Impact of wetting and drying cycles on soil structure dynamics
Introduction
Soil structure is an important indicator of the ecological status of soil as it both controls many ecosystem functions of soil and is shaped by them (Rabot et al., 2018; Young, 2004). It defines the pathways for water and nutrient fluxes (Jarvis, 2007; Köhne et al., 2009), shapes microhabitats in soil (Baveye et al., 2018; Bottinelli et al., 2015; Young et al., 2008) and controls the micro-environmental conditions for chemical reaction patterns in soil (Totsche et al., 2018; Wilcke and Kaupenjohann, 1997). The spatial distribution of substrate and structure-mediated pathways of oxygen diffusion exert a major control on aerobic and anaerobic soil respiration (Keiluweit et al., 2017; Kuzyakov and Blagodatskaya, 2015; Smith et al., 2003) and is considered as one of the major factors controlling long-term carbon stabilization in soil (Kravchenko and Guber, 2017; Lehmann and Kleber, 2015).
Soil structure can in the broadest sense be defined as the spatial heterogeneity of the different components or properties of soil (Dexter, 1988). There are two approaches to characterize soil structure (Rabot et al., 2018). One is centered on the structure of solid components and usually focused on aggregates as the building blocks of soil. The other is focused on the pore perspective and explores the spatial arrangement of voids in undisturbed soil. This dichotomy is particularly relevant with respect to soil structure dynamics by biotic and abiotic agents which typically manifests itself through the steady formation and destruction of pores, but not of solid particles. Furthermore, soil structure dynamics might not even be detectable based on standard measures of the pore structure such as porosity and pore size distribution, since these macroscopic measures may stay constant in dynamic equilibrium, when the microscopic destruction and formation of pores are in balance. In order to resolve this problem a structure labeling approach was recently proposed that enables a direct estimation of soil structure dynamics with a combination of X-ray microtomography and a novel structure labelling approach. (Schlüter and Vogel, 2016). Aggregates are coated with small, inert garnet particles. Garnet is an iron-bearing mineral that evokes good X-ray contrast. The garnet particles are in direct contact with inter-aggregate pores after coating, except for those that get occluded in the contact area of adjacent aggregates after repacking. This results on average in shorter distances between garnet particles and nearest pores than the distance of arbitrary soil matrix locations and nearest pores. This represents an analogy to pool dilution experiments to measure carbon turnover with stable isotopes. The “short distance pool” is highly enriched in garnet particles and soil structure dynamics, or soil structure turnover as it was coined in Schlüter and Vogel (2016), will lead to a dilution of this pool by randomization of distances between garnet particles and pores through the formation of new uncoated pores or occlusion of particles through the destruction of old pores Soil structure dynamics measured by this randomization of passively translocated garnet particles has direct consequences for soil carbon turnover as the formation of new pores may expose previously occluded organic matter, which is one main explanation for the so-called Birch effect (Borken and Matzner, 2009; Lopez-Sangil et al., 2018; Navarro-García et al., 2012) Likewise, the destruction of pores may protect organic matter in its vicinity against mineralization (Beylich et al., 2010; Haas et al., 2016).
It was demonstrated that compaction, a typical abiotic structure-changing process does not lead to a randomization of particle locations (Schlüter and Vogel, 2016). The “tracer particle”-pore distances increase through compaction, but this is the same for any location within the soil. Therefore it was hypothesized that other structure-forming processes may be more efficient to induce structure dynamics in the sense of its reorganization with time measured by the randomization of “tracer particle”-pore distances. In this paper, crack dynamics through wetting-drying cycles are investigated as another important abiotic process that is known to modify soil structure.
Crack dynamics mainly depend on the clay content and its mineralogy, as some minerals (i.e. kaolinite or illite) have low to no swelling potential, while this is high for others (i.e. smectite, vermiculite). A second important factor is the heterogeneity in the assembly of soil particles, as more heterogeneous soil matrices tend to crack more easily, e.g. (Fiès and Bruand, 1998; Wang et al., 2018). The aperture and width of cracks are governed by the initial water content, the drying intensity and the antecedent moisture regime. Other important factors are bulk density, the content of soil organic matter (SOM), particulate organic matter (POM), and sesquioxides (Peng et al., 2007; Tang et al., 2011; Zhang et al., 2016). However, studies regarding crack dynamics in soil often evaluate two-dimensional crack patterns in drying soil suspensions and are seldom carried out in intact, structured soils as three-dimensional crack patterns are hard to investigate in opaque media. Here, X-ray tomography provides an opportunity to study crack dynamics through non-invasive imaging. The objectives of this paper are two-fold. First, the capacity of repeated wetting-drying cycles to induce soil structure changes in general and soil structure dynamics in terms of particle-pore distances in particular are investigated. Second, the effect of important soil properties like texture, bulk density, soil organic matter content and clay mineralogy on soil structure changes through wetting-drying cycles is explored.
Section snippets
Materials and methods
We examined three top soils from Germany, which differ in texture, organic matter content and clay mineralogy. The luvisol from Bad Rotthalmünster (RM) has both low clay and low SOM content (Kögel-Knabner et al., 2008), the chernozem from Bad Lauchstädt (BL) has low clay and medium SOM content (Altermann et al., 2005), and the gleysol soil from Giessen (GI) has both high clay and SOM content (Jürgen Böttcher, personal communication) (Table 1). The BL and RM soil are not only similar in terms of
Visual assessment
At the low bulk density (Bd1), inter-aggregate pores are clearly visible in the wet stage (W2) (Fig. 3). This means that a large part of the garnet particles is in direct contact with connected pores. The structural changes through drying (W2 → D2) can be recognized very well in the segmented images. The difference in connected porosity between the two moisture levels seems to increase in the order RM < BL < GI. The proportion of pores classified as occluded is relatively small.
At the high bulk
Impact of soil properties on pore space dynamics
The three soils were chosen such that the impact of texture, SOM content and clay mineralogy on structure dynamics during desiccation and rewetting could be investigated. The initial pore structure prior to the first wetting (W1) is roughly the same for all soils due to identical sieving and packing. The main difference between the luvisol (RM) and the chernozem (BL) is a higher SOM content in the BL soil which entailed a larger increase in macroporosity through drying especially at the low
Conclusions
The magnitude of crack dynamics due to repeated wetting-drying cycles depended on a number of soil properties. The magnitude in macroporosity changes was largest in the clay-rich soil and was on average larger at the lower bulk density (1.22 g/cm3) than at the higher bulk density (1.48 g/cm3). A higher soil organic matter content led to a higher density of cracks with smaller aperture. In none of the investigated soils did the repeated wetting-drying cycles lead to a randomization of distances
Acknowledgments
We are grateful to the editor and two anonymous reviewers for their helpful comments. This study was partially funded by the Deutsche Forschungsgemeinschaft through the research unit DFG-FOR 2337: Denitrification in Agricultural Soils: Integrated Control and Modelling at Various Scales (DASIM). We thank Jürgen Böttcher (Leibniz University Hannover), for sharing texture and organic matter data and Reinhold Jahn, Klaus Kaiser and Sonia Banze (Martin Luther University Halle-Wittenberg) for sharing
References (42)
- et al.
Evaluation of soil compaction effects on soil biota and soil biological processes in soils
Soil Tillage Res.
(2010) - et al.
Why is the influence of soil macrofauna on soil structure only considered by soil ecologists?
Soil Tillage Res.
(2015) Advances in characterization of soil structure
Soil Tillage Res.
(1988)- et al.
Connectivity and percolation of structural pore networks in a cultivated silt loam soil quantified by X-ray tomography
Geoderma
(2017) - et al.
Minimum error thresholding
Pattern Recogn.
(1986) - et al.
A review of model applications for structured soils: a water flow and tracer transport
J. Contam. Hydrol.
(2009) - et al.
Soil pores and their contributions to soil carbon processes
Geoderma
(2017) - et al.
Microbial hotspots and hot moments in soil: concept & review
Soil Biol. Biochem.
(2015) - et al.
Drying and rewetting conditions differentially affect the mineralization of fresh plant litter and extant soil organic matter
Soil Biol. Biochem.
(2018) - et al.
When structure means conservation: effect of aggregate structure in controlling microbial responses to rewetting events
Soil Biol. Biochem.
(2012)