Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Biological soil crusts (biocrusts) are truly organo-sedimentary systems in which both the organic and the inorganic mineral components play dynamic roles in determining the architecture and evolution of the system, as they interact between themselves and with the physical environment. The relevant organic components include the organismal fraction as well as the extracellular organic substances; the mineral components include the parent soil materials and allochthonous minerals trapped by the crust, and some authigenically precipitated minerals. The underpinning of the interactions between biocrust components occurs at the microscale, but results in emergent properties that affect ecosystem-scale phenomena as important as microclimate and hydrological regimes. A basic understanding of the microstructural interactions and dynamics seems thus imperative to attain mechanistic models of biocrust functions and potential responses to a changing environment. Yet, microstructural studies have been few, remain largely descriptive, and have only rarely been integrated with other aspects of biocrust ecology.

Given that the inputs of light energy, most allochthonous minerals, and several nutrients originate from the atmosphere, a strong vertical forcing is imposed upon biocrust architecture. Most generalizations in terms of structure respond to this up/down directionality. While lateral organizational forces do occur, and horizontal variability certainly exists in biocrusts, patterns in this dimension do not lend themselves as easily to generalization and are usually folded into the concept of patchiness. Notably, however, some progress has been attained in describing the structuring influence of aspect exposure in highly three-dimensional, pedicelled crusts.

2 Microstructure of Biological Soil Crusts

2.1 Physical Microstructure

Biocrusts can show a strong vertical layering in their physical microstructure, typically consisting of a physically cohesive upper layer variously called simply crust, topcrust, or bio-rich layer (Fig. 13.1). Topcrust depth is likely determined by the maximal penetration of light into the sedimentary matrix (only a few mm, Garcia-Pichel 1995) that allows for the development of photosynthetic, crust-forming microbial populations. This depth depends on soil texture and the concentrations of microbial pigments. The topcrust layer can often be peeled off and can hold itself together against gravity. The main binding mechanisms may vary. For soils at large [see reviews by Amézketa (1999), and by Bronick and Lal (2005)], abiotic binding mechanisms for soil particles can be the aggregation by carbonates, clay minerals, salts, (hydr)oxides, and phosphates as well as ionic bridging by polyvalent cations. Biotic stabilization can be caused by organic carbon from primary producers as well as the activity of the soil fauna (e.g., earthworms and termites) and microorganisms. The combination and intensity of these mechanisms in biocrusts likely vary, but microbial stabilization is clearly determining in all of them, and abiotic factors have not been examined in much detail. If cyanobacteria are dominant, the stabilization effect can be related to the excreted exo-polysaccharides, often organized as a network of extracellular sheaths (Fig. 13.1). Stabilization by clays in crust roll-ups is demonstrably short-lived compared to stabilization by microbial extracellular polymeric substances (EPS; Beraldi-Campesi and Garcia-Pichel 2010). Cyanobacterial filaments serve as main binding components in cyanobacterial crusts, but fungal hyphae and cyanobacteria may take on that task in lichen-dominated crusts. In moss-dominated crusts, moss rhizoids and fungi become the relevant binding agents (Lan et al. 2012). Free-living fungi do not seem to play a major role in holding non-lichen crusts together, given their low relative biomass (Bates and Garcia-Pichel 2009; Bates et al. 2010).

Fig. 13.1
figure 1

Microstructure of biocrusts: the topcrust. (a): A topcrust from the Jornada del Muerto Valley, New Mexico, Chihuahuan Desert, can hold its own against gravity and can be easily peeled off the soil. (b) Bundle-forming filaments of Microcoleus vaginatus weave the organo-sedimentary matrix of a topcrust. (c) Recurrent burial events can result in multilayered crusts, as in this cross-sectioned example from Bardenas Reales in Spain. (d) The topcrust often develops an internal stratification by trapping allochthonous fines from dust, as seen in this example from the Negev Desert, in which a geological thin section is observed with epifluorescence microscopy to highlight red-fluorescent cyanobacterial filaments (Raanan et al. 2016). (e) In a larger view, the biocrust on sandy soils from the Colorado Plateau is shown in landscape form (courtesy of Estelle Couradeau)

Full size image

But a biocrust area of influence does often extend deeper than this topcrust, into what has been called a “subcrust” or “undercrust.” For the sake of clarity, we will use the term subcrust throughout this chapter. Several studies describe significant differences between this thin, biologically rich soil surface crust cover, and the underlying soil material (Garcia-Pichel et al. 2003; Malam Issa et al. 2009; Lan et al. 2012; Williams et al. 2012; Drahorad and Felix-Henningsen 2013). Penetration resistance (PR) data of needle-type electronic micro-penetrometers are illustrative (Fig. 13.2) because they delimit this small-scale boundary between topcrust and subcrust particularly well (Drahorad and Felix-Henningsen 2013; Drahorad et al. 2013), which is characterized by a reduction in stability at a few mm in soil depth. In this particular case, the subcrust is partly indurated, but this induration does not occur in other crusts. Williams et al. (2012) describe planar voids at this boundary in Mojave Desert crusts. The PR minimum is typical for a breaking point as described by Callebaut et al. (1985) and may indicate a movement of grains into the pore system below or simply a decrease in binding organic substances. PR values among topcrusts vary significantly around the world. Some studies yield values around 49–392 kPa (Guo et al. 2008; Thomas and Dougill 2006, 2007), whereas others report pressures an order of magnitude higher (Maestre et al. 2002). Systematic studies relating PR to biochemical or textural parameters have not been carried out.

Fig. 13.2
figure 2

Microstructure of a biocrust: topcrust and subcrust revealed by penetration resistance (PR). Graph shows the PR for a biocrust-covered Arenosol (solid line, n = 10) and unconsolidated dune sand (dashed line, n = 17). The dotted lines indicate the boundaries of biocrusts between top- and subcrust. From Drahorad and Felix-Henningsen (2013)

Full size image

There is also evidence that the topcrust itself often contains significant microstructural differences. Early on, it was demonstrated using electron microscopy (Verrecchia et al. 1995) that significant quantities of allochthonous dust fines are trapped in the surface of biocrusts (see Fig. 13.1). In some cases these trapped fines can be transported downwards, even beyond the boundaries of the topcrust. Enrichment in fines has been documented as deep as 5–10 mm, well into the subcrust, in the Negev Desert dunes (Yair 1990), but it is unclear if these are the result of transport or may instead correspond to buried surfaces (see below).

Perhaps not surprisingly, the subcrusts are rarely devoted much attention. Mager (2010) defined a “nonorganic” subcrust in the Kalahari Desert, as did Williams et al. (2012) in the North American Mojave, who defined it as “bio-poor.” However, Garcia-Pichel et al. (2003) showed that zones under the topcrust are still significantly enriched in bacterial populations compared to bulk desert soil (Fig. 13.3). In cryptogam-dominated crusts, rhizines of specific crust organisms can easily reach into the subcrust and influence soil organic matter and stabilization there (Jimenez Aguilar et al. 2009). Beraldi-Campesi et al. (2009) demonstrated that a variety of subcrusts across different biogeographic provinces in North America differ significantly in geochemical character from the topcrust, the latter being enriched in biogenic elements (C, N, P). Both the topcrusts and the subcrusts are depleted in most other micronutrients and metals compared to neighboring uncrusted soils, speaking for the leaching effect of microbes on micronutrients, and for the necessity and role of allochthonous dust inputs to maintain crust fertility.

Fig. 13.3
figure 3

Biogeochemical differentiation in biocrusts as measured in dark cyanobacterial crusts from the Colorado Plateau. The depth (in mm) of the consolidated crust or topcrust is defined by the presence of large phototroph (primary producer) populations, revealed by high concentrations of chlorophyll-a (left panel). However, the influence on live microbial biomass (gauged by the concentration of DNA; middle panel) extends into, and defines the character of, the undercrust. Red vertical lines reflect the levels of DNA and chlorophyll-a in not crusted, bulk soil from the same area. The differential geochemical character of crust and undercrust is reflected in elemental concentrations (right panel), where the relative size of the font for each element denotes relative enrichment or depletion between crust and undercrusts. Data from Garcia-Pichel et al. (2003) and Beraldi-Campesi et al. (2009)

Full size image

When a crust becomes covered by sand, the organisms undergo severe stresses (Rao et al. 2012). Motile organisms can easily move upward to recolonize the new surface (Garcia-Pichel and Pringault 2001), as long as a certain threshold of sand burial depth is not exceeded. In laboratory experiments, Rao et al. (2012) found that a burial by less than 1 cm of sand has little long-term effect on the crust. Sometimes, in this manner, a new topcrust is created, leading to the buildup of a new subcrust, with perhaps mixed characteristics (Fig. 13.4 and see also Fig. 13.2, where the subcrust shows evidence of an old matrix of cyanobacterial sheaths). Deeper in the soil, ancient buried crusts can sometimes be found (Malam Issa et al. 2009; Drahorad and Felix-Henningsen 2013; Felde et al. 2014), in some cases even recurrently, forming layered sediments (Beraldi-Campesi and Garcia-Pichel 2010). Recently some of these layered fossilized biosurfaces have been recovered from terrestrial sediments as old as 1.2 billion years (Beraldi-Campesi et al. 2014; see Chap. 3 by Beraldi-Campesi and Retallack).

Fig. 13.4
figure 4

Evolution of pores and biopores. (a) Vesicular horizons develop in the subcrust of cyanobacterial biocrusts of the Sonoran Desert (Beraldi-Campesi et al. 2014). (b) Computer tomography cross section of a cyanobacterial biocrust from the Negev with two vesicular micro-horizons. In addition to the unconnected vesicles that are known to impede water movement, alternating layers of fine and coarse grains cause pore discontinuities, which create a capillary barrier effect that further reduces water infiltration (Felde et al. 2014). (c) Computer tomography cross section of a lichen-dominated crust from the Negev. Increased surface roughness results in a higher dust-trapping efficiency and therefore higher amount of silt and clay. This, combined with the high shrink-swell activity of the lichens, causes expansion-contraction during wetting-drying cycles and ultimately leads to the creation of shrinkage cracks. These cracks can vary in their orientation from horizontally to vertically, which is likely to have different impacts on water and gas fluxes in the crust

Full size image

Pore formation seems to be a significant structural trait in many biocrusts. Williams et al. (2012), working in the Mojave Desert, reported that the separation of a crust into two layers coincided with a planar void, likely generated by anastomosis of vesicular pores. These subsurface vesicular pore systems (or vesicular horizons) are common in crusts from various regions (Miralles-Mellado et al. 2011; Badorreck et al. 2013; Beraldi-Campesi et al. 2014; Felde et al. 2014). Unless actively destroyed (e.g., by root growth), they do preserve well in the fossil record and can be used as indicators for paleoenvironmental conditions and “markers” of paleo-crusts (Beraldi-Campesi et al. 2014). Vesicular layers can further decrease infiltration of a crusted soil surface (Hillel et al. 1998). Dietze et al. (2012) conducted comprehensive laboratory experiments that shed light on the genesis of vesicular pores, pointing to a combination of repeatedly descending wetting fronts with a soil surface that is puddled by fines. This sealed surface, due to the water table on top of the puddled surface, leads to an increased gas pressure and decreased inter-grain connection in the soil, which ultimately results in the formation of spherical, isolated pores. There exists a positive correlation between the size and sphericity of the pores with wetting rates and sand content, but a negative correlation with CaCO3 content. A necessary biotic influence of microbial metabolic gases can be effectively ruled out because vesicles also formed in sterilized soil (Dietze et al. 2012). For a more detailed description of the connection between soil crusting and vesicular pore formation, see also Chap. 9 by Colesie et al. and Felde et al. (2014).

Compelling evidence has been presented that EPS play a role in the architecture of biocrusts. Malam Issa et al. (2009) report that EPS can enhance the formation of pores and affect crust geometry. EPS clearly control the formation of microbially induced sedimentary structures (MISS) like roll-ups, folds, desiccation polygons, etc., particularly in non-cohesive substrates like sandy soils (Beraldi-Campesi et al. 2014). Rossi et al. (2012) showed that a simple nondestructive extraction of biocrust extracellular polysaccharides resulted in a much more compacted, “caked up” form that lost significant albedo and the ability to absorb water, all expected outcomes of a collapse of the pore system.

2.2 Small-Scale Spatial Structure of Biological Components

Organisms are not spread homogeneously within biocrusts. As was the case with abiotic components, and while lateral patterns (i.e., patchiness) clearly exist, most generalizations can be made regarding the small-scale vertical distribution of organismal biomass and composition.

Examining Colorado Plateau cyanobacterial crusts at millimeter resolution, Garcia-Pichel et al. (2003) found that both bacterial populations and DNA concentrations were tenfold larger in the top centimeter than in deeper, bulk soils. But the maxima in DNA concentration were clearly within the topcrust, either in the surface or the shallow subsurface (1–2 mm), depending on the crust type. Analyses of community structure along the vertical dimension, based on DNA sequencing and microbial community fingerprinting, showed clearly that community composition varied as well, pointing to the presence of differentiated niches within the crust.

Most descriptions of organismal spatial distribution in biocrusts involve the phototrophs, which are visible and largely recognizable by microscopy. Mostly as a consequence of the steep vertical gradients in light and, in arid environments, water availability within the crust, photosynthetic organisms are confined to the topcrust that they help delimit. This is usually only one to a few millimeters deep. Garcia-Pichel et al. (2003) reported that 75 % of all extractable chlorophyll-a can be found in the top 2 mm in Colorado Plateau crusts. But even within this euphotic zone, there is clear evidence of niche partitioning (Elliott et al. 2014). The very surface of the crusts represents a zone of light trapping, where, because of intense light scatter by mineral particles, the irradiance can be significantly higher than that incident down-welling from the outside (Garcia-Pichel 1995). The first colonizers of biocrusts in dryland regions (filamentous cyanobacteria and sometimes green algae) avoid it, and tend to establish subsurface populations below it, a few hundred microns to a few mm deep (Fig. 13.1b, d). Here the light intensities are moderate, there is little chance of abrasion by moving sand, and desiccation happens less quickly. This gives rise to a subsurface “organic” layer with a largely mineral layer on top. Crusts at this stage are very cryptic and typically referred to as light crusts in the literature. Examples of this are abound in the literature from a variety of locale and climatic regions, for cyanobacteria and eukaryotic populations (Hu et al. 2003; Pringault and Garcia-Pichel 2004; Smith et al. 2004; Zhang 2005; Zhang et al. 2006; Beraldi-Campesi et al. 2009; Chen et al. 2009; Wu et al. 2011; Lan et al. 2012; Rajeev et al. 2013).

Many of these subsurface populations of filamentous cyanobacteria (Microcoleus vaginatus , Microcoleus steenstrupii , Trichocoleus (formerly Microcoleus) sociatus , Oscillatoria spp. ) undergo vertical migrations to the surface when the crust gets wet in the dim light of overcast, rainy skies, returning to their subsurface refugium by sensing impending drought, a unique feature among microbes (Garcia-Pichel and Pringault 2001; Pringault and Garcia-Pichel 2004; see videos in Rajeev et al. 2013). Because these cyanobacteria leave a trail of EPS as a mechanism for motility, the movement becomes an efficient form of “weaving” minerals together, and the consolidation encompasses not just their subsurface resting area but the entire topcrust (Fig. 13.1e). Even at this early stage, empty sheaths and EPS left at the surface probably contribute to the dust-trapping ability of early crusts.

The light-trapping surface (0–200 μm deep) is typically only colonized by cyanobacteria that produce large quantities of sunscreens, as in the genera Scytonema , Tolypothrix , Nostoc , Porphirosiphon, etc. (Garcia-Pichel et al. 2001; Hu et al. 2003; Yeager et al. 2007). Most of these cyanobacteria are nonmotile and require the soil to be stabilized before they can thrive, being secondary colonizers after the initial motile filamentous forms (Garcia-Pichel 2002). Their advent during maturation increases a crust’s nitrogen-fixing capacity (Yeager et al. 2007) and lowers significantly soil albedo, due to the high content of scytonemin sunscreens. The top surface can also be colonized by aerophytic cryptogams (lichens and mosses), but these components do not occur everywhere.

Very detailed (submillimeter scale) vertical distributions of cyanobacterial and algal components of crusts from Shapotou (Tengger Desert, China) based on microscopic observations have been presented (Hu et al. 2003; Wu et al. 2011; Lan et al. 2012). These indicate that further vertical organization is possible, depending on the particular locale, but many of the components described in these studies do not seem to be of widespread distribution. Interestingly, they also report a preference of nematodes and protozoa for the topcrusts.

Analogously to cyanobacterial and algal biocrusts, Wu et al. (2011) also observed a layering of the biotic components in lichen-dominated crusts below the lichen thalli, which extend upward from the crust surface (Fig. 13.5). This included a top 1 mm layer with typical phototrophs, underlain by a layer dominated by lichen rhizoids between 1 and 2 mm, and a subcrust below it (subrhizoid layer, down to 6–8 mm). Within moss-dominated crusts, Lan et al. (2012) also distinguished three different layers, i.e., an upper “stem-leaf” layer (~0–2 mm), a rhizoid layer (~2–6 mm), and a subrhizoid layer (~6–15 mm). Pedicellation into mounds in microbiotic crusts appears to produce distinctive microhabitats that result in organismal differential distributions and differential organismal cover (George et al. 2000). This is particularly conspicuous for lichens and mosses. Presumably milder micro-aspects (north-north-west, east-north-east, and top) supported greater lichen and moss cover than the warmer, windward, and more xeric micro-aspects (west-south-west and south-south-east (Bowker et al. 2006).

Fig 13.5
figure 5

Biological succession in biocrusts (depth ~1 cm), conceptualized progressing from left to right, for North America crusts, according to Garcia-Pichel (2002). Early colonization by bundle-forming cyanobacteria (depicted as green filaments, with their sheaths in gray) stabilizes the soil, contributes to the creation of a layer of fines, and enables secondary colonization by heterocystous cyanobacteria (brownish globules and cylinders on soil surface), with eventual colonization by lichens or mosses. Aridity limits the end-point climax of this succession, and the time of recovery to each successive stage increases exponentially

Full size image

There have been few recent studies on the microbial distributions of non-photosynthetic microorganisms in biocrusts (Steven et al. 2013; Elliott et al. 2014; Maier et al. 2014; see Chap. 5 by Maier et al.). In a first investigation on that topic, Johnson et al. (2005) investigated aerobic chemolithotrophic ammonia-oxidizing bacteria, which build populations covering up to 2.5 mm in the suboxic confines of the active crusts. Here they find optimal “rain down” of ammonium released by cyanobacteria, moderate but sufficient oxygen, and a refuge from intense surface light intensities, to which they are sensitive. Their activity, in turn, contributes to creating a maximum of nitrate concentration at this depth (Johnson et al. 2007). Surprisingly, there are to date no studies on the small-scale distribution of heterotrophic microbes (bacteria, archaea, or fungi) beyond the anecdotal report of Garcia-Pichel et al. (2003), although one could envision a variety of potential niche differences regarding oxygen, pH, light (Garcia-Pichel and Belnap 1996), nutrients (Johnson et al. 2007; Beraldi-Campesi et al. 2009), and organic carbon (Baran et al. 2015).

3 Temporal Dynamics: Biological Succession, Structural Maturation, and Weathering

3.1 Biological Succession

Long-term studies on biocrust changes over time are rare. Therefore a “space-for-time” approach has often been applied to infer biocrust temporal dynamics. This “space-for-time” approach includes biocrust types of different perceived developmental stages, in most cases separated into cyanobacteria-dominated crusts (incipient or “light” and more mature or “dark”), lichen crusts, and moss crusts, often assuming that, given sufficient time, crusts will develop in such sequence (see Chaps. 9 and 23 by Colesie et al. and Weber et al., respectively). This is not always the case, given that in many pristine unaltered environments, terminal succession crusts can be lichen, cyanobacteria, or moss dominated (also see Chap. 23 by Weber et al.). Whenever lichen or moss crusts do develop, it is likely that they are preceded by one or more cyanobacterial- or algal-dominated stages. A generalization of the initial stages of biological succession in unconsolidated soils is presented in Fig. 13.5, based largely on results from crusts in Western North America. Loose mineral soil is initially colonized by pioneer cyanobacteria, which are typically motile, filamentous cyanobacteria that form supra-filamentous aggregations into bundles of ropes (Garcia-Pichel and Wojciechowski 2009; Fig. 13.1). They colonize the subsurface, and because of their large size, the rope-shaped organisms stabilize soil particles that come in contact with the extracellular sheaths. Migrations help weave a web of extracellular sheath trails that stabilizes the soil surface. This stage is what is typically referred to as “light crusts” in the literature. Once the surface is stabilized, secondary colonization by other organisms is possible. At depth, we find a variety of other cyanobacteria, which nevertheless remain typically less abundant than the pioneer Microcoleus spp. The most significant change occurs close to the surface with the advent of populations of heterocystous, nonmotile cyanobacteria. The three most common clades of such cyanobacteria are Nostoc sp. , Tolypothrix sp. , and Scytonema (Yeager et al. 2007). The advent of these heterocystous cyanobacteria brings about significant abiotic changes in albedo and crust surface temperature, that may, in turn, further affect the crust community composition (Couradeau et al. 2016). In a third stage, cryptogamic populations (lichens and/or mosses) may come in. Their presence brings about changes in the original populations as their thalli, external to the soil, can significantly shade the crusts interior, and as their rhizoids and protonemata penetrate and rework the topcrust and subcrust, thereby altering the structure and geometry of the pore system (Felde et al. 2014), which can drastically alter gas and water fluxes in and through the crust. Similar successions have been described from the Asian deserts (Hu et al. 2003; Zhang et al. 2006). These studies emphasize also the associated changes in visible organization into layers that logically follow the biological succession. Moss thalli constitute, in turn, a habitat for specialized populations of epiphytic cyanobacteria, which are relevant to crust fertility (Veluci et al. 2006).

3.2 Structural Development

During ongoing succession of Gurbantunggut Desert crusts, a transition from uncrusted soils to “light crusts” resulted in significant structural maturation, including the formation of a surface layer dominated by aeolian fines, with a dense and tightly packed matrix over an “organismal” layer dominated by cyanobacteria filaments, devoid of fines and developing large voids (Zhang et al. 2006). Studies carried out in the SE Tengger Desert of China showed that not only the vertical stratification was altered with crust successional stage but also that these changes were accompanied by a general increase in thickness and porosity and therefore a decrease in bulk density (Lan et al. 2012). And yet, typically biocrusts of older development stages tend to show increased penetration resistance. Thomas and Dougill (2007) found an almost threefold increase from 55 to 147 kPa with maturation; Guo et al. (2008) report a much more moderate difference: 125 kPa for algal crust vs. 168 kPa for a moss-dominated crust.

In cyanobacterial crusts from the NW Negev Desert of Israel, the porosity of both topcrust and subcrust changes with development. These crusts develop a vesicular micro-horizon and pore discontinuities associated with changes of pore and grain size (Fig. 13.4), which in turn causes a capillary barrier effect that reduces water infiltration . These crusts do show low infiltration compared to the bare soil or more mature crusts (Yair et al. 2011). With ongoing development, lichen crusts establish that develop shrink-swell dynamics caused by wetting cycles of lichen thalli and increased clay contents. This leads to the formation of shrinkage cracks (Fig. 13.4). The number of biopores increases at this stage, due to the growth of annual plants (roots) and lichen rhizoids, leading to increased water infiltration capacity, unless the crust is significantly detached from the soil underneath (Miralles-Mellado et al. 2011). In the successional stages of the Negev crusts, moss crusts show the highest abundance of biologically formed pores. In addition, moss rhizoids and protonemata pierce the complete crust, creating pathways for preferential infiltration and changing the geometry of the pore system from a tortuous to a straight pore network, while increasing pore diameters. The initial layering of coarse and fine grains, which was present in cyanobacterial crusts, disappears from these older crusts, likely as a consequence of active bioturbation. Within this ecosystem, cyanobacterial subcrusts show the highest porosity due to vesicular structures, with the topcrust porosity increasing steadily toward fully formed moss crusts (Felde et al. 2014).

3.3 Weathering and Geochemical Alteration

Investigations of biocrust weathering patterns, mechanisms, and rates are in their infancy. What we can predict to be relevant comes largely from geomicrobial interactions that have been studied in similar environments, i.e., endolithically within rocks, or on temporarily desiccated marine benthic environments. These comparisons have clearly their limits, since the lithic habitat is much more prone to weathering/erosional processes and biocrusts, at the opposing end of the geological spectrum, form on sedimentary settings that are the very results of such processes of weathering and erosion. A crust’s main geological role is its stabilizing, erosion-preventing effect on the sedimentary bed it covers. Given the accumulation of relatively large populations of microbes within the confines of a small space, however, biocrusts promote the internal formation of chemical microenvironments that can deviate significantly from the bulk values of the soil (Garcia-Pichel and Belnap 1996). Such microenvironments are thought to be one of the driving forces for increased rates of weathering on the abiotic components of biocrusts, and so the process of weathering cannot be understood but in the light of microstructural considerations. Beraldi-Campesi et al. (2009) surveyed the overall composition of a variety of soil crusts in North America with the specific aim to test the presence of chemical leaching as a process of relevance within biocrusts. The approach they took was to compare the concentrations of a set of 25 elements between crusts and immediately adjacent, uncrusted soils. They also analyzed the concentrations in the crusts and subcrusts. The survey revealed that crusts expectedly enriched the soil with biogenic elements (C, N, P) and generated leaching for many elements, significantly so for several important metals and metalloids (Ca, Cr, Mn, Zn, As, Zr). This leaching effect included the subcrust, indicating that the export of these mineral constituents and important micronutrients extended beyond the reaches of the crust. The authors argued that this leaching effect explains the necessity for allochthonous (dust) inputs to maintain the fertility of biocrusts in the long run.

Some of the physical weathering mechanisms ascribed to cryptogams and microbes in the lithic environment, such as the expansion and contraction of biomass in crevices or pores during wetting and drying (Chen et al. 2000), are clearly less relevant in sedimentary settings, but some of the mechanisms of biochemical weathering can be assumed to potentially work in a similar manner in the clastic milieu of the soil. These include the excretion of oxalic acid, the generation of respirative CO2 causing an acidification of the surrounding medium, and the production of biochemical compounds with complexing ability (Chen et al. 2000). In addition, two mechanisms of potential relevance are the alkalization of the substrate caused by CO2 consumption and the net OH− increase in the medium (Büdel et al. 2004) and the dissolution of carbonates caused by an active intracellular transport of free Ca2+ away from the weathering front and into the medium (Garcia-Pichel 2006).

Generally, silicate substrates are resistant to solubilization by organic acids and acid volatiles like CO2. Only some minerals, like the micas, phlogopite, and muscovite, are prone to acid hydrolysis (Serstevens et al. 1978). Bachmann (1904) described the preferential growth of endolithic lichens between the muscovite mica lamellae of granite. This euendolithic growth was restricted to granites containing mica, whereas in others, organisms were delimited to endolithic growth within fine preexisting cracks or crevices. The exfoliation of micas in granite by growth of chasmoendolithic cyanobacteria has been described by de los Rios et al. (2003, 2007). These mechanisms of weathering could potentially impact the mineralogy of soil crusts wherever micas form a substantial mineral component. Beraldi-Campesi and Garcia-Pichel (2010), however, did not find micas to be important silicates in a variety of cyanobacteria soil crust settings, in which silicate clay composition was in fact not altered by the presence/activity of large cyanobacterial populations in those crusts. In young crusts developing in temperate soils impacted by mining, Fischer et al. (2010) found no measurable weathering on silicates after 3 years of development. By contrast, Chen et al. (2009) found a general trend of decrease in primary silicate (K-feldspar) and an opposing increase in secondary hydrous micas along a full-scale space-for-time crust succession, particularly in crusts dominated by lichens and mosses. This is consistent with the well-known role of fungi in driving silicate mineral transformations.

In porous siliceous rocks , organisms form a continuous layer parallel to the rock surface at a few millimeters depth, including lichens (Friedmann et al. 1988; Wessels and Schoeman 1988) and free-living cyanobacteria (Friedmann et al. 1967; Eichler 1981; Friedmann 1982; Bell et al. 1986; Wessels and Büdel 1995; Weber et al. 1996). For the endolithic habitats in South Africa and Antarctica, exfoliative weathering mechanisms along the growth zone of the organisms were observed (Friedmann 1982; Büdel et al. 2004). Büdel et al. (2004) demonstrated that exfoliation results from an alkalization of the substrate by cyanobacteria, which causes both dissolution of silicate in the binding material and mineral grains and precipitation of carbonate crystals. The phenomenon of silicate dissolution by alkali was also predicted to occur in crusts (Garcia-Pichel and Belnap 1996), since photosynthesis-driven alkalization can drive local pH microenvironments above ten, but direct demonstrations are still lacking.

Carbonate rocks and carbonate sediments are colonized and weathered by a variety of microorganisms, such as cyanobacteria, algae, fungi, and lichens. Some grow euendolithically, creating a habitat by the active dissolution of the mineral (e.g., Bachmann 1913; Danin et al. 1983). For heterotrophs, it is thought that the mechanism consists of direct carbonate dissolution by acidification through metabolic CO2 production (Golubić et al. 1979). Also autotrophic organisms can temporarily show a net release of CO2 when respiration prevails, which may happen quickly enough to cause dissolution of the substrate (Weber et al. 2011). A new mechanism, by which autotrophs (algae and cyanobacteria) are able to dissolve carbonates, has been proposed by Garcia-Pichel (2006). There is now good evidence that euendolithic cyanobacteria bore by taking up Ca2+ ions from the medium selectively at the boring front, thus promoting dissolution of calcium carbonates there, transporting the ions intracellularly and excreting them at the opposing end (Garcia-Pichel et al. 2010; Ramirez-Reinat and Garcia-Pichel 2012). It now remains to be studied if this mechanism is restricted to cyanobacteria or if it also occurs in eukaryotic phototrophs. Again, in this case there is no evidence that any of these weathering mechanisms are of relevance in soil crusts developing in carbonate-rich soils. A review of molecular tallies of cyanobacteria in a variety of soil crusts, in fact, does not yield any typical euendolithic phylotypes. If any organisms are of importance, it is likely to be lichens, which could contribute to carbonate dissolution through acidification of the medium. Other organisms known to contribute significantly to carbonate dissolution in limestone through the production of nitric acid are the nitrifying microbes. These organisms are widespread and numerous in biocrusts (Marusenko et al. 2013).

4 Conclusion

Biological soil crusts (biocrusts) are organo-sedimentary systems in which both the organic and the inorganic mineral components play dynamic roles in determining the architecture and evolution of the system, as they interact between themselves and with the physical environment. We critically reviewed advances in the description of the microstructure of biocrusts with respect to their abiotic and biological components, as well as the interactions between the two in time and space that result in important properties of environmental relevance. We paid special attention to the processes of crust biological and physical succession and to mineral weathering processes.