The effect of industrial pollution on soil respiration measured in situ has been studied along two gradients formed by emissions from large copper smelters: the Middle Ural Copper Smelter (spruce-fir...

The rate of soil respiration was measured in situ under native pine stands in a large industrial city (Yekaterinburg, Russia) and beyond it. The compared sites differed significantly in the two...

Soil respiration modeling (i.e., simulation of carbon dioxide emissions from the soil surface) makes it possible to analyze and forecast changes in the carbon cycle in terrestrial ecosystems. Along with classical regression models, researchers currently use machine learning methods based on neural networks or regression tree ensembles. However, models produced using these methods are often used as ‘black boxes’, which hinders the analysis of process mechanisms. This paper demonstrates capabilities of the Random Forest algorithm that can be successfully used to estimate effects exercised by various factors on soil respiration based on features importance measurements. Using variance separation, predictors have been classified either as spatial (biotope type, soil type, vegetation type, and soil moisture content) or temporal (soil and air temperature, NDVI, LAI, FPAR, and SPEI). Several models were produced based on 5670 respiration measurements performed during five growing seasons (2012–2016) on 30 sampling plots in south-taiga pine forests and meadows that feature different vegetation and soil types but are confined to the same small area. Different models include different sets of predictors (all predictors, temporal predictors only, spatial predictors only, and temperature and humidity only), and their accuracy reaches R² = 0.88 (MSE = 0.47). It is established that soil respiration depends primarily on temporal factors whose importance ranges from 76 to 91%. In forests, the effect of spatial factors on respiration is stronger than in meadows.

A significant portion of the current knowledge regarding the use of iron nanoparticles for remediating metal-contaminated soils is derived from laboratory experiments, leaving several unanswered questions. This article presents a field experiment comparing the efficacy of magnetite nanoparticles and microparticles for the immobilization of metals and the growth of plants in metal-contaminated soils. This study aimed to investigate the effects of magnetite particle size on metal immobilization and plant growth in soils exposed to airborne pollution from the Middle-Urals Copper Smelter in the southern taiga subzone near Revda, Russia, 50 km from Ekaterinburg. Magnetite nano- and microparticles were added to forest litter at a 4 % w/w dose. The total metal contents in litter from the study plots were 1-2 orders of magnitude higher than background metal concentrations. The magnetite nanoparticle treatment was found to decrease the concentration of exchangeable copper in soil and improve the growth of red fescue (Festuca rubra L.) on polluted soil compared to the control. In contrast, magnetite microparticles did not show any statistically significant effects. These findings are in line with laboratory results that demonstrated the superior metal adsorption properties of magnetite nanoparticles compared to microparticles. However, this study was limited in duration (2 months), and longer field studies would be necessary to confirm the role of iron particle size in the rehabilitation of metal-contaminated soils.

The effect of individual trees on soil and litter respiration in forests polluted with heavy metals from copper smelter emissions was investigated for the first time. We tested the hypothesis that the portion of spatial variability in soil respiration explained by the distance from the tree trunk decreases on polluted plots in comparison with the background area. The study was performed in the southern taiga spruce–fir and birch forests under long-term pollution from the Middle Ural Copper Smelter in the Revda City, Sverdlovsk region, Russia. Measurement points were located near spruce and birch trees at different distances from tree trunks (tree-base plot, the middle of the crown projection, and canopy gap). The total CO₂ emission, litter respiration, litter contribution to soil respiration, specific respiratory activity of litter, and litter stock were measured at each point. In the background area, soil respiration decreased from the tree trunk to the canopy gap. The hypothesis was only partially confirmed: the portion of respiration variance explained by the distance from tree trunks decreased in polluted areas in comparison with the background areas in spruce forests but did not change in birch forests. The observed change in spruce forests was related to a drop in specific respiratory activity of litter, though litter stock was considerably higher near the tree trunk than in the canopy gap. We propose to locate measurement points in the middle of the crown projection, i.e., at a sufficient distance from tree trunks and away from the canopy gaps, to reduce potential bias in soil respiration estimates.