Forest practices are rapidly becoming mechanised in France, resulting in unknown consequences for the current and long-term functioning and functions of ecosystems. Degradation of forest soil porosity cannot be remediated artificially, and restoration via natural processes is slow and not likely to include deep soil layers, where tree roots take up large amounts of water and nutrients. In 2007 and 2008, two experimental sites were set up in the Lorraine Plateau (France). The soils are Ruptic Luvisols and are classified as highly sensitive to compaction. We observed multiple parameters and studied weakly bound soil solutions, which are known to react to changes in ecosystem functioning. We hypothesised that (i) soil solution chemistry is a relevant indicator to assess soil changes after compaction and dynamics of soil recovery in the decade following compaction, (ii) restoration is greater at the more fertile site and (iii) soil pH is a relevant parameter to explain the behaviour of soil solutions. Our results showed that soil solution concentrations changed drastically after compaction and that restoration was more effective in the less fertile soil than in the more fertile one. Finally, a soil pH threshold of 4.5 was relevant for explaining the behaviour of nitrate, which is particularly useful for monitoring solution geochemistry in these acidic soils. Liming at the less fertile site increased the pH to more than 4.5, which changed the behaviour of nitrate and reinforced the utility of this threshold for explaining soil biogeochemical functioning. Soil solution is a relevant indicator of current soil functioning after compaction; however, this study demonstrated that additional information is required to understand its meaning accurately. This study highlights that long-term observation is needed to identify the consequences of soil compaction on long-living ecosystems.

The mechanisation of forestry operations is too recent in France to have enough perspective and scientific knowledge of the effects on extensively managed soils. Two experimental sites on soils sensitive to compaction (silty acidic soils laying on top of a clayey subsoil) were set in Lorraine (France). A gentle and controlled compaction of these physically very analogous soils was carried out during spring 2007 at the Azerailles site (AZ) and during spring 2008 at the Clermont-en-Argonne site (CA). Immediate changes were observed for all ecosystem physical, chemical and biological functioning indicators. Soil restoration dynamics were followed using continuous monitoring of numerous parameters, including soil moisture and occurrence of a perched water table (PWT) (monitored at a daily time step) and PWT chemistry (monthly time step). Mid-term monitoring (7-8 years) results showed that both soils shifted towards a hydromorphic soil type attested by the increase in PWT frequency and duration, which invaded the soil upper layers, leading to characteristic Fe, Mn and S mobility and a change in organic compounds stability. Though soil types were very similar, the PWT characteristics varied notably between both sites. The mean residence time of the PWT was shown to be a major driver of its geochemistry, but with strong interaction with soil characteristics. A previous study made on both sites suggested that soil clay content and clay mineralogy controls the PWT dynamics through effects on aggregation. From a PWT perspective, seven years of monitoring were insufficient to observe any soil compaction recovery in the richer soil of AZ but a partial recovery was observed for the chemically poorer soil (CA). Compaction durably impacted the two studied soils and the probability that a new compaction event may occur before the complete recovery from the first disturbance must be considered in forest management.

Processes and rate of macroporosity changes following heavy traffic in forest ecosystems are seldom studied. The aim of this study was to determine the ability of forest soils to regenerate their macroporosity naturally. The study was performed on 2 silty temperate-forest soils classified as sensitive to compaction located in north-eastern France. Macroporosity was measured in control and trafficked plots at 3 depths (0–7, 15–30 and 30–45cm) over 2–3years. Soil macroporosity characteristics (shape, size and orientation) were assessed on polished sections through 2D-image analysis and micromorphic observations. Immediately after heavy traffic, macroporosity decreased by 96 to 49% from 0 to 45cm in depth. Natural regeneration of macroporosity occurred in the upper 7cm of soil, while the soil below remained compacted. Small and medium macropores (0.05–0.8mm2) dominated by rounded and irregular pores regenerated completely. Large macropores (>0.8mm2) originally dominated by vughs, mammilated vughs and channels rarely regenerated and were gradually replaced by horizontally oriented planar pores. Our results suggest that initial stages of natural macroporosity recovery are likely due to plant-root penetration and physical processes (shrink–swell, freeze–thaw), whereas recovery due to fauna activities appears later.

An experimental site was set up in northeast France on a Luvisol (ruptic) soil to examine the duration of physical, chemical and biological disturbances in the soil following mechanized forest harvesting. Soil carbon dioxide (CO2) efflux (SE) and concentration ([CO2]) in the silt loam layer (0–50 cm) were measured from March 2008 to March 2010 in the trafficked (T) and control (C) plots of this site. This study aimed to validate these two measurements as indicators for long‐term soil monitoring following disturbance by heavy traffic in 2007. Throughout the sampling period, SE in the T‐treatment was significantly reduced relative to that in the C‐treatment. The response of [CO2] to traffic depended on the season; it decreased during summer and increased during winter and spring. The combination of the two measurements indicated an increase in the frequency and duration of anoxic conditions resulting from poor gas diffusion after heavy forest traffic. The relationships between soil climatic properties (temperature, water content and water table level) on one hand and SE or [CO2] on the other, demonstrated a strong control of SE by soil biological activity and a double control of [CO2] by gas production/consumption and gas transfer. Our findings suggest that [CO2] and SE are sensitive to soil degradation by forest harvesting and that the impact of soil compaction provides complementary information on the processes involved in regulating CO2 production and efflux. However, their use as simple indicators is questionable as the impact varied with time and was probably dependent on the soil type.

Case studies are necessary to assess the effects of changes to tree species on the physicochemical and chemical properties of soils. To achieve this, the fine earth under five tree species was investigated. This study was performed in the Breuil-Chenue experimental forest site located in the Morvan Mountains (France). This site contains two adjacent blocks with replicated stands. The native forest (old beech and oak coppice with standards) was partially clear-felled and replaced in 1976 with mono-specific plantations of European beech, Norway spruce, Laricio pine and Douglas fir. The same changes in soil properties were revealed in both blocks, thus confirming the tree species effect. The percentage of exchangeable acidity on the cation exchange capacity (CEC) was greater under spruce, Douglas fir and pine than under the other species. Spruce stands, and to a lesser extent those of Douglas fir and pine, had a less acidic soil pH than hardwood stands (which was unusual in view of the data in the literature) and smaller CEC values. The small quantities of carbon added to the soil under these tree species provide an explanation for these effects through a partial control of both CEC and pH. This case study thus demonstrated that the tree species effect was not unequivocal and different criteria are necessary for its interpretation. Tree species significantly influenced certain aspects of the chemical properties of topsoil and have the potential to have an impact on current soil fertility.

Rhizosphere soil can play a central role in the maintenance of the soil-plant system and influencing the biogeochemistry of forest ecosystems. However, rhizospheric studies in situ to understand the ecosystem functions are still lacking. Therefore, the objectives of the present study were to examine the differences in the chemical properties such as the pH, organic C, exchangeable base cations (EBC=Mg2+ + Ca2+ + K+), exchangeable acidity and cation exchange capacity (CEC) of three soil fractions (Bulk, B; Rhizosphere, R and Rhizosphere Interface, RI) and the evolution of chemical properties of soil samples collected in March and June from a Douglas fir ecosystem located in the Beaujolais Mounts in France.Most of the variables measured (organic matter, CEC and EBC) increased significantly in the same order (B < R < RI), indicating that the rhizosphere was a favorable interface for tree nutrition. These processes were more pronounced in June than in March, as the temperature and biological activities are normally higher in June than in March. The temporal variations of Al charge in R and B seem to depend on OH-, organic complexation and H+ production by roots and/or the organic matter degradation in the bulk soil. In the rhizosphere, the H+ or OH- production depends on the N cycle. The results of this study and the support of independent mineralogical study using the same soil fractions, and of other field studies on the same site have all pointed at the importance of rhizosphere as an excellent indicator for the understanding of the ecosystem dynamics in both short- and long-term. (c) 2006 Elsevier B.V. All rights reserved.

We present evidence of surface-controlled and proton-promoted chemical weathering of primary silicates in a brown acidic soil (Vauxrenard, Rhône, France). We used aqueous silica (Si) in soil solutions held at high matric potential (180–1600 kPa), which are representative of solutions reacting with soil solids. Si concentration was well correlated with H+ concentration and to a lesser extent with dissolved organic carbon (DOC), which showed a significant affect (P<0.05) only in the surface layer (0–15 cm). Significant negative linear relationships were obtained between log(Si) and pH at the profile scale, at each soil depth and for most sampling dates. We found no significant influence of soil temperature (P>0.05). Geochemical modelling showed that primary silicates dissolved under far-from-equilibrium conditions, and that organic ligands (modelled with a triprotic analogue) may have a weak but significant effect on the variations in log(Si) at the profile scale and at both 15–30 and 30–45 cm depths. Comparison of Si/Al ratios to literature data and observed soil mineralogy demonstrated that significant linear relationships found in the activity diagram between log[Al3+]+3pH and log[H4SiO4°] may not have been caused by the reversible formation of secondary Al–Si phases in the soil. Instead, the apparent trends may arise from relationships between log(Si) and pH and the control of Al-mobility by the reversible formation of Al-hydroxides in the vermiculite interlayer. All these results indicated that active, in situ chemical weathering of silicates may be surface-controlled and mostly proton-promoted. Mineralogy suggested that K-feldspar weathered much faster than albite and white mica, in contrast to the weathering gradient inferred from the mass balance between unweathered and soil material. This was certainly caused by differential changes in mineral reactive surfaces with time.

SummaryWeathering of soil minerals under forest seems to depend on the species present. To study the influence of tree species we placed unweathered vermiculites in the soil and assessed the impact in terms of saturation index of earth‐alkaline cations and cation exchange capacity in 64 forest stands, dominated by different species and growing side by side at 20 sites on acidic soils. The vermiculites were of two types, one with a large charge and the other with a small charge. Minerals were maintained in the soil for 1 and 3 years.The minerals placed in the topsoil and in soils with low buffering capacity were more acidified and weathered than those deeper in the soil and in less acid conditions. The vermiculites were transformed into hydroxylated interlayered vermiculites, and the formation of hydroxides in the interlayer space decreased the vermiculites' cation exchange capacities. The high‐charge vermiculite had a greater affinity for aluminium than the low‐charge variety. The effect of tree species was significant but small compared with factors such as soil type, depth and duration of incubation. Nevertheless, we can rank the acidifying and weathering caused by the trees in the following order: Picea abies, Abies alba > Pinus sylvestris, Pseudotsuga menziesii > Quercus spp., Fagus sylvatica. This in situ experimental approach enabled us to study potential trends in pedogenesis in few years.

A procedure for quantifying minerals in particle size fractions of fine earth (<2 mm) is presented. Mineral fractions of the acid forest soils are characterised by various transition products. Transitional minerals between illite and vermiculite or smectite, polyhydroxy-Al polymers interlayered in smectite and vermiculite, as well as poorly crystalline compounds are present. Normative calculations with standard minerals, generally applied to rocks, cannot be used to quantify precisely minerals of these soil samples. Conventional qualitative and quantitative chemical analyses are therefore completed by microanalyses of well-identifiable minerals. The calculation procedure Is based mainly on a system of simultaneous linear equations. These are set up from the chemical contents of a particle size fraction on the one hand and from the corresponding clement contents of single minerals on the other hand. The contribution of single minerals to overall parameters can be calculated by programmed algorithms. The methodology was adjusted to different particle size fractions.

Seven forest tree species were tested for their impact on the rate of soil acidification, using both analytical and experimental methods. Soils were investigated by usual analytical methods, and two experiments were carried out on 20 acidic sites where several tree species had been planted. Test-minerals and cation-exchange resins were inserted into soil layers (litter; A and B horizons) and then maintained over various periods. The results of these experiments were interpreted with reference to the theory of acidolysis and acidocomplexolysis developed by Robert et al. [Robert, M., Razzaghe, M., Vicente, M.A., Veneau, G., 1979. Rôle du facteur biochimique dans l'altération des minéraux silicatés. Sci. Sol. 2 (3), 153–174.]. The main conclusions were as follows: (i) The nature of the parent material strongly influenced the weathering of the test-mineral; (ii) Acidification and weathering took place particularly during the dormant period; (iii) Norway spruce (Picea abies Karsten) enhanced the acidification of soils and weathering of test-minerals in comparison with the other tree species tested under the same conditions; (iv) In this study, the effects of tree species were minor in comparison to the other factors.

SummaryThe cation exchange capacity (CEC) of 18 samples (<2 mm) of five acid soils developed on granite or gneiss was measured before and after H2O2 treatment from pH‐unbuffered extraction solution (1 N KCl). The pH in water of the soils ranged from 3.4–4.7. The samples were separated into seven fractions (<2 μm, 2–20 μm, 20–50 μm, 0.05–0.2 mm, 0.2–0.5 mm, 0.5–1 mm and 1–2 mm) and the CEC of each measured. Thus the contribution of each size fraction and of organic matter to the CEC was obtained. From CEC measurements on the different fractions and particle size distribution data, a balance calculation was carried out to verify the methodology. According to the hypotheses used and soil characteristics, the CEC of organic matter was found to vary from 35 to 165 cmolc kg−1 and so represented from 10 to 85% of the total soil CEC in the upper soil horizons. In these sandy soils developed on granite or gneiss which were frequently affected by hydrothermal alteration, the CEC of silt and sand fractions was large. It can represent from 35 to 80% of the total soil mineral CEC. The specific contribution of the 0.2–2 mm fraction can reach 50% of the total soil mineral CEC with values ranging from 0.5–5.2 cmolc kg−1. The mineralogical description showed that hydrothermal white micas and neoformed smectites (precipitated during weathering) were present in all the fractions even in the coarsest ones. In some subsoils, albite grains containing smectite have a CEC as large as 21 cmolc kg−1. This study shows that the CEC of acid soils was not always located in the organic matter and in the clay fraction. The sand fraction can contribute substantially to the soil CEC.

The ways through which ectomycorrhizal fungi benefit tree growth and nutrition have not been totally elucidated. Our study was therefore aimed at assessing fungal access to soil exchangeable and stable mineral reserves. The growth of different ectomycorrhizal fungi in bi-compartment Petri dishes with NH4+- or Ca-saturated vermiculite led to cation exchange reactions and to crystal lattice weathering. The presence or absence of soluble NH4+ or Ca did not seem to affect fungal mobilization potential. Oxalic acid appears to be involved in vermiculite weathering by Paxillus involutus and Ca ions could limit the acid dissolution of Ca-saturated vermiculite. The quantitative significance of such cation mobilization and mineral weathering still has to be assessed in situ and in association with ectomycorrhizal plants.

Two strains of ectomycorrhizal fungi proved able in vitro to alter ammonium saturated vermiculite, freeing bound ammonium ions. Such release is partly associated with crystal lattice dissolution.

In the Landes du Médoc (France), the soil mantle is organized in short soil toposequences (10–20 metres) made of podzols (Haplaquods) in the less waterlogged upper part, and of hydromorphic soils (Haplaquepts) in the lower part. In the upper part, podzolization is shown by the strong redistribution of OM-Al complexes in a Spodic horizon. Because of the presence of these complexes in the groundwater, this process is likely to be active at present. In the eluvial horizons of podzols, the study of the clay minerals has revealed a weathering process by acidocomplexolysis. On the other hand, weathering in the hydromorphic soil has given intergrade minerals which reflects a simple acidolysis. Samples of a test-vermiculite were introduced in the main horizons of the soils. They were removed and reexamined after six months (winter period or summer period) or one year of in situ experiment. The weathering of the test-vermiculite showed distinctive features according to the soils and the seasons. After the summer period no change could be identified in the test-vermiculite from any of the two soils. After the winter period a few interlayers were blocked at 14 Å (interlayered hydroxy-Al) in the vermiculite from the A1 horizon of the hydromorphic soil. The phenomenon was more pronounced for the vermiculite from the E and Bh horizons of the podzol where most of the interlayers were blocked. According to experimental models such a behaviour is due to the presence of a non-complexing acidic medium in these horizons: this does not correspond to the results reduced from the study of the soil clay minerals.

The changes in forest soil functioning due to silvicultural modifications were studied by an in situ experimental method. This method used the evolution of a test mineral as an indicator of modifications in the environment. The processes operating in the soil can be interpreted with reference to laboratory data obtained on the same mineral. After a 3-year contact in acidic soils, the test material (an interstratified mineral containing vermiculite) was studied; CEC, exchangeable and non-exchangeable elements, mineralogy (XRD traces) and compared with the control. The results show that the evolution of the chemical and the mineralogical properties characterize the effects of soil types, soil horizons and forest species. In the acid brown soils the test mineral is highly desaturated while exchangeable acidity increases. In this soil, Al is hydroxylated. The change of forest species (spruce versus mixed broadleaved) produces an important mobilization of Al and tends to form a more stable Al intergrade. In the podzol the quantity of Al is greater than in the acid brown soil but the formation of a very stable Al intergrade mineral was not observed. The change of forest species (spruce versus beech) tends to produce the same relative effect as observed in the acid brown soil. The processes of soil function can be defined as follows: acidolysis is the process which discriminates the function of brown soils; acido-complexolysis is the one which characterizes the A1, A2 horizons of the podzolic soils. We can identify the mechanisms involved in the effect of species on soil functioning. For example, the spruce increases acidity which mobilizes more Al and which tends to form more stable Al intergrade than the broadleaved species. There is a strong interaction between soil and species. Statistical analyses show the level of significance of the effects (soil, species) and allow a clear visualization of the results. This method, with its unique analytical and experimental reference, allows us to compare numerous situations and to characterize the mechanisms of soil functioning in relation to soil type and plant species.

The changes in forest soil function due to silvicultural modifications were studied by an in situ experimental method. This method used the evolution of a test-mineral as an indicator of change in the physico-chemical environment of the soil.The processes operating in the soil can be identified with reference to laboratory data obtained on the same mineral.After 3 years in acidic soils, the test-mineral (an interstratified mineral containing vermiculite) was studied; CEC, exchangeable and non exchangeable elements, mineralogy (XRD traces), and compared with the control.The results showed that if the chemical data and the mineralogy are studied, one can characterize the effects of soil types, soil horizons and forest species. In the acid brown soils the test mineral is highly desaturated while exchangeable acidity increases. In this soil, Al is rapidly hydroxylated. The change of forest species (spruce versus mixed broadleaved) produces an important mobilization of Al and forms a more shable Al intergrate. In the podzol, the quality of Al is greater than in the acid brown soil but the formation of a very stable Al intergrade mineral was not observed. The change of forest species (spruce versus beech) tends to produce the same relative effect as observed in the acid brown soil.The processes of soil function can be defined as follows:- acidolysis is the processes which discriminates the function of acid brown soils;- acido-complexolysis characterizes the A1, A2 horizons of the podzolic soils.We can identify the mechanisms involved in the effect of species on soil function; for example the spruce increases acidity which mobilizes more Al and which tends to form a more stable Al integrate than the broadleaved species; there is a strong interaction between soil and species. Statistical analyses show the level of significance of the effects (soils, species) and allow a clear vizualisation of the results. This method with its unique experimental reference allows us to compare numerous situations and to characterize the mechanisms of soil function in relation to soil type and species.The different processes of soil function that we have recognized, correspond to very different aluminium speciation and cation distribution on the CEC, and consequently to different behaviour for the nutrition and the development of forest species.

SUMMARYA mica was inserted into the soil at four sites to try to identify, in situ, the physico‐chemical changes in the soil caused by the introduction of different tree species, and especially conifers. Changes in the chemistry and mineralogy of the mixed‐layered mica, mica‐chlorite, vermiculite, after 1 1/2 and 31/2 years in the soil, showed good differentiation between soil horizons and between soil types. Differences due to tree species were present, but small. The mechanisms of mineral evolution can be defined with reference to experimental data obtained in the laboratory. In litter layers, the mica‐chlorite sheet was transformed into a mica‐vermiculite, and exchangeable cations predominated on the exchange sites. In the organo‐mineral horizons, changes varied depending on the soil type: in the podzol, transformation was similar to that in the litter layers, but in the sol brun acide, changes were dominated by the Al fixation on the exchange sites and its rapid hydroxylation which blocks these sites, leading to the formation of Al intergrade minerals.Different tree species produced deviations linked to their ability to cycle cations, and also to the different organic materials produced which can attack the mineral.The method has the potential to define the present soil environment and weathering processes, but it needs a more complete experimental design and the use of statistics to clarify the effects of the parameters and their interactions.