al. [20] determined the relationship between the contents of total and active organic carbon in soils on the basis of a large volume of factual data and obtained linear regression equations allowing calculation of the content of active organic carbon from data on the total carbon content in a wide range of soils. We used suggested equations in our work for rough estimates of the content of biologically active organic matter in the upper layer (0–20 cm) of studied soils:

y = 0.039x + 0.025 for soddy-podzolic soils;

y = 0.024x + 0.038 for alluvial soddy and alluvial meadow soils,

where у is the carbon content of the active organic matter, %; and х is the total content of organic carbon in soil, %.

The carbon content of active organic matter obtained from these calculations was used to estimate active carbon stocks. These approximate estimates (within the order of magnitude) allowed us to compare the studied soils. More exact estimates can be performed according to the results of direct measurements.

The obtained data were treated using Statistica 12 software.

RESULTS AND DISCUSSION

Evaluation of carbon stock in litter. In forests, and especially in spruce forests, spruce as an edificatory tree structures the space by tesserae with certain spatial regularities in the distribution of carbon stock in litter. On one hand, needle litter of most coniferous trees is characterized by the slow decomposition; on the other hand, it stabilizes water regime of litter horizons [33], Thus, the quantity of falling needle litter, which depends on the position relative to the trunk and crown, greatly affects the properties of forest litter horizons in spruce forests.

Wood sorrel spruce forest on soddy-podzolic soil in the gently sloping position on the interfluve is characterized by considerable accumulation of forest litter. Litter stock varies depending on the position in tessera. The maximum litter stock and its maximum spatial variability are characteristic of the positions near the trunks of spruce trees. In these loci, humified litter predominates. The presence of the F and H subhorizons in the forest litter near the trunks of spruce trees attests to the decelerated decomposition of organic matter. Under the crowns and in intercrown spaces, the F horizon predominates. In intercrown spaces, destructive litters composed mostly of the falloff of recent years forming the L horizon can also be found. These patterns in the forest litter distribution have been described by us in earlier studies [21, 24].

Litter horizons near tree trunks accumulate up to 20 t С/ha (subhorizons F and Н contain 80% of this stock), whereas carbon stocks in litter under the central part of spruce crowns and in open intercrown are about ten times smaller (Fig. 1).

Fig. 1.
figure 1

Carbon stock in litter (a) and its spatial variability (b) in tesserae of wood sorrel–yellow archangel spruce forest: (1) mean, (2) standard error, and (3) standard deviation.

Full size image

Fractional composition of litter in spruce forest. Subhorizon L of litter in all components of tessera is dominated by difficultly decomposable fractions (such as twigs, needles, and rarely cones). Carbon stock in detritus as the most slowly decomposable fraction accounts for 8–15% of the total stock of Сorg in subhorizon L. Carbon stock of the most easily decomposable components—leave and herb litter—reaches about 0.1–0.6 t С/ha. There are no statistically significant differences in the contents of carbon of easily decomposable components and detritus in subhorizon L between the sites near the trunk, under crowns, and in open intercrown spaces because of the high spatial variability of this indicator (Fig. 2).

Fig. 2.
figure 2

Relationship between carbon stocks of (a) detritus and (b) easily decomposable fractions in litter of wood sorrel–yellow archangel spruce forest and location in the tessera: (1) mean, (2) standard error, and (3) standard deviation.

Full size image

Thus, the total stock of carbon in the litter horizons of studied spruce forests ranges from 2 to 20 t С/ha and reaches maximum values near tree trunks. More than 90% of the total carbon stock in the litter horizons of spruce forests is accumulated in the subhorizons of humification and fermentation (if they are present) and in difficultly decomposable fractions (bark, twigs, cones, needles) of the L subhorizon.

In herbaceous small-leaved forests, a decrease in the rate of plant falloff decomposition with an increase in the degree of hydromorphism is observed and leads to changes in the structure of litter horizons and an increase in the total litter stock. This corresponds to the earlier obtained information on the contribution of organic (litter) horizons to the total carbon stock in the one-meter layer, which can reach 50% under semihydromorphic and especially hydromorphic conditions [19]. Unlike coniferous species, falloff of small-leaved tree species is less resistant to decomposition, and this does not favor the accumulation of considerable litter stocks, other factors being equal [18].

Both destructive and fermentative litters are found in the studied small-leaved forests, but the fraction of fermentative litter increases from 63% to 78 and 100% with an increase in the soil hydromorphism, which agrees with common understanding of the influence of moistening on the rate of organic matter decomposition [32, 34].

Carbon stock in litter increases by more than three times from sedge birch forest to tufted hairgrass birch forest (Fig. 3). However, even the maximum C stock in the most waterlogged tufted hairgrass birch forest is inferior to C stock in litter of spruce forest.

Fig. 3.
figure 3

Carbon stock in litter and its spatial variability (a) and distribution of carbon stock by litter subhorizons (b) in small-leaved forests of different degrees of hydromorphism: (1) mean, (2) standard error, and (3) standard deviation.

Full size image

Fractional composition of litter in small-leaved forests. According to obtained data, the stock of detritus fraction increases with an increase in the decree of moistening. The stocks of detritus fraction in subhorizon L vary from 0.05–0.1 t/ha (sedge birch forest) to 0.3–0.48 t/ha (tufted hairgrass birch forest) reaching 25% of the total stock of subhorizon L. Tessera patterns are not pronounced in waterlogged small-leaved forests, so it seems impossible to trace regularities of changes in carbon stocks of the detritus fraction depending on the particular loci. The stock of easily decomposable organic matter components in subhorizon L is determined by the features of plant falloff; the maximum content and stock of this fraction are observed under birch–aspen stands. The mean weighted ash content in subhorizon L, the stock of ash from easily decomposable litter components, and its contribution to the total ash stock in subhorizon L are also maximal in birch–aspen forests, and this cannot not but affect the biological cycle and the rate of organic matter decomposition.

In meadow phytocenoses, both on the interfluves and on the floodplains, litter horizons consist of the litter of herbs of previous years (destructive herbaceous litter). Litter stocks in this case are determined by the quantity of incoming falloff of herbs, which depends on the stock and composition of aboveground phytomass and on the rate of falloff decomposition. On one hand, the phytomass stock is, as a rule, higher in the floodplain meadows; on the other hand, the presence of groundwater enriched with nutrients can either slow down or accelerate biological cycle. As a result, calculated carbon stocks in litter horizons of upland meadows regraded agrosoddy-podzolic soils) practically did not differ from those calculated for floodplain meadows (alluvial gray-humus gley soils) and range from 0.8 to 1.0 t С/ha.

In comparison with forest litter, litter horizons in meadow soils are inferior to litter horizons of small-leaved forests even at the most drained sites. Hence, the contribution of meadow ecosystems to carbon accumulation in litter is less than that of other natural ecosystems. As for agroecosystems (plowed fields), aboveground plant residues are not accumulated on the soil surface because of their annual removal in the course of harvesting.

Thus, according to the results of our studies of the litter layer, six groups of biogeocenoses can be distinguished with respect to the possible accumulation of carbon in litter horizons. Complete absence of carbon accumulation in litter is typical of plowed lands; insignificant accumulation (within destructive herbaceous litter, 0.3–1.6 t С/ha) is typical of upland and floodplain meadows. Low carbon accumulation in herb–leave and leave–herb destructive litter (1.1–2.0 t С/ha) is typical of small-leaved forests with mesophilic herbs developed on drained soils. Moderate accumulation (2.8–5.3 t С/ha) in destructive herbaceous and fermentative needle–leave or leave litter is typical of mixed spruce–broadleaved forests and swampy small-leaved forests with tall grasses in the ground cover. Intense litter accumulation (1.7–6.1 t С/ha) is typical of spruce forests in drained positions, which eliminates the presence of the H subhorizon; in these forests, litter mainly consists of the destructive leafy (L) and fermentative (F) subhorizons. Finally, extremely intense carbon accumulation (2.0–20.0 t С/ha) in the humified and fermentative litters is characteristic of typical spruce forests developed on slightly and moderately drained slopes. The studied wood sorrel–yellow archangel spruce forest belongs to this category.

Assessment of organic carbon stock in soil. The stock of soil carbon is determined by a set of factors. The leading role in natural ecosystems in the studied region belongs to the geomorphic position and parent material, as they determine the features of plant communities (productivity, quantity, qualitative composition, and the character of litter input) and the water regime of soil affecting the rate of organic matter mineralization. History of land use and the character of applied agrotechnologies additionally affect carbon stock in soils of agroecosystems. The stock of Сorg in soils of Chashnikovo is characterized by estimates presented in Fig. 4.

Fig. 4.
figure 4

Carbon stocks in soils of Chashnikovo (without litter): (1) soddy-podzolic (Albic Retisols), (2) soddy-podzolic gleyed (Albic Retisols (Stagnic)), (3) gray-humus alluvial (Umbric Fluvisols), (4) gray-humus gleyed alluvial (Umbric Fluvisols (Oxyaquic)), (5) agrosoddy-podzolic (Anthric Albic Retisols (Loamic)), (6) regraded agrosoddy-podzolic (Albic Retisols (Anthric, Loamic)), (7) agrosoddy-podzolic gleyed (Albic Retisols (Anthric, Stagnic, Loamic)), and (8) agrohumus alluvial (Anthric Umbric Fluvisols). Bars reflect error of the mean.

Full size image

As seen from this figure, the obtained estimates vary in a wide range, which is caused by the diversity of combinations of soil forming factors controlling the level of organic carbon accumulation in soils. The stock of Сorg reaches maximum in soils under meadow vegetation on the Klyazma River floodplain. In the one-meter layer of gray-humus alluvial soil, it reaches 206 t С/ha; in the gleyed alluvial soils with an increased moistening, the carbon stock in the one-meter layer increases to 288 t С/ha, though carbon stocks in the upper 30 cm of these soils are characterized by close values. The diversity of moistening conditions on the floodplain explains the wide range of carbon stock in the one-meter layer of alluvial gley soils: from 80 to 567 t С/ha. Agricultural use of alluvial soils leads to a decrease in their organic carbon stock; in the layer of 0–30 cm of plowed agrohumus alluvial soils, it is about 44 t С/ha, which is 2.4 times lower than in the gray-humus alluvial soils under native meadow vegetation.

Soddy-podzolic soils of wood sorrel–yellow archangel spruce forest were studied in more detail among soils of forest ecosystems. In these soils, typical values of carbon stock corresponding to lower and upper quartiles range from 40 to 54 and from 76 to 100 t С/ha in the layers of 0–30 and 0–100 cm, respectively. The difference between minimum and maximum values in the upper layer reaches 67 t С/ha and decreases to 45 t С/ha in the layer of 0–100 cm. The lower range of variation in the C stock in the one-meter layer is apparently connected with a decrease in variation of the carbon content down the soil profile, as well as with a smaller sample volume (6 vs 22). Carbon stock in soddy-podzolic gleyed soils under birch forests is higher than that in soils of spruce forests, and this agrees with the data on the increase in the carbon stock of forest soils with an increase in the degree of soil hydromorphism [27].

Very close estimates were obtained for the mean carbon stocks in the layer of 0–30 cm of the arable agrosoddy-podzolic soils (56 t С/ha) and regraded agrosoddy-podzolic regraded soils under unmanaged fallow (58 t С/ha) (Fig. 5). However, the range of the values of carbon stock in the fallow soils reaches 143 t С/ha, and this is two times greater than in the arable soils. Such a great variation in carbon stock of fallow soils is determined to a great extent by their different ages controlling the stage of plant succession and the level of humus accumulation in overgrowing fallows.

Fig. 5.
figure 5

Carbon stock and its variability in agrosoddy-podzolic and regraded agrosoddy-podzolic soils: (1) mean, (2) standard error, and (3) standard deviation.

Full size image

Thus, carbon stock in the one-meter soil layer without litter in forest ecosystems of Chashnikovo varies from 28 to 129 t С/ha. A half of this stock is accumulated in the upper 30 cm. In the plowed agrosoddy-podzolic soils of interfluves, carbon stock varies from 17 to 89 t С/ha, and typical values are within the range from 47 to 63 t С/ha. In floodplain meadows of different moistening, carbon stocks in the layers of 0–30 and 0–100 cm vary within 25–228 and 123–567 t С/ha, respectively.

Assessment of potential CO2 production. Biologically active organic matter of soil is direct source of greenhouse gases. It includes potentially mineralizable, quickly utilized by microorganisms, capable of chemical and biochemical interactions, and chemically and physically unprotected organic substances with the life time of less than 3–10 years [20]. The stock of active organic matter in soil can serve as one of indicators of soil potential to produce climatically active gases. It is comparable with carbon dioxide emission from soils during the growing season [20, 25].

Assessment of potential CO2 production by litters. Decomposition of organic matter accumulated in litter plays an important role in carbon dioxide production. In this regard, structural and functional parameters of litters can serve as indicators of the intensity of biological cycle and, hence, the proportion between the processes of organic matter accumulation and destruction in litter horizons. These regularities are reflected in the litter typology. Easily decomposable litters of meadow cenoses and small-leaved forests of well-drained territories are characterized by the maximum potential for carbon dioxide production. The studied swampy small-leaved forests are characterized by moderate potential CO2 production and wood sorrel–yellow archangel spruce forests are characterized by the intense accumulation of litter, which suggests their relatively low potential for carbon dioxide production.

The ratio of the thickness or carbon stock of subhorizon L to the thickness (or carbon stocks) in other subhorizons of litter is another informative indicator pf the intensity of litter decomposition. For example, the ratio between the thickness of subhorizon L to the thickness of subhorizons (F + H) in the wood sorrel–yellow archangel spruce forest is 0.43, 1.14, and 1.33 for the area near tree trunk, under the main part of crown, and in intercrown space, respectively. The corresponding ratio for carbon stocks is 0.16, 0.69, and 2.12. It can be concluded that the rate of organic matter decomposition and CO2 production crease from the near-trunk area to intercrown space.

In small-leaved forests, the ratio of carbon stocks in the L and F subhorizons decreases with increasing soil hydromorphism from 4.5 to 2.5 and 0.9. The intensity of organic matter decomposition in moderately waterlogged herbaceous small-leaved forests gains maximum in marginal parts of waterlogged watershed depressions under sedge birch forests.

The content of the detritus fraction in subhorizon L can be one more criterion of potential CO2 production. The greater the detritus content, the lower the rate of organic matter decomposition and, hence, the rate of the CO2 emission.

The stock of biologically active organic matter in the soil layer of 0–20 cm (without litter). Estimates of the stock of biologically active organic matter in alluvial soils and in soddy-podzolic soils on the basis of regression equations from [20] are presented in Table 2.

Table 2.   Calculated stocks of active organic carbon in soils of Chashnikovo station (t С/ha in the layer of 0–20 cm)
Full size table

According to published data, 0.4 to 5.4 t C/ha of potentially mineralizable organic matter are accumulated in the layer of 0–20 cm in soils of different bioclimatic zones. Its stock in arable soils is 1.2–2.4 times lower than in soils under natural vegetation. The stock of carbon in the biologically active organic matter in alluvial meadow soils is higher than that in soddy-podzolic soils. The range of changes in the carbon stock of biologically active organic matter in meadow alluvial soils varies within 1.2–4.2 with the mean value of 2.1. The range of variation of this indicator in soddy-podzolic soils is somewhat narrower (0.6–2.8), and the mean value is 1.1 t С/ha [20]. Our estimates for gray-humus alluvial soils in the territory of Chashnikovo agree well with literature data. The mean stock of biologically active organic matter in soddy-podzolic soils is two times higher than the mean reported in [20] for analogous soils, but it lies within the range of reported values.

Note that our estimates of carbon stock in the biologically active organic matter of soils of natural and agricultural ecosystems in Chashnikovo can serve as indirect approximate estimates of the potential for CO2 production by soils in the studied territory.

CONCLUSIONS

The levels of carbon accumulation in litter and mineral horizons of soils predominating in the area of Chashnikovo station in the subzone of coniferous–broadleaved forests of European Russia have been estimated. Forest litters in spruce biogeocenoses on interfluves are characterized by considerable accumulation of organic carbon, the stock of which depends on the soil position in the tessera (from the area near spruce trunks towards the area under the main crown and intercrown open space). It reaches 20 t С/ha near the trunks and decreases by almost ten times in the soils of intercrown spaces. Carbon stocks in litters of swampy small-leaved forests of semihydromorphic landscapes is determined by the degree of hydromorphism and increases with an increase in the soil moisture level, though remains lower than that in the litter of spruce stands near the trunks. Forest litter typology based on the presence and thickness of litter subhorizons, the content of plant detritus fraction in subhorizon L, and the proportions between thicknesses (or stocks) of subhorizon L and lower (F and H) subhorizons of litter are important indicators of the intensity of organic matter decomposition in forest litters. Herbaceous (meadow) ecosystems are characterized by the minimal accumulation of carbon in the litter horizon.

Maximum carbon stocks in the mineral part of soil profiles have been found for meadow alluvial soils on the Klyazma River floodplain. An increase in the degree of hydromorphism of these soils leads to the increase in the organic carbon stocks. whereas agricultural use of floodplain soils decreases soil carbon stocks. Carbon stock in the one-meter layer of forest soils is two times lower than that in the meadow alluvial soils.

The maximum potential for carbon dioxide production is characteristic of litter horizons in meadow soils and in the soils of small-leaved forests in drained positions. Swampy small-leaved forests were are characterized by the moderate accumulation of carbon in the litter horizons, and the soils of spruce forests are characterized by the active accumulation of carbon in the litter layer, which attests to a relatively low potential of these soils for carbon dioxide production.

Maximums stocks of active organic carbon in the upper (0–20 cm) horizons are found in meadow alluvial soils. The stocks of this carbon are lower in the soils of forest ecosystems. Indirect approximate estimates of the potential CO2 production by soils can be obtained from these data. Judging the structural and functional parameters of litter horizons and data on carbon stocks in the biologically active soil organic matter (in the upper 20 cm), we conclude that the maximum potential for CO2 production among the studied soils is characteristic of the soils of meadow ecosystems.

Thus, the level of organic carbon accumulation and the potential for CO2 production in soils of typical ecosystems in the Chashnikovo territory are determined by the character of vegetation, degree of soil hydromorphism (influence of temporary perched water or groundwater), and specific features of the agricultural use of soils at present or in the past.