Dynamics of Distribution of Carbon Stocks in Soils and Structure of Plant Communities under Natural Reforestation on Fallow Lands

This study analyzes the effect of the spatial-species structure of plant communities formed during postagrogenic reforestation of former agricultural lands on the changes in soil organic matter (Corg) stocks. The distribution and dynamics of Corg were estimated with the application of the EFIMOD3 simulation modeling system taking as example the fallow areas of agro-gray soil (Luvisol) in the broadleaf forest zone (54.833770 N, 37.568915 E). We simulated the reforestation on the abandoned soils with pine (Pinus sylvestris L.) or birch (Betula spp.) trees typical for European Russia, with random, clustered, or gradient tree patterns within the fallow area, with the same initial density (5000 trees/ha). According to the estimates, over 50 years, a forest floor pool corresponding to an average of 1.5–2.5 kg C/m2 is formed under birch stands, with spatial variation from 0.5–1.0 to 3.0–3.5 kg C/m2 depending on the distribution pattern of trees. For pine overgrowth, forest floor accumulation is lower, averaging 1.0–1.5 kg C/m2, with a higher variation ranging from 0.5 to 4.0 kg C/m2. In the former arable horizon of fallow lands overgrown with birch, calculations show an increase in the carbon stock by an average of 1.0–2.0 kg C/m2 from the initial value; the greatest increase is observed in the random placement option. In case of pine, the estimates obtained correspond to a decrease in the forest floor Corg stocks in the first 25–30 years, with subsequent recovery and a slight increase in the average value of Corg stocks for the site by 0.5–1.0 kg C/m2 only for the group pattern. An about twofold difference was obtained between the maximum and minimum estimates of the total (in the forest floor and former arable horizon) accumulation of Corg over 50 years in the soils of the simulated area, corresponding to the random pattern of birch (73.6 t C/ha) and pine (38.4 t C/ha), whereas for the clustered one, the model, on the contrary, shows the small differences between birch and pine stands of 59.2 and 60.8 t C/ha, respectively. Numerical experiments have shown that to obtain correct spatial estimates of carbon storage, it is necessary to consider not only the species composition, but also the spatial structure of forests formed on abandoned lands.