Ea

Ya

b

Sc

d

a

ARRA

KBCESSS

1

citoaotBlBm

EG

h0

Applied Soil Ecology 80 (2014) 6–14

Contents lists available at ScienceDirect

Applied Soil Ecology

jo u r n al homep ag e: www.elsev ier .com/ locate /apsoi l

ffects of biological soil crusts on soil enzyme activities in revegetatedreas of the Tengger Desert, China

anmei Liua,b,∗, Hangyu Yangc, Xinrong Lib, Zisheng Xingd

School of Life Science and Chemistry, Tianshui Normal University, Tianshui 741001, ChinaShapotou Desert Research and Experiment Station, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy ofciences, Lanzhou 730000, ChinaGansu Forestry Technological College, Tianshui 741020, ChinaFaculty of Forestry and Environmental Management, University of New Brunswick, PO Box 4400, 28 Dineen Drive, Fredericton, NB, Canada E3B 5A3

r t i c l e i n f o

rticle history:eceived 18 September 2013eceived in revised form 16 March 2014ccepted 27 March 2014

eywords:iological soil crustsrust typelapsed time since dune stabilizationeasonoil depth

a b s t r a c t

Biological soil crusts (BSCs) cover up to 70% of the sparsely-vegetated areas in arid and semiarid regionsthroughout the world and play a vital role in dune stabilization in desert ecosystems. Soil enzyme activ-ities could be used as significant bioindicators of soil recovery after sand burial. However, little is knownabout the relationship between BSCs and soil enzyme activities. The objective of this study was to deter-mine whether BSCs could affect soil enzyme activities in revegetated areas of the Tengger Desert. Theresults showed that BSCs significantly promoted the activities of soil urease, invertase, catalase anddehydrogenase. The effects also varied with crust type and the elapsed time since sand dune stabi-lization. All the soil enzyme activities tested in this study were greater under moss crusts than undercyanobacteria–lichen crusts. The elapsed time since sand dune stabilization correlated positively with thefour enzyme activities. The enzyme activities varied with soil depth and season, regardless of crust type.

oil enzyme activities Cyanobacteria–lichen and moss crusts significantly enhanced all test enzyme activities in the 0–20 cmsoil layer, but negatively correlated with soil depth. All four enzyme activities were greater in the summerand autumn than in spring and winter due to the vigorous growth of the crusts. Our study demonstratedthat the colonization and development of BSCs could improve soil quality and promote soil recovery indegraded areas of the Tengger Desert.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

Biological soil crusts (BSCs) are a sub-ecosystem or a micro-osm (Castillo-Monroy et al., 2011) and cover up to 70% of thenterspaces between sparse vegetation in semiarid and arid regionshroughout the world (Belnap, 1995). They are a complex mosaicf soil, green algae, lichens, mosses, micro-fungi, cyanobacteriand other bacteria (Belnap and Lange, 2003). They all developn the surface of stabilized sand dunes and vary from 2 mmhick, relatively homogeneous cyanobacteria crusts (Zaady andouskila, 2002; Li et al., 2011) to complex crusts dominated by

ichens and mosses that are up to 30 mm thick (Li et al., 2002).SC succession generally starts with colonization by large fila-entous cyanobacteria (such as Microcoleus sp.), which adhere

∗ Corresponding author at: Tianshui Normal University; Cold and Arid Regionsnvironmental and Engineering, School of Life Science and Chemistry, Tianshui,ansu province 741001, China. Tel.: +86 0938 2726327; fax: +86 0938 2726327.

E-mail address: lym-781118@163.com (Y. Liu).

ttp://dx.doi.org/10.1016/j.apsoil.2014.03.015929-1393/© 2014 Elsevier B.V. All rights reserved.

to soil particles by secreting gelatinous polysaccharide materials(Belnap, 2006; Zaady et al., 2010). Smaller pigmented cyanobacte-ria (such as Nostoc sp. and Scytonema sp.) and green algae theninvade the spaces within the filamentous cyanobacteria. Subse-quently, lichens and mosses may grow and colonize as the soilsurface stability increases and/or moisture availability improves(Eldridge and Greene, 1994; Kidron et al., 2008; Yu et al., 2012).BSCs have been found to play a number of important ecologicalroles in desert ecosystems, including enhancing soil aggregationand stability (Belnap, 1996; Guo et al., 2008), adjusting soil temper-ature and moisture (Belnap, 1995; George et al., 2003), improvingsoil aeration and porosity (Harper and Marble, 1988; Belnap et al.,2006), adjusting local hydrology (Evans and Johansen, 1999; Belnapand Lange, 2003), promoting vascular plant colonization (Zhaoet al., 2011) and improving soil invertebrate and microbial diver-sity (Darby et al., 2007, 2010; Neher et al., 2009; Liu et al., 2011,

2013).

Soil enzyme activity changes quickly if soil conditions alterand are, therefore, sensitive indicators of changes in soil quality(Puglisi et al., 2006; Trasar-Cepeda et al., 2008; Lebrun et al., 2012).

oil Eco

MeuacaeaZaatgw(aiaisopduBdB

riogac

2

2

gadmfowaad2sbcda((C

tbrbe

Y. Liu et al. / Applied S

oreover, it is relatively easy to determine the activity of soilnzymes. This means that soil enzyme activity has been widelysed to study the impacts of environmental changes and humanctivity on soil quality in recent years (Lebrun et al., 2012). Thelose relationship between BSCs and soil enzyme activities hasttracted increasing attention in recent years. For example, Wut al. (2009) and Zhao et al. (2010) suggested that nitrogenasectivity was influenced by the different BSC successional stages.hang et al. (2012) reported that catalase, urease, dehydrogenasend sucrase activities were greater in BSCs than in bare soil without

crust. Miralles et al. (2012) found that, in the Tabernas Desert,he activities of several hydrolase enzymes (i.e., arylsulphatase, ß-lucosidase, casein-protease, cellulase and phosphomonoesterase)ere greater in BSCs than in the bare substrate. Zelikova et al.

2012) measured the relationship between extracellular enzymectivities and BSCs as conditions grew warmer and precipitationncreased. Although the relationship between BSCs and soil enzymectivities has been studied, relatively little is known about thempacts that BSCs, crust type and the elapsed time since sand dunetabilization have on soil enzyme activities in revegetated areasf the Tengger Desert. In particular, in the Tengger Desert, whereeriods of high temperature are often accompanied by temporaryroughts, the spatio-temporal variations in soil enzyme activitiesnder BSCs have rarely been studied. Understanding the impact ofSCs on below-ground soil enzyme activities and spatio-temporalynamics could provide important information that may enhanceSC’s role in the recovery of degraded desert areas.

The aim of this study was to investigate the BSCs role in the soilecovery process through the promotion of soil enzyme activitiesn degraded desert areas. As a first step, we evaluated the impactf BSCs on soil enzyme activities in revegetated areas of the Teng-er Desert in northern China. Then we quantified the soil enzymectivity variation patterns under BSCs as the season and soil depthhanged.

. Materials and methods

.1. Site description

The study area is located at the southeast fringe of the Teng-er Desert in northern China (37◦32′N and 105◦02′E) and has anltitude of 1339 m. This area is a typical transitional zone fromesertified steppe to sand dunes, delineating oases and deserts. Theean annual precipitation is approximately 186 mm, 80% of which

alls between May and September. The mean annual potential evap-ration is about 3000 mm. The mean annual temperature is 10.0 ◦Cith a coldest mean monthly temperature of −6.9 ◦C in January

nd a warmest of 24.3 ◦C in July. A northwest wind prevails withn average wind velocity of 3.5 m s−1. The mean annual number ofays with dust storms caused by wind is approximately 59 (Li et al.,011). The soil is loose, infertile and mobile and can thus be clas-ified as orthic sierozem and aeolian sandy soil, with a very low,ut constant moisture content of 3–4% and a soil organic matterontent of 1.5–4.0 g kg−1 (Li et al., 2007b; Liu et al., 2011). The pre-ominant plants are semishrubs (Salsola passerine Bunge, Oxytropisciphylla Ledeb. and Ceratoides lateens Reveal et Holmgren), shrubsRemuria soongorica Maxim and Caragana korshinskii Kom.), forbsCarex stenophylloides Krecz) and grasses (Stipa breviflora Griseb andleistogenes songorica Ohwj) (Li et al., 2004).

To protect the natural desertified steppe from sand burial due tohe constant expansion of sand dunes in the Shapotou region, sand-

inding vegetation was first established along the Baotou–Lanzhouailway in 1956 by erecting 1 m × 1 m straw checkerboard sandarriers on shifting sandy surfaces. Parallel stabilized areas werexpanded one by one along the railway line in 1964, 1981 and 1991.

logy 80 (2014) 6–14 7

After the shifting sandy surfaces were stabilized, xerophytic shrubs(Artemisia ordosia Krasch, C. korshinskii Kom. and Caragana micro-phylla Lam.) were planted within the checkerboard and they reliedon rain water for growth. BSCs colonized and developed once thexerophytic vegetation was established. The initial cyanobacteria-dominated crusts remain the dominant crust type in areasrevegetated in 1991 and the late successional lichen and moss-dominated crusts are the dominant crust type in areas revegetatedin 1956 and 1964. In the study areas, moss crusts were 8–20 mmthick and green under wet conditions, while cyanobacteria–lichencrusts were 2–3.5 mm thick and dark brown or black with a clearsurface microtopography. Crusts cover more than 80% of the reveg-etated areas (Jia et al., 2008). The BSCs on the sand surface havealtered the physical and chemical properties of the topsoil, whichhas resulted in increases in the soil clay and silt contents, soil pH,electrical conductivity (EC), organic C, total N, total phosphorousand the thickness of soil crust and subsoil. Furthermore soil bulkdensity has decreased in revegetated areas compared to mobilesand dunes (Table 1; Li et al., 2007a; Liu et al., 2013).

2.2. Soil sample collection and preparation

Soil samples were collected in July 2011 from revegetated areasthat had been stabilized in 1956, 1964, 1981 and 1991. Mobile sanddunes served as the control. Within each revegetated area, fivesub-plots (10 m × 10 m), containing both cyanobacteria–lichen andmoss crusts, were established. There was at least 20 m betweentwo adjacent sub-plots. Meanwhile, five sub-plots (10 m × 10 m)without crusts were established in the mobile sand dunes andthese were also at least 20 m apart. Within each sub-plot, soilsamples from under the cyanobacteria–lichen crusts were col-lected from each soil layer at depths of 0–10 cm, 10–20 cm and20–30 cm using a 4 cm inner diameter × 10 cm depth core sampler.Each soil sample consisted of five soil sub-samples from under thecyanobacteria–lichen crusts, which were taken from the five sub-plots at the same soil depth. Soil samples from under the mosscrusts and from the mobile sand dunes were also collected usingthe same method as for the cyanobacteria–lichen crusts. In total,72 soil samples from under the crusts and nine sand dune soilsamples were used to analyze soil enzyme activities. To moni-tor seasonal variations in soil enzyme activities under the crusts,the areas revegetated in 1956 and natural vegetation areas in theHongwei of Shapotou area were chosen to represent the artifi-cially revegetated areas and natural vegetation areas, respectively.Five sub-plots (10 m × 10 m) containing both cyanobacteria–lichenand moss crusts, that were at least 20 m apart, were establishedin the areas revegetated in 1956 and in the natural vegetationareas. Soil samples from under the crusts were collected in April,July, September and December 2011 in areas revegetated in 1956and from the natural vegetation areas using the method describedabove. A total of 144 soil samples from under the crusts were col-lected over the different seasons and were used to analyze seasonalvariations in soil enzyme activities.

Each soil sample was placed in an individual plastic bag andtaken back to the laboratory. A 2 mm sieve was used to remove anylarge plant parts and stones before the samples were stored at 4 ◦Cuntil needed. Soil moisture content was determined gravimetri-cally using 20 g of field moist soil samples that had been oven-driedat 105 ◦C for 24 h.

2.3. Soil enzyme activities

Urease (EC 3.5.1.5) activity was assayed according to the methodused by Yang et al. (2007). A 5 g soil sample was incubated for24 h at 37 ◦C with 10 ml of 10% urea and 20 ml of citrate buffer(pH 6.7). It was then mixed with 1 ml methylbenzene for 15 min.

8 Y. Liu et al. / Applied Soil Ecology 80 (2014) 6–14

Table 1The physico-chemical properties of topsoil (0–10 cm) in revegetated areas, natural vegetation areas and mobile sand dunes.

Soil properties Natural vegetation areas Years after revegetation (y)

1956 1964 1981 1991 0 (Mobile sand dunes)

Sand (%) 13.54 66.39 68.28 71.54 78.87 99.67Silt (%) 72.00 22.59 24.79 23.59 15.60 0.12Clay (%) 14.45 11.01 6.93 4.87 4.45 0.21Organic carbon (g kg−1) 20.54 7.74 7.59 4.32 1.65 0.37Total nitrogen (g kg−1) 2.07 1.02 0.74 0.54 0.39 0.17Total phosphorous (g kg−1) 1.38 0.77 0.75 0.71 0.44 0.4Soil water content (%) 3.79 1.96 1.87 1.61 1.55 1.2pH 8.28 7.99 7.95 7.90 7.82 7.42EC (m s−1) 1.28 0.19 0.17 0.15 0.14 0.09Bulk density (g cm−3) 1.13 1.44 1.47 1.5 1.52 1.53

Ao4aua

mi5madAas5

mb0dtww

dioc2ecptTs

2

adwthtu

The thickness of soil crust and subsoil (cm) 4.87

fter incubation, the mixtures were immediately filtered and 1 mlf the supernatant was mixed with 10 ml of distilled water at 37 ◦C,

ml sodium phenate (1.35 M) and 3 ml sodium hypochlorite (0.9%ctive chlorine) for 20 min. The amount of N NH+

4 released fromrea hydrolysis was measured in the supernatant at 578 nm. Ureasectivity is expressed as �g N NH+

4 g−1 h−1.Invertase (EC 3.2.1.26) activity was assayed according to the

ethod used by Jin et al. (2009). Five gram soil samples werencubated for 24 h at 37 ◦C with 15 ml of 8% sucrose solution and

ml phosphate buffer (pH 5.5). They were then mixed with 0.2 mlethylbenzene for 15 min. The mixtures were immediately filtered

nd 1 ml of the supernatant was poured into a tube and 3 ml 3,5-initrosalicylic acid solution and 5 ml distilled water were added.ll the tubes were then placed in a boiling water bath for 5 min andllowed to cool to room temperature. The amount of glucose in theupernatant, as a result of the sucrose hydrolysis, was measured at08 nm. Invertase activity is expressed as �g glucose g−1 h−1.

Catalase (EC 1.11.1.6) activity was analyzed according to theethod used by Jin et al. (2009). Catalase activity was analyzed

y back-titrating residual H2O2 with KMnO4 and was expressed in.1 mol l−1 KMnO4 ml g−1 h−1. A 5 g soil sample was added to 40 mlistilled water and 5 ml 0.3% hydrogen peroxide solution. The mix-ure was oscillated at 25 ◦C for 20 min. After oscillation, the reactionas stopped by adding 5 ml of 1.5 mol l−1 sulfuric acid. The mixtureas filtered and titrated using 0.1 mol l−1 KMnO4.

Dehydrogenase (EC 1.1.1.1) activity was determined asescribed by Belén Hinojosa et al. (2004) with the following mod-

fication: a 15 g soil sample, which had been mixed with 0.15 gf CaCO3, was incubated with 1 ml 3% 2,3,5-triphenyltetrazoliumhloride (TTC) and 10 ml distilled water at 37 ◦C in the dark. After4 h, 10 ml methanol was added and the suspension was homog-nized, filtered and washed with methanol until the reddish coloraused by the reduced TTC (triphenylformazan, TPF) had disap-eared from the soil. The optical density at 485 nm was comparedo the TPF standards. Dehydrogenase activity is expressed as �gPF g−1 h−1. The enzyme assay was repeated three times for eachoil sample.

.4. Statistical analyses

Analysis of variance was used to evaluate whether soil enzymectivities differed between crust types, the elapsed time since sandune stabilization and season. A multivariate analysis of varianceas used to evaluate effects that the interactions between crust

ype, the elapsed time since sand dune stabilization and soil depthad on enzyme activities. The correlations between the elapsedime since sand dune stabilization and soil depth were examinedsing Pearson’s correlation coefficients. Differences obtained at the

2.5 2.2 1.4 0.72 0

p < 0.05 level were considered significant and SPSS 16.0 softwarewas used for the statistical analyses.

3. Results

3.1. Effects of BSCs on soil enzyme activities

BSCs significantly increased the soil urease, invertase, cata-lase and dehydrogenase activities in revegetated areas. Soil ureaseactivity was greater under the cyanobacteria–lichen and mosscrusts than in the mobile sand dunes for the 0–20 cm soil layer(p < 0.05) (Fig. 1A and B) and soil invertase activity was also greaterunder the cyanobacteria–lichen and moss crusts than in the mobilesand dunes for the 0–30 cm soil layer (p < 0.05) (Fig. 1C and D).Soil urease and invertase activities varied with crust type, theelapsed time since sand dune stabilization and soil depth (Table 2,p < 0.001). Moss crusts had greater soil urease and invertase activ-ities than cyanobacteria–lichen crusts and the effects of crust typeon their activities were interactively determined by the elapsedtime since sand dune stabilization and soil depth (Table 2, p < 0.01).The elapsed time since sand dune stabilization correlated positivelywith soil urease and invertase activities (r = 0.967 and r = 0.906,respectively, p < 0.05) and the effects of the elapsed time sincesand dune stabilization on their activities were also dependentupon crust type and soil depth (Table 2, p < 0.01). Soil urease activ-ity under cyanobacteria–lichen and moss crusts declined with soildepth (Table 2, p < 0.001) and had a negative correlation with soildepth (r = 0.956 and r = 0.970, respectively, p < 0.05). Soil invertaseactivity under cyanobacteria–lichen and moss crusts also declinedwith increasing soil depth (Table 2, p < 0.001) and had a signifi-cant negative correlation with soil depth (r = 0.954 and r = 0.909,respectively, p < 0.05).

Soil catalase and dehydrogenase activities (Fig. 1E and G) weregreater beneath the cyanobacteria–lichen crusts than in the mobilesand dunes for the 0–20 cm soil layer (p < 0.05). The soil catalase anddehydrogenase activities were also greater beneath the moss cruststhan in the mobile sand dunes for the 0–20 cm soil layer (p < 0.05)(Fig. 1F and H). The soil catalase and dehydrogenase activities var-ied with crust type, the elapsed time since sand dune stabilizationand soil depth (Table 2, p < 0.001). The elapsed time since sanddune stabilization correlated positively with soil catalase and dehy-drogenase activities (r = 0.908 and r = 0.956, respectively, p < 0.05)and the effect of the elapsed time since sand dune stabilization ontheir activities also depended on crust type and soil depth (Table 2,

p < 0.001). Catalase and dehydrogenase activities under the mosscrusts were greater than under the cyanobacteria–lichen crusts andthe effect of crust type on their activities also depended on theelapsed time since sand dune stabilization (Table 2, p < 0.001). Soil

Y. Liu et al. / Applied Soil Ecology 80 (2014) 6–14 9

s in re

cds(daga

3

ut

Fig. 1. Soil enzyme activities under BSC

atalase activity under the cyanobacteria–lichen and moss crustseclined as the soil depth increased (Table 2, p < 0.001) and theoil crusts had significant negative correlations with soil depthr = 0.979 and r = 0.993, respectively, p < 0.05). Similarly, soil dehy-rogenase activity under the cyanobacteria–lichen and moss crustslso declined as soil depth increased (Table 2, p < 0.001), which sug-ested that it had a negative correlation with soil depth (r = 0.903nd r = 0.893, respectively, p < 0.05).

.2. Seasonal variations in soil enzyme activities

The soil urease, invertase, catalase and dehydrogenase activitesnder the cyanobacteria–lichen and moss crusts changed withhe season in the 0–10 cm, 10–20 cm and 20–30 cm soil layers in

vegetated areas of the Tengger Desert.

both the revegetated and natural vegetation areas (Figs. 2 and 3).Soil urease and invertase activities under the cyanobacteria–lichenand moss crusts increased from April onwards, where the val-ues were at their lowest, and reached their maximum values insummer (July). There was a significant difference between sum-mer and the spring and winter months (p < 0.05). The valuesdropped back after July. They achieved their second highest val-ues in autumn (September) and their third highest values in winter(December) in the revegetated (Fig. 2A–D) and natural vegetationareas (Fig. 3A–D). The soil catalase and dehydrogenase activities

under the cyanobacteria–lichen and moss crusts were ranked, fromhighest to lowest, as summer (July), autumn (September), spring(April) and winter (December) in the three soil layers of the reveg-etated (Fig. 2E–H) and natural vegetation areas (Fig. 3E–H).

10 Y. Liu et al. / Applied Soil Ecology 80 (2014) 6–14

Table 2Statistical analysis of soil enzyme activities under cyanobacteria–lichen and moss crusts.

Source of variation Type III sum of squares df Mean square F

Soil urease activityCorrected model 936.881 29 32.306 112.086Crust type 78.932 1 78.932*** 273.854Soil depth 416.680 2 208.340*** 722.831The elapsed time since dune stabilization 246.647 4 61.662*** 213.935Crust type × soil depth 14.010 2 7.005*** 24.304Crust type × the elapsed time since dune stabilization 28.275 4 7.069*** 24.525Soil depth × the elapsed time since dune stabilization 138.650 8 17.331*** 60.131Crust type × soil depth × the elapsed time since dune stabilization 13.686 8 1.711*** 5.935

Soil invertase activityCorrected model 108466.367 29 3740.220 38.079Crust type 5695.419 1 5695.419*** 57.985Soil depth 47934.421 2 23967.210*** 244.010The elapsed time since dune stabilization 25980.595 4 6495.149*** 66.127Crust type × soil depth 9706.558 2 4853.279*** 49.411Crust type × the elapsed time since dune stabilization 1895.429 4 473.857** 4.824Soil depth × the elapsed time since dune stabilization 13627.354 8 1703.419*** 17.343Crust type × soil depth × the elapsed time since dune stabilization 3626.592 8 453.324*** 4.615

Soil catalase activityCorrected model 2.195 29 0.076 26.735Crust type 0.226 1 0.226*** 79.986Soil depth 0.918 2 0.459*** 162.098The elapsed time since dune stabilization 0.708 4 0.177*** 62.537Crust type × soil depth 0.013 2 0.007 2.381Crust type × the elapsed time since dune stabilization 0.085 4 0.021*** 7.469Soil depth × the elapsed time since dune stabilization 0.234 8 0.029*** 10.317Crust type × soil depth × the elapsed time since dune stabilization 0.011 8 0.001 0.477

Soil dehydrogenase activityCorrected model 90.958 29 3.136 253.646Crust type 5.169 1 5.169*** 418.025Soil depth 49.598 2 24.799*** 2.005E3The elapsed time since dune stabilization 10.128 4 2.532*** 204.767Crust type × soil depth 8.402 2 4.201*** 339.728Crust type × the elapsed time since dune stabilization 1.318 4 0.330*** 26.654Soil depth × the elapsed time since dune stabilization 14.220 8 1.777*** 143.745

2

4

necTBtm2asBNsta2tcC(taianef

Crust type × soil depth × The elapsed time since dune stabilization

** p < 0.01, *** p < 0.001.

. Discussion

Enzyme activities can be used to evaluate soil changes in bothatural and agro-ecosystems (Trasar-Cepeda et al., 2000; Puglisit al., 2006). BSCs significantly enhanced the soil urease, invertase,atalase and dehydrogenase activities in revegetated areas of theengger Desert, which was a similar result to Zhang et al. (2012).SCs may create a more favorable environment (suitable soilemperature and moisture) with higher soil nutrient and organic

atter levels (Belnap and Lange, 2003; Li et al., 2007a; Darby et al.,007, 2010; Zhao et al., 2011), which would increase soil enzymectivities (Acosta-Martínez et al., 2007; Brockett et al., 2012). Thisuggestion is supported by our previous finding that soil underSCs had 11.3 times more organic matter, 10.4 times more total, a 3.25-fold greater C:N ratio, 1.57 times more soil moisture and

howed a 1.13-fold decrease in daytime temperatures comparedo uncrusted soil in the study area (Li et al., 2011). Soil invertasend urease are crucial to soil C and N cycling, respectively (Ge et al.,010). Thus, the greater soil invertase and urease activities underhe crusts compared to the mobile sand dunes meant that BSCsould accelerate soil C and N turnover, which would enhance soil

and N cycling in desert ecosystems. This is supported by Li et al.2009), who found that BSCs were the dominant factors affectinghe input and exchange of C and N in desert ecosystems. Catalasend dehydrogenase are produced mainly by soil microbes and arenvolved in microbial oxidoreductase metabolism (Iyyemperumal

nd Shi, 2008). Therefore, the increased catalase and dehydroge-ase activities under BSCs in revegetated areas suggest that BSCsnhance soil microbial activities. The improved activities of theour enzymes under BSCs in our study may indicate that BSCs

.122 8 0.265*** 21.453

play a significant role in improving soil quality and promoting soilrecovery in degraded areas of the Tengger Desert.

Soil invertase and urease activities under late-stage moss crustswere greater than under early-stage cyanobacteria–lichen crustsand this indicated that late-stage moss crusts have a greaterability to increase C and N cycling compared to early-stagecyanobacteria–lichen crusts in revegetated desert areas. This resultwas consistent with previous findings that late-stage crusts couldimprove C and N inputs and exchange more than early-stage crusts(Li et al., 2009; Darby et al., 2010). Free-living cyanobacteria wereubiquitous in late-stage moss crusts due to the favorable condi-tions created by the mosses (Veluci et al., 2006). For example,we observed epiphytic cyanobacteria on late-stage moss crustsin revegetated areas of the Tengger Desert, which was similar toVeluci et al. (2006). Therefore, late-stage crusts that contain notonly dark-colored mosses and lichens, but also cyanobacteria, canfix more C and N than light-colored, early-stage crusts, which aredominated by light cyanobacteria (Housman et al., 2006; Groteet al., 2010; Zelikova et al., 2012). Soil catalase and dehydrogenaseactivities in late-stage moss crusts were greater than in early-stagecyanobacteria–lichen crusts and this suggests that soil microbialactivities were greater in the late-stage moss crust than in theearly-stage crusts. Similarly, Miralles et al. (2012) reported thatlate-stage crusts increased hydrolytic enzyme activities more thanearly-stage crusts. These findings reinforced previous findings thatthe soil under late-stage crusts had 1.26 times more organic mat-

ter and 1.88 times more total N than the soil under early-stagecrusts in revegetated areas (Li et al., 2011). There was also a posi-tive correlation between nutrient content and high enzyme activity(Aon and Colaneri, 2001; Acosta-Martínez et al., 2007; Zwikel et al.,

Y. Liu et al. / Applied Soil Ecology 80 (2014) 6–14 11

acteria

2ftwtVsdsrtssha

Fig. 2. Seasonal variations of soil enzyme activities under cyanob

007). In addition, late-stage crusts increased soil moisture 1.29-old, decreased daytime soil temperatures 1.15-fold and increasedhe pH 1.02-fold more than the early-stage crusts (Li et al., 2011),hich improved enzyme activities because soil temperature, mois-

ure and pH influence soil enzyme activities (Brockett et al., 2012).eluci et al. (2006) reported that BSCs moderate fluctuations inoil climate, in part, by reducing evaporation and by using theirark pigments to absorb solar radiation. Ecologists usually con-ider cyanobacteria to be primary colonizers that are successivelyeplaced or substituted by moss–lichen crusts. Our results suggesthat BSC succession stages could be identified by the changes in

ome enzyme activities because the late-stage crusts had greateroil enzyme activities than the early-stage crusts. Some reportsave suggested that early-stage crusts had greater soil invertasectivity than late-stage crusts (Miralles et al., 2012), which is in

–lichen crusts in revegetated areas and natural vegetation areas.

contrast to our findings. This may have been caused by differencesin crust species compositions, climate conditions and geographicalfactors etc.

The linear increase in soil urease, invertase, catalase anddehydrogenase activities as the elapsed time since sand dune sta-bilization increased suggested that soil C and N cycling and soilmicrobial activity under the crusts may be accumulated as theelapsed time since sand dune stabilization rose in revegetatedareas. These benefits to soil C and N cycling and soil microbialactivity led to improvements in soil quality. This is the first studyto show that four enzyme activities in our study correlated pos-

itively with elapsed time since sand dune stabilization in desertecosystems. A similar observation has been reported for restoredspoil heap ecosystems after mining of brown coal (Baldrian et al.,2008). Enzyme activities also generally increased with succession

12 Y. Liu et al. / Applied Soil Ecology 80 (2014) 6–14

moss

a1stcieiTe

td(t

Fig. 3. Seasonal variations of soil enzyme activities under

ge in 6–150 year old alpine glacier foreland soils (Ohtonen et al.,999). The improvement in soil enzyme activities as elapsed timeince sand dune stabilization increases may be due to the increasedhickness of the soil crust and subsoil. The crust thickening pro-ess that occurred as elapsed time since sand dune stabilizationncreased was concomitant with the accumulation of soil nutri-nts (i.e., organic C, total N and P; Table 1) and environmentalmprovements (more soil water, suitable temperature and pH).hese processes lead to an increase in soil enzyme activities (Lit al., 2005; Hamman et al., 2007; Kara et al., 2008).

The soil urease, invertase, catalase and dehydrogenase activi-

ies under the crusts showed a gradual decline with increasing soilepth and this was in agreement with the findings of Niemi et al.2005) and Enowashu et al. (2009). The significant differences inhe four enzyme activities between the crusted soils and mobile

crusts in revegetated areas and natural vegetation areas.

sand dunes in the 0–20 cm soil layers suggested that BSCs couldsignificantly promote the cycling of soil C and N and soil microbialactivity in the upper soil layers. In the 20–30 cm soil layer, althoughthe four enzyme activities under the crusts were greater than inmobile sand dunes, not all of the increases were significantly dif-ferent. A possible explanation is that the greater soil organic matter,nutrient contents, moisture and temperature retention in the topsoil (Jia et al., 2007; Liu et al., 2011) can support greater cycling ofsoil C and N and soil microbial activity, leading to an increase in thefour enzyme activities, whereas this was not happening to such agreat extent in the deeper soil layer (Fierer et al., 2003). Significant

interactions between the elapsed time since sand dune stabiliza-tion, soil depth and enzyme activities were found. This may implythat BSCs can enhance soil C and N cycling and soil microbial activ-ity in the top soil first (in the 0–20 cm soil layer), but the deeper soil

oil Eco

lbgvi

sttutb(pmYtbcvaTp(tg(titiaamtsTc

5

sretsctccbopduassmimf

e

Y. Liu et al. / Applied S

ayers only benefit later as the elapsed time since sand dune sta-ilization continues to increase. The result may indicate that BSCsradually improved soil quality and promoted soil recovery in aertical direction as the elapsed time since sand dune stabilizationncreased in revegetated areas.

As far as we can ascertain, this is the first study to indicate thatoil enzyme activities under crusts showed seasonal variations inhe sandy soils found in the Tengger Desert. These seasonal varia-ions in soil urease, invertase, catalase and dehydrogenase activitiesnder the crusts may be caused by climatic factors and soil proper-ies. While climatic factors, such as temperature and moisture haveeen identified as the leading factors affecting seasonal variationCriquet et al., 2004; Wittmann et al., 2004), the effects of tem-erature and moisture on enzyme activities may only be indirectlyediated by crust growth. In previous studies, Jin et al. (2009) and

ao et al. (2011) reported that plant growth can be an important fac-or that regulates seasonal variations in soil enzyme activities. Theetter sunlight conditions (776 h sunshine) and relatively high pre-ipitation (78.0 mm) in summer meant that crust growth was moreigorous and this led to increased aboveground biomass, covernd species composition than in the autumn, spring and winter.he vigorous crust growth and improved soil moisture and tem-erature created a more suitable habitat for soil microorganismsZwikel et al., 2007) and promoted soil enzyme activities underhe crusts in the summer compared to the other seasons. The crustrowth slowed down in autumn due to relatively low precipitation59.4 mm), low illumination (467 h sunshine) and low soil tempera-ures, which reduced soil microorganism and soil enzyme activitiesn autumn compared to summer. The low precipitation and lowemperature in spring and winter also restricted crust growth, lead-ng to reduced soil microorganism activities. Thus, soil enzymectivities decreased in spring and winter compared to summer andutumn. Our finding that enzyme activities under the crusts wereore sensitive to seasonal variations in the topsoil layer (0–10 cm)

han in the deeper soil layer was similar to previous reports aboutoils in other regions (Aon and Colaneri, 2001; Zwikel et al., 2007).he finding suggests that enzyme activities in the topsoil layerould be sensitive indicators of seasonal changes in crust growth.

. Conclusion

This study demonstrated that BSCs significantly increasedoil urease, invertase, catalase and dehydrogenase activities inevegetated areas of the Tengger Desert. Crust type and thelapsed time since sand dune stabilization significantly affectedhe four enzyme activities. All four enzymes demonstrated con-istently greater activities in late-stage moss than in early-stageyanobacteria–lichen crusts due to more suitable soil tempera-ures and moistures and improved soil organic matter and nutrientontents under the late-stage moss crusts compared to early-stageyanobacteria–lichen crusts. The elapsed time since sand dune sta-ilization correlated positively with the four enzyme activities inur study, which showed that, in revegetated areas, there was arogressive improvement in soil quality as elapsed time since sandune stabilization increased. In addition, the four enzyme activitiesnder the crusts also varied with soil depth and season in the studyreas. BSCs significantly promoted enzyme activities and improvedoil quality in the 0–20 cm soil layer. The enzyme activities alsohowed an apparent seasonal pattern. They were highest in sum-er, followed by autumn, and were lowest in spring and winter

n our study. The seasonal patterns in soil enzyme activities were

ainly due to the crust growth being affected by the way climatic

actors interact with crust development.The study demonstrated that BSCs significantly enhanced soil

nzyme activities in revegetated areas and promoted soil C and N

logy 80 (2014) 6–14 13

cycling and soil microbial activity. The colonization and develop-ment of BSCs could improve soil quality and benefit the recoveryof degraded areas in the Tengger Desert.

Acknowledgments

This research was supported by China National Funds forRegional Science (grant no. 41261014), a school project grant fromTianshui Normal University (grant no. TSA0923) and a grant fromthe China National Funds for Distinguished Young Scientists (grantno. 40825001).

References

Acosta-Martínez, V., Cruz, L., Sotomayor-Ramírez, D., Pérez-Alegría, L., 2007. Enzymeactivities as affected by soil properties and land use in a tropical watershed. Appl.Soil Ecol. 35, 35–45.

Aon, M.A., Colaneri, A.C., 2001. II. Temporal and spatial evolution of enzymatic activ-ities and physico-chemical properties in an agricultural soil. Appl. Soil Ecol. 18,255–270.

Baldrian, P., Trögl, J., Frouz, J., Snajdr, J., Valásková, V., Merhautová, V., Cajthaml, T.,Herinková, J., 2008. Enzyme activities and microbial biomass in topsoil layerduring spontaneous succession in spoil heaps after brown coal mining. Soil Biol.Biochem. 40, 2107–2115.

Belén Hinojosa, M., Carreira, J.A., García-Ruíz, R., Dick, R.P., 2004. Soil moisturepre-treatment effects on enzyme activities as indicators of heavy metal-contaminated and reclaimed soils. Soil Biol. Biochem. 36, 1559–1568.

Belnap, J., 1995. Surface disturbances: their role in accelerating desertification. Envi-ron. Monit. Assess. 37, 39–57.

Belnap, J., 1996. Soil surface disturbances in cold deserts: effects on nitrogenaseactivity in cyanobacterial-lichen soil crusts. Biol. Fertil. Soils 19, 362–367.

Belnap, J., 2006. The potential roles of biological soil crusts in dryland hydrologicalcycles. Hydrol. Processes 20, 3159–3178.

Belnap, J., Lange, O.L., 2003. Biological Soil Crust: Structure, Function and Manage-ment. Springer-Verlag, Berlin, pp. 3–30.

Belnap, J., Phillips, S.L., Troxler, T., 2006. Soil lichen and moss cover and speciesrichness can be highly dynamic: the effects of invasion by the annual exoticgrass Bromus tectorum, precipitation, and temperature on biological soil crustsin SE Utah. Appl. Soil Ecol. 32, 63–76.

Brockett, B.F.T., Prescott, C.E., Grayston, S.J., 2012. Soil moisture is the major factorinfluencing microbial community structure and enzyme activities across sevenbiogeoclimatic zones in western Canada. Soil Biol. Biochem. 44, 9–20.

Castillo-Monroy, A.P., Bowker, M.A., Maestre, F.T., Rodríguez-Echeverría, S., Mar-tinez, I., Barraza-Zepeda, C.E., Escolar, C., 2011. Relationships between biologicalsoil crusts, bacterial diversity and abundance, and ecosystem functioning:insights from a semi-arid Mediterranean environment. J. Veg. Sci. 22, 165–174.

Criquet, S., Ferre, E., Farnet, A.M., Le Petit, J., 2004. Annual dynamics of phosphataseactivities in an evergreen oak litter: influence of biotic and abiotic factors. SoilBiol. Biochem. 36, 1111–1118.

Darby, B.J., Neher, D.A., Belnap, J., 2007. Soil nematode communities are ecologicallymore mature beneath late- than early-successional stage biological soil crusts.Appl. Soil Ecol. 35, 203–212.

Darby, B.J., Neher, D.A., Belnap, J., 2010. Impact of biological soil crusts and desertplants on soil microfaunal community composition. Plant Soil 328, 421–431.

Eldridge, D.J., Greene, R.S.B., 1994. Microbiotic soil crusts: a view of their roles insoil and ecological processes in the rangelands of Australia. Aust. J. Soil Res. 32,389–415.

Enowashu, E., Poll, C., Lamersdorf, N., Kandeler, E., 2009. Microbial biomass andenzyme activities under reduced nitrogen deposition in a spruce forest soil. Appl.Soil Ecol. 43, 11–21.

Evans, R.D., Johansen, J.R., 1999. Microbiotic crusts and ecosystem processes. Crit.Rev. Plant Sci. 18, 183–225.

Fierer, N., Schimel, J.P., Holden, P.A., 2003. Variations in microbial community com-position through two soil depth profiles. Soil Biol. Biochem. 35, 167–176.

Ge, C.R., Xue, D., Yao, H.Y., 2010. Microbial biomass, community diversity, andenzyme activities in response to urea application in tea orchard soils. Commun.Soil Sci. Plan. 41 (7), 797–810.

George, D.G., Roundy, B.A., St Clair, L.L., Johansen, J.R., Schaalje, G.B., Webb, B.L., 2003.The effects of microbiotic soil crusts on soil water loss. Arid Land Res. Manage.17, 113–125.

Grote, E., Belnap, J., Housman, D., Sparks, J., 2010. Carbon exchange in biologicalsoil crust communities under differential temperatures and soil water contents:implications for global change. Global Change Biol. 16, 2763–2774.

Guo, Y.R., Zhao, H.L., Zuo, X.A., Drake, S., Zhao, X.Y., 2008. Biological soil crust devel-opment and its topsoil properties in the process of dune stabilization, InnerMongolia, China. Environ. Geol. 54, 653–662.

Hamman, S.T., Burke, I.C., Stromberger, M.E., 2007. Relationships between microbialcommunity structure and soil environmental conditions in a recently burnedsystem. Soil Biol. Biochem. 39, 1703–1711.

Harper, K.T., Marble, J.R., 1988. A role for nonvascular plants in managementof arid and semiarid rangelands. In: Tueller, P.T. (Ed.), Vegetationa Science

1 oil Eco

H

I

J

J

J

K

K

L

L

L

L

L

L

L

L

L

L

M

N

4 Y. Liu et al. / Applied S

Applications for Rangeland Analysis and Management. Kluwer Academic Pub-lishers, Dordrecht, The Netherlands, pp. 135–169.

ousman, D., Powers, H., Collins, A., Belnap, J., 2006. Carbon and nitrogen fixation dif-fer between successional stages of biological soil crusts in the Colorado Plateauand Chuhuahuan Desert. J. Arid Environ. 66, 620–634.

yyemperumal, K., Shi, W., 2008. Soil enzyme activities in two forage systems follow-ing application of different rates of swine lagoon effluent or ammonium nitrate.Appl. Soil Ecol. 38, 128–136.

ia, R.L., Li, X.R., Liu, L.C., Gao, Y.H., Li, X.J., 2008. Responses of biological soil cruststo sand burial in a revegetated area of the Tengger Desert, Northern China. SoilBiol. Biochem. 40, 2827–2834.

ia, X.H., Li, X.R., Li, Y.S., 2007. Soil organic carbon and nitrogen dynamics during there-vegetation process in the arid desert region. Chin. J. Plant Ecol. 31, 66–74 (inChinese).

in, K., Sleutel, S., Buchan, D., De Neve, S., Cai, D.X., Gabriels, D., Jin, J.Y., 2009. Changesof soil enzyme activities under different tillage practices in the Chinese LoessPlateau. Soil Tillage Res. 104, 115–120.

ara, Ö., Bolat, ì., Cakıroglu, K., Öztürk, M., 2008. Plant canopy effects on litter accu-mulation and soil microbial biomass in two temperate forests. Biol. Fertil. Soils45, 193–198.

idron, G.J., Vonshak, A., Abeliovich, A., 2008. Recovery rates of microbiotic crustswithin a dune ecosystem in the Negev desert. Geomorphology 100, 444–452.

ebrun, J.D., Trinsoutrot-Gattin, I., Vinceslas-Akpa, M., Bailleul, C., Brault, A., Mougin,C., Laval, K., 2012. Assessing impacts of copper on soil enzyme activities in regardto their natural spatiotemporal variation under long-term different land uses.Soil Biol. Biochem. 49, 150–156.

i, X.R., He, M.Z., Duan, Z.H., Xiao, H.L., Jia, X.H., 2007a. Recovery of topsoil physi-cochemical properties in revegetated sites in the sand-burial ecosystems of theTengger Desert, Northern China. Gemorphology 88, 254–265.

i, X.R., Jia, R.L., Chen, Y.W., Huang, L., Zhang, P., 2011. Association of ant nestswith successional stages of biological soil crusts in the Tengger Desert, NorthernChina. Appl. Soil Ecol. 47, 59–66.

i, X.R., Kong, D.S., Tan, H.J., Wang, X.P., 2007b. Changes in soil and in vegetationfollowing stabilisation of dunes in southeastern fringe of the Tengger Dersert,China. Plant Soil 300, 221–231.

i, X.R., Wang, X.P., Li, T., Zhang, J.G., 2002. Microbiotic crust and its effect on vegeta-tion and habitat on artificially stabilized desert dunes in Tengger Desert, northChina. Biol. Fertil. Soils 35, 147–154.

i, X.R., Xiao, H.L., Liu, L.C., Zhang, J.G., Wang, X.P., 2005. Long-term effect of sand-fixed vegetation on restoration of biodiversity in Shapotou region in TenggerDesert. J. Desert. Res. 25 (2), 173–181 (in Chinese).

i, X.R., Zhang, Y.M., Zhao, Y.G., 2009. A study of biological soil crusts: recent devel-opment, trend and prospect. Adv. Earth Sci. 24 (1), 11–24 (in Chinese).

i, X.R., Zhang, Z.S., Zhang, J.G., Wang, X.P., Jia, X.H., 2004. Association between veg-etation patterns and soil properties in the southeastern Tengger Desert, China.Arid Land Res. Manage. 18, 369–383.

iu, Y.M., Li, X.R., Jia, R.L., Huang, L., Zhou, Y.Y., Gao, Y.H., 2011. Effects of biologicalsoil crusts on soil nematode communities following dune stabilization in theTengger Desert, northern China. Appl. Soil Ecol. 49, 118–124.

iu, Y.M., Li, X.R., Xing, Z.S., Zhao, X., Pan, Y.X., 2013. Responses of soil microbialbiomass and community composition to biological soil crusts in the revegetatedareas of the Tengger Desert. Appl. Soil Ecol. 65, 52–59.

iralles, I., Domingo, F., Cantón, Y., Trasar-Cepeda, C., Leirós, M.C., Gil-Sotres, F.,

2012. Hydrolase enzyme activities in a successional gradient of biological soilcrusts in arid and semi-arid zones. Soil Biol. Biochem. 53, 124–132.

eher, D.A., Lewins, S.A., Weicht, T.R., Darby, B.J., 2009. Microarthropod communitiesassociated with biological soil crusts in the Colorado Plateau and Chihuahuandeserts. J. Arid Environ. 73, 672–677.

logy 80 (2014) 6–14

Niemi, R.M., Vepsäläinen, M., Wallenius, K., Simpanen, S., Alakukku, L., Pietola, L.,2005. Temporal and soil depth-related variation in soil enzyme activities and inroot growth of red clover (Trifolium pratense) and timothy (Phleum pratense) inthe field. Appl. Soil Ecol. 30, 113–125.

Ohtonen, R., Fritze, H., Pennanen, T., Jumpponen, A., Trappe, J., 1999. Ecosystemproperties and microbial community changes in primary succession on a glacierforefront. Oecologia 119, 239–246.

Puglisi, E., Del Re, A.A.M., Rao, M.A., Gianfreda, L., 2006. Development and validationof numerical indexes integrating enzyme activities of soils. Soil Biol. Biochem.38, 1673–1681.

Trasar-Cepeda, C., Leirós, M.C., Gil-Sotres, F., 2008. Hydrolytic enzyme activities inagricultural and forest soils, some implications for their use as indicators of soilquality. Soil Biol. Biochem. 40, 2146–2155.

Trasar-Cepeda, C., Leirós, M.C., Seoane, S., Gil-Sotres, F., 2000. Limitationof soil enzymes as indicators of soil pollution. Soil Biol. Biochem. 32,1867–1875.

Veluci, R.M., Neher, D.A., Weicht, T.R., 2006. Fixation and leaching of nitrogen bybiological soil crust communities in mesic temperate soils. Microb. Ecol. 51,189–196.

Wittmann, C., Kähkönen, M.A., Ilvesniemi, H., Kurola, J., Salkinoja-Salonen, M.S.,2004. Areal activities and stratification of hydrolytic enzymes involved in thebiochemical cycles of carbon, nitrogen, sulphur and phosphorus in podsolizedboreal forest soils. Soil Biol. Biochem. 36, 425–433.

Wu, N., Zhang, Y.M., Downing, A., 2009. Comparative study of nitrogenase activityin different types of biological soil crusts in the Gurbantunggut Desert, north-western China. J. Arid Environ. 73, 828–833.

Yang, C.L., Sun, T.H., Zhou, W.X., Chen, S., 2007. Single and joint effects of pesticidesand mercury on soil urease. J. Environ. Sci. 19, 210–216.

Yao, H.Y., Bowman, D., Shi, W., 2011. Seasonal variations of soil microbial biomassand activity in warm- and cool-season turfgrass systems. Soil Biol. Biochem. 43,1536–1543.

Yu, J., Kidron, G.J., Pen-Mouratov, S., Wasserstrom, H., Barness, G., Steinberger,Y., 2012. Do development stages of biological soil crusts determine activityand functional diversity in a sand-dune ecosystem? Soil Biol. Biochem. 51,66–72.

Zaady, E., Ben-David, E.A., Sher, Y., Tzirkin, R., Nejidat, A., 2010. Inferring biologicalsoil crust successional stage using combined PLFA, DGGE, physical and biophys-iological analyses. Soil Biol. Biochem. 42, 842–849.

Zaady, E., Bouskila, A., 2002. Lizard burrows association with successionalstages of biological soil crusts in an arid sandy region. J. Arid Environ. 50,235–246.

Zelikova, T.J., Housman, D.C., Grote, E.E., Neher, D.A., Belnap, J., 2012. Warmingand increased precipitation frequency on the Colorado Plateau: impli-cations for biological soil crusts and soil processes. Plant Soil 355,265–282.

Zhang, W., Zhang, G.S., Liu, G.X., Dong, Z.B., Chen, T., Zhang, M.X., Dyson, P.J., An, L.Z.,2012. Bacterial diversity and distribution in the southeast edge of the TenggerDesert and their correlation with soil enzyme activities. J. Environ. Sci. 24 (11),2004–2011.

Zhao, H.L., Guo, Y.R., Zhou, R.L., Drake, S., 2011. The effects of plantation developmenton biological soil crust and topsoil properties in a desert in northern China.Geoderma 160, 367–372.

Zhao, Y., Xua, M., Belnap, J., 2010. Potential nitrogen fixation activity of different aged

biological soil crusts from rehabilitated grasslands of the hilly Loess Plateau,China. J. Arid Environ. 74, 1186–1191.

Zwikel, S., Lavee, H., Sarah, P., 2007. Temporal dynamics in arylsulfatase enzymeactivity in various microenvironments along a climatic transect in Israel. Geo-derma 140, 30–41.