Abandoned farmlands are increasing due to socio-economic changes and land marginalization, and they require sustainable land management practices. Biocrusts are a common cover on the topsoil of abandoned farmlands and play an important role in improving soil stability and erosion resistance. The critical functions of biocrusts are known to mostly rely on their biofilaments and extracellular polymeric substances (EPS), but how these components act at microscopic scale is still unknown, while rheological methods are able to provide new insights into biocrust microstructural stability at particle scale. Here, bare soil and two representative types of biocrusts (cyanobacterial and moss crusts) developed on sandy (Ustipsamments) and sandy loam (Haplustepts) soils in abandoned farmlands in the northern Chinese Loess Plateau were collected at a sampling depth of 2 cm. Changes in the rheological properties of the biocrusts were analyzed with respect to their biofilament network and EPS contents to provide possible explanations. The rheological results showed that compared with bare soil, storage and loss moduli were decreased by the biocrusts on sandy soil, but they were increased by the biocrusts on sandy loam soil. Other rheological parameters tau max, gamma L, gamma YP, and Iz of biocrusts on both soils were significantly higher than those of bare soil, showing higher viscoelasticity. And the moss crusts had about 10 times higher rheological property values than the cyanobacterial crusts. Analysis from SEM images showed that the moss crusts had higher biofilament network parameters than the cyanobacterial crusts, including nodes, crosslink density, branches, branching ratio and mesh index, and biofilament density, indicating that the biofilament network structure in the moss crusts was more compact and complex in contrast to the cyanobacterial crusts. Additionally, EPS content of the moss crusts was higher than that of the cyanobacterial crusts on both soils. Overall, the crosslink density, biofilament density, and EPS content of the biocrusts were significantly and positively correlated with their gamma YP and Iz. The interaction between crosslink density and biofilament density contributed 73.2 % of gamma YP, and that between crosslink density and EPS content contributed 84.0 % of Iz. Our findings highlight the biocrusts-induced changes of abandoned farmland soil rheological properties in drylands, and the importance of biocrust biofilament network and EPS in maintaining abandoned farmland soil microstructural stability to resist soil water/wind erosion and degradation, providing a new perspective for sustainable management of abandoned farmlands.

Background and aims Vascular plants and moss biocrusts are known to coexist in drylands, wherein vascular plant cover is known to be a major influencing factor for biocrusts development. Vascular plants produce litter which may affect moss biocrusts when covering them. However, to which extent the cover of litter may affect the physiology, e.g., photosynthetic activity, of moss biocrusts remains poorly understood.MethodsWe studied the effect of the litter covering on biocrust-forming mosses on the northern Chinese Loess Plateau over four-month period. We used litter from shrubs of Artemisia ordosica and Caragana korshinskii with two levels of litter thickness, and monitored moss greenness, and several indicators of moss physiological activity.ResultsLitter covering reduced moss greenness, content of chlorophyll a and b, soluble sugar, and soluble protein, suggesting a reduced photosynthetic and metabolic activity of mosses under litter cover. On the other hand, mosses covered by litter showed higher contents of malondialdehyde, proline, and catalase activity compared to those mosses without any litter cover, suggesting that litter covering increased oxidative stress in mosses and triggered a protective response against oxidative damage. Moreover, we found litter thickness exerted a more significant impact on the physiological indices of mosses than litter type.ConclusionsOur results demonstrate the detrimental effects of litter covering on the physiological activity of biocrust-forming mosses. The findings provide a mechanistic understanding of the reductions in mosses in ecosystems with high shrub cover, highlighting the importance of litter in mediating the relationships between moss biocrusts and shrub patches.

Soil respiration is one of the dominant fluxes of CO2 from terrestrial ecosystems to the atmosphere. Accurate quantification of soil respiration is essential for robust projection of future climate variation and for reliable estimation of paleoatmospheric CO2 levels using soil carbonates. Soil-respired CO2, which is the most uncertain factor in estimating atmospheric CO2 concentration, has been calculated from modern observations of surface soils and from proxy indicators of paleosols formed during time periods of known atmospheric CO2. However, these estimations provide a wide range of S(z) values from past to present. To directly compare modern observation with past estimation, here we first monitored soil CO2 profiles in a Holocene profile on the western Chinese Loess Plateau (CLP) for two years, providing direct measurements of soil-respired (CO2 )at the depth where carbonate nodules likely formed. We then collected carbonate nodules below last interglacial paleosol (S1) from two N-S-aligned transects across the CLP to back-calculate soil-respired CO2. The mean back-calculated S(z) from S1 carbonate nodules vary from 539 +/- 87 ppm to 848 +/- 170 ppm in the sections on the northwestern and southeastern CLP, respectively. The mean value of directly measured soil-respired CO2 on the western CLP is 572 + 273 ppm before the onset of summer monsoon, consistent with the back-calculated S(z) in northwestern sections. Our results suggest that spatial S(z) variations are mainly controlled by monsoonal precipitation during the summer season on the CLP. To better constrain the high end of S(z), more monitoring work is needed in higher precipitation areas on the southeastern CLP.