lable at ScienceDirect

European Journal of Soil Biology 74 (2016) 93e103

Contents lists avai

European Journal of Soil Biology

journal homepage: http : / /www.elsevier .com/locate/ejsobi

Original article

Biocrusts, inside and outside resource islands of Mimosa luisana(Leguminosae), improve soil carbon and nitrogen dynamics in atropical semiarid ecosystem

Ana Lidia Sandoval P�erez a, b, Sara Lucía Camargo-Ricalde a, *, No�e Manuel Monta~no a,Felipe García-Oliva b, Alejandro Alarc�on c, Susana Adriana Monta~no-Arias a,Manuel Esper�on-Rodríguez d

a Departamento de Biología, Divisi�on de Ciencias Biol�ogicas y de la Salud, Universidad Aut�onoma Metropolitana-Iztapalapa, A.P. 55-535, C.P. 09340, D. F.,M�exico, Mexicob Instituto de Investigaciones en Ecosistemas y Sustentabilidad, Universidad Nacional Aut�onoma de M�exico, campus Morelia, AP. 27-3, 58090, Morelia,Michoac�an, Mexicoc �Area de Microbiología, Colegio de Postgraduados, campus Montecillo, C.P. 56230, Montecillo, Estado de M�exico, Mexicod Laboratorio de Ecofisiología Tropical, Instituto de Ecología, Universidad Nacional Aut�onoma de M�exico, C.P. 04510, D. F., M�exico, Mexico

a r t i c l e i n f o

Article history:Received 22 October 2015Received in revised form11 February 2016Accepted 16 March 2016

Handling Editor: C.C. Tebbe

Keywords:Biological-soil-crustsLegumesSoil nutrientsSoil fertilitySoil biologyThorny-tropical scrub

* Corresponding author.E-mail address: slcr@xanum.uam.mx (S.L. Camarg

http://dx.doi.org/10.1016/j.ejsobi.2016.03.0061164-5563/© 2016 Elsevier Masson SAS. All rights res

a b s t r a c t

In the semiarid Valley of Zapotitl�an Salinas, Puebla, Mexico, biocrusts may be found inside Mimosa lui-sana Brandegee (Leguminosae)-resource islands (RI) or outside them (ORI). We studied the seasonalvariation of the effect of three microenvironments: i)M. luisana-RI þ biocrusts (Biocrusts-RI), ii) biocrustsoutsideM. luisana-RI (Biocrusts-ORI), and iii) Open areas (OA), on the soil C and N dynamics. In both rainyand dry seasons, moss species richness and cover were higher at Biocrusts-RI than in Biocrusts-ORI;opposite pattern to lichens. Soil organic C, labile C, and total N were the highest at Biocrusts-RI, inter-mediate at Biocrusts-ORI, and lowest at OA. This agrees with high microbial C and N, and C minerali-zation. We suggest that C availability regulated soil N availability under both Biocrusts-RI and Biocrusts-ORI by stimulating microbial biomass and N mineralization. Biocrusts-RI and Biocrusts-ORI did not differin soil NH4

þ and NO3� concentration, but N mineralization was higher at both microenvironments than in

OA in the dry season. In contrast, in the rainy season, nitrification was higher and decreased from OA,Biocrusts-ORI to Biocrusts-RI. It supports that both Biocrusts-RI and Biocrusts-ORI may be forming“mantles of fertility”, and highlight their functional role on microbial dynamics and N transformationslinked to changes in C availability, providing a hypothetical model for a better understanding of soilbiology within this tropical semiarid ecosystem.

© 2016 Elsevier Masson SAS. All rights reserved.

1. Introduction

Arid and semiarid ecosystems are highly heterogeneous due torainfall seasonality [1] and spatial variability created by the exis-tence of vegetation patches with different sizes and forms [2], andisolated plants creating resource islands (RI), which redistributesoil resources (i.e. soil nutrients, and water) and improve micro-environmental conditions [3]. Understanding how spatial andtemporal heterogeneity affect the functioning of these ecosystems

o-Ricalde).

erved.

is critical to formulate frame-works for their conservation, man-agement or restoration. These ecosystems cover ca. 41% of theglobal terrestrial surface, and 60% in Mexican territory, but arehighly disturbed [4].

Current research [5�10] have demonstrated that legumes assome Mimosa and Prosopis species form RI, whose soil containshigher amounts of organic material, which promotes the microbialactivity, enabling the soil nutrients availability compared to baresoil. In addition, the legume-RI also provides a more benignmicroenvironment that reduces temperature and increases hu-midity, promoting strong plant-soil-microorganisms feedbacks, aswell as the establishment of other plants under their canopies[3,5�7]. Recently, on the soil of legume-RI such as those formed by

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e10394

Mimosa luisana Brandegee, the presence of communities consti-tuted by cyanobacteria, lichens and mosses (biocrusts) has alsobeen observed; although, they can also be found outside Mimosaluisana-RI [11], where they also contribute to soil fertility, protec-tion against rainfall water run-off and wind erosion, as in other dryecosystems [12].

As it occurs in the legume-RI, but in a micro-scale, it has alsobeen shown that biocrusts are able to build a favorable soilmicroenvironment for other organisms; enhancing rainfall waterinfiltration and soil water availability [13], and improving soil car-bon (C) and nitrogen (N) dynamics [14�16,24]; hence increasingthe nutritional status of plants and microorganisms [12], andfacilitating seed germination [17�19]; consequently, promotingplant growth and establishment under environmental adverseconditions [12,19]. In Mexico, most of the studies have been focusedon species composition and distribution of biocrusts in relation tosoil in both undisturbed [11,20�22] and disturbed conditions[23,24], keeping on the side functional aspects; for instance, theeffect of biocrusts on biological C and N transformations in the soil.

It is already known that some Mimosa species-RI, includingM. luisana-RI, have a forestry insularity effect on soil nutrients thatis defined by a decreasing gradient from species-trunk soil, with thehighest C and N concentrations, towards species-mid-foliage coversoil/species-edge-foliage cover soil, to the soil of open areas withthe lowest concentrations [7]. This insularity gradient may also bedue to the relationship between M. luisana and biocrusts under thecanopy of this legume. In addition, this plant could be alteringbiocrust constituents in relation to areas outside RI, as it has beenreported in biocrusts under the canopy of others desert plants[11,13,25], changing thus nutrients cycling [14,15,26]. However,there are no studies focused on the functional role of bothM. luisana-RI and biocrusts on soil C and N dynamics in semiaridecosystems. In turn, we hypothesized that biocrusts insideM. luisana-RI would differ in species composition and cover frombiocrusts outside, so they may differentially influence the avail-ability and biological transformations of subsurface soil C and N, inrelation to open area soils, especially during the rainy season, whenbiocrusts are metabolically active. Actually, there is a critical gap inrelation to the understanding on how biocrusts could regulate theactivity of different soil microbial groups, impacting C and N dy-namics [1,12,15], especially in tropical semiarid ecosystems.

Therefore, the aim of this study was to determine the effect ofthree microenvironments: i) Biocrusts inside M. luisana-RI (Bio-crusts-RI), ii) biocrusts outsideM. luisana-RI (Biocrusts-ORI), and iii)Open areas (OA), on the availability-mineralization of soil C and N,in dry and rainy seasons, in the tropical semiarid Valley ofZapotitl�an Salinas, within the Tehuac�an-Cuicatl�an BiosphereReserve, Puebla, Mexico.

2. Material and methods

2.1. Study area

The Valley of Zapotitl�an Salinas (18�200N, 97�280W), Puebla,Mexico, is part of the Tehuac�an-Cuicatl�an Biosphere Reserve(Fig. S1). Climate is semiarid, with an average annual temperatureof 21 �C and an annual rainfall ranging from 400 to 600 mm [27].Soils are sandy-clay-loams (41% sand, 37% silt, and 22% clay), poorin its structure, mainly derived from sedimentary andmetamorphicrocks, and classified as Calcisols in the FAO/UNESCO system [28].The dominant vegetation type is the tropical thorny scrub, wherethemost abundant species areM. luisana, Prosopis laevigata (Humb.& Bonpl. ex Willd.) MC. Johnston, Parkinsonia praecox (Gonz�alez-Ruiz and Pav�on) Harms., Myrtillocactus geometrizans (Mart.) DC.,and Neobuxbaumia tetetzo (Weber) Backeb., among others [29].

M. luisana is a shrub of 1e4.5 m tall, with foliage cover up to 6 m2,endemic to the Tehuac�an-Cuicatl�an region, where 16 Mimosa spe-cies have been recorded [30]. This plant species is either dominantor co-dominant in plant communities [5], and is used by localpeople as wood fuel, construction or in traditional medicine; henceit is consider as a multipurpose species [31].

2.2. Sampling design

Within the Valley of Zapotitl�an Salinas, seven plots weredistributed in different locations: P1: 18� 180 42.000 N, 97� 320

32.100 W (1615 m asl); P2: 18� 180 16.400 N, 97� 320 32.300 W (1580 masl); P3: 18� 170 55.000 N, 97� 310 21.500 W (1578 m asl); P4: 18� 170

9.000 N,97� 290 52.700 W (1590 m asl); P5: 18� 190 44.600 N, 97� 270

24.100 W (1509 m asl); P6: 18� 190 36.900 N, 97� 270 20.700 W (1460 masl); and P7: 18� 170 55.000 N, 97� 310 21.500 W(1455m asl). Every plotmeasured 1000 m2 (20 m � 50 m); when possible, it was estab-lished in a flat site; otherwise, the slope was�10� (Fig. S1). Per plot,three independent microenvironments were selected: i) Biocrustsinside RI formed by M. luisana (2e3 m height, and 4e5 m foliagecover) (Biocrusts-RI), ii) Biocrusts outside RI (Biocrusts-ORI), and iii)Open areas, without plants or biocrusts (OA) (Fig. S2). Biocrusts,litter and soil samples were collected in two seasons: i) Dry(March), and ii) Rainy (September) on points randomly distributedat each plot.

Biocrusts samples were collected of the microenvironmentsBiocrusts-RI and Biocrusts-ORI, at 15 points per plot, and they wereplaced in sterile Petri dishes sealed with parafilm. Litter wascollected on an area of 345 cm2, at nine points of each microenvi-ronment, per plot. Litter was placed in paper bags and stored atroom temperature. Soil samples were taken at 0e3 cm depth, aftercollecting the Biocrust or directly at the OA, in 15 points, permicroenvironment; within each plot, 15 subsamples were mixed toform a composite sample, per each microenvironment, per plot,which were stored in black plastic bags, refrigerated at 10 �C, andtransported to the laboratory.

2.3. Biocrusts species composition and cover

In laboratory, mosses, lichens and cyanobacteria were taxo-nomically identified [32]. Species richness for each biocrust wasregistered in terms of the observed species. A dissimilarity index bycluster analysis (Ward's method with Euclidean distances) for thespecies composition of mosses, lichens and cyanobacteria betweenmicroenvironments (Biocrusts-RI and Biocrusts-ORI) and seasonswas used. This index is based on the presence-absence data, beingequal to 0% under complete similarity, and 100% if there are notshared species [33]. Cover of the three main biotic components ofthe biocrusts (cyanobacteria, lichens and mosses) was determinedusing a 25 cm2 square (5 cm � 5 cm), centered on each collectedsample. This square was divided into a grid of 0.5 cm � 0.5 cm, andthe cover of each biotic-crust component was evaluated by thepoint sampling method [13].

2.4. Litter mass

Litter samples were passed through two sieves (2 mm and1 mm) to eliminate sand, carcasses and soil aggregates; litter infresh weight was registered. Samples were oven-dried at 70 �C toconstant weight to estimate dry mass. To include the fine litter thatis mixed with soil, a litter sub-sample was ground to 450 mm andcombusted at 600 �C/4 h in a muffle to estimate dry mass of littersoil free.

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103 95

2.5. Soil physical properties

Prior to sampling, soil temperature was recorded in situ using athermometer (OBH Nordic 4770) at nine points by microenviron-ment within each plot. Temperature data was gathered around 10a.m. to 19 p.m. (local time). The composite soil samples were passedthrough a 2 mm sieve. A soil sub-sample was oven-dried at 75 �C toconstant weight for soil moisture determination by the gravimetricmethod. Soil pH was measured in deionized water (soil:solutionratio, 1:2 ratio w/v) using a pH meter (Corning). Soil electricalconductivity (EC) was measured in a suspension with a ratio of 1:5w/v in deionized water and a conductivity meter (Model CL30).Only in the dry season, the soil bulk density (depths: 0e3 and0e10 cm) was determined using the tube method, as well as soiltexture by the Bouyoucos hydrometer method.

2.6. Total and available soil C and N

All soil C forms were determined with a Total Carbon Analyzer(TOC, Mod. CM5012) [34], while soil N forms were determinedcolorimetrically using a Bran-Luebbe Auto Analyzer III (Norder-stedt, Germany). Soil was dried and ground in an agate mortar andpestle prior to total C and N analyses. Total C and inorganic C weredetermined by combustion and coulometric detection in the TOC[34]. Thus, soil organic C was obtained by the difference betweentotal C and inorganic C. Total N was determined after an aciddigestion by the macro-Kjeldahl method [35], and measuredcolorimetrically with the auto-analyzer [36]. Available forms of Cand N were measured in fresh soil samples. Hence, labile C wasextracted with 0.5 M K2SO4 and determined by total and inorganicC with the TOC analyzer (see above). The difference between theseC forms provided the organic labile C, here used as an index of Cavailability. Ammonium (NH4

þ) and nitrate (NO3�) were extracted

using 2 M KCl, followed by filtration through Whatman No. 1 filters[37], and determined with the auto-analyzer colorimetrically bythe phenol-hypochlorite method [36].

2.7. Microbial biomass, potential C mineralization and net Ntransformations

Microbial C and N were determined in fresh soil samples by thechloroform fumigation-extraction method [38]. Fumigated andnon-fumigated samples were incubated during 24 h at 25 �C, andconstant moisture. Microbial C was extracted from both fumigatedand non-fumigated samples using 0.5M K2SO4, and passed throughWhatmanNo. 42 filters [38]. Total and inorganic C from soil extractswere measured in the TOC analyzer (see above), and organic C wasestimated by the difference between these two C forms. Microbial Cwas calculated by subtracting the extracted C in non-fumigatedsamples from that of fumigated samples and dividing it by a KEC

value of 0.45 [38]. Microbial N was extracted with the same pro-cedure used for microbial C, but was filtered through a WhatmanNo. 1 paper. The filtrated was acid digested by the macro-Kjeldahlmethod, and determined as total N as was previously described.Microbial N was calculated similar to microbial C, but divided by aKEN of 0.54 [39].

Potential C mineralization, net N mineralization and net nitri-fication were measured in 21-day laboratory aerobic incubations[37]. Soil fresh samples were placed in PVC (polyvinyl-chloride)tube cores, with a 0.250 mm mesh at the bottom. Each sample waswetted to field water holding capacity with deionized water andincubated in 1 L jars at 25 �C, which were regularly aerated andadjusted in soil moisture every two days. Potential C mineralizationwas estimated as evolved CO2eC collected in 1 M NaOH traps.Carbonates were precipitated by adding 1.5 M BaCl2, and titrated

with 1 M HCl. Furthermore, before and after the incubation, bothNH4

þ and NO3� were determined with the previously described

methods. Differences between post- and pre-incubation values ofinorganic N (NH4

þ, plus NO3�) and NO3

� were used to calculate Nmineralization and nitrification, respectively [37].

2.8. Statistical analysis

Data were expressed on a dry weight basis, and subjected to arepeated measures analysis of variance (RMANOVA, P � 0.05) withone between-subject factor (microenvironment), and two within-subject factors (season and microenvironment-season interac-tion). When the RMANOVA indicated significant effects, meancomparisons were performed with a Tukey's HSD test. A one-wayANOVA for soil bulk density and texture, in dry season, was alsoperformed, where the main factor was the microenvironment. Inboth cases, data were log-transformed to meet ANOVA assump-tions [40]. A principal components analysis (PCA) was used toexplore the interdependence of all soil variables and their re-lationships with microenvironments and biocrusts cover. In thePCA, a matrix Pearson correlation was included to explorer thesignificance of each variable. A stepwise multiple-regression anal-ysis was also conducted to examine the control on soil C and Ndynamics under laboratory incubations during the two seasons. Allanalyses were performed with Statistica 6 software, and a P � 0.05was taken to be significant.

3. Results

3.1. Species composition and cover of biocrusts

Ten species of mosses, eight of lichens and seven of cyanobac-teria were identified. The maximum number of species was recor-ded in the dry season, where the Biocrusts-RI microenvironmentshowed higher species richness of mosses, lichens and cyanobac-teria, than the Biocrusts-ORI (Table 1). The highest index ofdissimilarity corresponded to both Biocrusts-RI and Biocrusts-ORIduring the dry season (66%). In contrast, the lowest dissimilarity(12%) between these microenvironments was found in the rainyseason (Fig. 1). The microenvironment affected moss cover;whereas lichens cover was affected by microenvironment � seasoninteraction (Table 2). Thus, mosses cover was higher in Biocrusts-RIthan in Biocrusts-ORI; while lichens cover was greater in Biocrusts-ORI than in Biocrusts-IR for both sampling seasons, and only inBiocrusts-ORI it decreased from the dry to the rainy season. Incontrast, cyanobacteria cover did not vary between microenviron-ments or seasons (Table 2; Fig. 2).

3.2. Litter mass

Litter mass was affected by microenvironment � season inter-action (Table 2). The highest litter mass was obtained on theBiocrusts-RI during the dry season; whereas, Biocrusts-ORI and OAhad similar litter mass. However, in the rainy season, litter mass onBiocrusts-RI and Biocrusts-ORI was similar; whereas, the lowestlitter mass was recorded on OA (Table 3).

3.3. Soil physical properties

Soil moisture in the dry season was not different among mi-croenvironments; however, during the rainy season, soil moisturewas the highest at Biocrusts-RI, intermediate at Biocrusts-ORI, andthe lowest at the OA. Soil pH was less alkaline at the Biocrusts-RIthan at the Biocrusts-ORI or OA; differing also between seasons,soil pH was more alkaline in the rainy season than in the dry one.

Table 1Species composition of mosses, lichens and cyanobacteria species reported from each biocrust studied, within twomicroenvironments: i)M. luisana-RIþ biocrusts (Biocrusts-RI), and ii) Biocrusts outside M. luisana-RI (Biocrusts-ORI), during dry and rainy seasons, at the Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�an Biosphere Reserve, Puebla,Mexico.

Season/Microenvironment

Dry Rainy

Biocrusts-RI Biocrusts-ORI Biocrusts-RI Biocrusts-ORI

MossesAloina hamulus (C. Muell.) Broth x x x xBrackymenium exile (Dosz et Molk.) Bosch et Lac x x xBrackymenium sp. Schwaegr x x x xBryum argenteum Hedw xDidymodon rigidulus var. gracilis (Hook. Grev.) Zand x xNeohyophyla sp.Pseudocrossidium aureum (Bartr.) Zand x x x xPseudocrossidium replicatum (Tayl.) Zand x xTrichostomum brachydontium Bruch ex F. Muell xWeissia sp. Hedw xLichensGloeoheppia sp. x x xLempholemma sp. x x xLichinella sp.1 x xLichinella sp. 2 x x x xPeltula euploca (Ach.) Poelt x xPeltula patellata (Bagl.) Swinscow & Krog x x xPlacidium laciniatum (Ach.) Breuss x x x xPsora crenata (Taylor) Reinke x x x xCyanobacteriaAnabaena sp. x xChroococcus turgidus (Kutzing) N€ageli xMicrocoleus chthonoplastes (Mertens) Zanardini x x xOscillataria sp. xOscillatoria agardhii (Gomont) xSchizothrix sp. xScytonema javanicum (Kütz) Bonet ex Born et Flah x x

Fig. 1. Dissimilarity index by cluster analysis (Ward's method and Euclidean distances) showing differences in composition of mosses, lichens and cyanobacteria species of bio-crusts, in two microenvironments: i) M. luisana-RI þ biocrusts (Biocrusts-RI), and ii) Biocrusts outside M. luisana-RI (Biocrusts-ORI); in the dry and rainy seasons, at the Valley ofZapotitl�an Salinas, Tehuac�an-Cuicatl�an Biosphere Reserve, Puebla, Mexico.

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e10396

Soil electrical conductivity (EC) was higher at the Biocrusts-RI andOA microenvironments than at the Biocrusts-ORI, decreasing fromthe dry to the rainy season. Soil bulk density at 0e3 cm(1.43 g cm�3) and 0e10 cm (0.91 g cm�3), as well as soil texture

(56% sand, 25% silt, and 18% clay), were not modified by themicroenvironment. Furthermore, the highest soil temperature insituwas recorded at the Biocrusts-ORI and OA microenvironments;while Biocrusts-RI had the lowest temperature in both seasons; this

Table 2F-ratios and significance levels of the repeated measures analysis, RMANOVA (P� 0.05), for soil properties at three microenvironments: i)M. luisana-RIþ biocrusts (Biocrusts-RI), ii) Biocrusts outsideM. luisana-RI (Biocrusts-ORI), and iii) Open areas (OA), as a control; in the dry and rainy seasons, at the Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�anBiosphere Reserve, Puebla, Mexico.

Variables Source of variation M � S

Between subject Within subjects

Microenvironment (M) Season (S)

F P F P F P

Soil physical propertiespH (1:2) 8.8 <0.01 880.7 <0.0001 1.0 0.38Electrical Conductivity 4.1 <0.05 12.6 <0.01 1.9 0.17Moisture 4.6 <0.05 376.7 <0.0001 5.4 <0.01Temperature 8.6 <0.01 24.3 0.0001 3.0 0.07Litter mass and soil nutrientsLitter mass 5.3 0.01 16.5 <0.001 7.1 <0.01Total C 0.5 0.6 0.03 0.8 1.0 0.3Inorganic C 0.1 1.0 0.5 0.5 1.9 0.2Organic C 4.4 <0.05 0.01 0.9 1.9 0.2Total N 14.4 <0.001 0.7 0.4 7.5 <0.01Labile C 5.1 <0.01 59.1 <0.0001 2.6 0.1NH4

þ 2.2 0.1 275.3 <0.0001 1.3 0.3NO3

� 3.0 0.07 19.3 <0.001 1.8 0.2Microbial nutrients and processesMicrobial C 7.9 <0.001 6.5 <0.05 0.07 0.93Microbial N 9.9 <0.001 25.3 <0.0001 1.9 0.17Microbial C:N ratio 5.2 <0.01 21.5 <0.0001 1.2 0.32Potential C mineralization rate 9.3 0.001 6.08 <0.05 1.5 <0.05Net N-mineralization 1.7 0.20 10.65 <0.004 3.6 <0.05Net nitrification 0.32 0.73 8.74 <0.008 2.1 <0.05Biocrusts coverMosses 55.4 <0.0001 0.5 0.5 0.04 0.8Lichens 47.4 <0.0001 39.1 <0.0001 4.8 <0.05Cyanobacteria 3.5 0.09 1.1 0.3 2.4 0.1

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103 97

trend was also showed along the day (Tables 2 and 3; Fig. S3).

3.4. Total and available forms of soil C and N

Soil total C and inorganic C did not vary among microenviron-ments or seasons. Soil organic C varied only among microenvi-ronments (Table 2), thus in both seasons it decreased fromBiocrusts-RI to OA, with an intermediate concentration at theBiocrusts-ORI (Table 3). Soil total N varied depending on themicroenvironment� season interaction (Table 2). For both seasons,Biocrusts-RI showed the highest N concentration, and it increasesfrom the dry to the rainy season. In contrast, in the dry season, bothBiocrusts-ORI and OA had similar soil N concentrations; however,in the rainy season, soil N concentration was higher at Biocrusts-ORI than at OA (Table 3).

3.5. Available and microbial forms of soil C and N

Soil labile organic C was the highest at Biocrusts-RI, interme-diate at Biocrusts-ORI, and the lowest at OA in both seasons, and itdecreased from the dry to the rainy season. Concentrations of NH4

þ

and NO3� were affected by seasonality. NH4

þ concentration washigher in the dry season than during the rainy season. In contrast,NO3

� concentration was lower in the dry season than in the rainyseason (Tables 2 and 4). Soil microbial C and N were the highest inBiocrusts-RI, intermediate at Biocrusts-ORI and the lowest at OA;however, microbial C decreased from the dry to the rainy season;while microbial N increased from the dry to the rainy season. Thus,microbial C:N ratio was the highest at OA in both seasons; however,in the rainy season, it decreased at the three microenvironments(Tables 2 and 4).

3.6. Soil C and N mineralization, and nitrification

Soil C and N transformations were affected bymicroenvironment � season interaction (Table 2). In the incubatedsamples of soil collected at Biocrusts-RI, potential C mineralizationwas higher than those collected at Biocrusts-ORI and OA, in the dryseason. Additionally, in the rainy season, the soil collected atBiocrusts-RI and Biocrusts-ORI had similar and higher potential of Cmineralization during the incubation than the one collected at OA(Fig. 3a; Fig. S4). Net N mineralization was lower at OA than atBiocrusts-RI and Biocrusts-ORI, in the dry season; whereas, in therainy season, N mineralization did not vary among microenviron-ments. Merely, Biocrusts-RI and Biocrusts-ORI had higher Nmineralization in the dry season than in the rainy one (Fig. 3b). Netnitrification, in incubated samples of soil at Biocrusts-ORI and OA,was higher than at Biocrusts-RI, in the rainy season; whereas, in thedry season, the nitrification did not vary among microenviron-ments, but it was higher in the rainy season than in the dry one(Fig. 3c).

3.7. Relationships among biocrusts, labile nutrients, and microbialprocesses

In the principal components analysis, the two axes accountedfor 61.4% of the variation in the data (Fig. 4). This indicated that allthe studied microenvironments were different between both sea-sons. In dry season, Biocrusts-RI were separated from bothBiocrusts-ORI and OA, being associated to a higher lichens cover,microbial C, labile C, NH4

þ and N mineralization than the other twomicroenvironments. Meanwhile, in rainy season, Biocrusts-RI andBiocrust-ORI were similar and separated fromOA, being related to ahigher moss cover, moisture, NO3

�, litter, total N, microbial N, and Cmineralization (Fig. 4). Likewise, the correlation coefficients of the

Fig. 2. Seasonal variation on the cover of mosses, lichens and cyanobacteria of bio-crusts collected in two microenvironments: i) M. luisana-RI þ biocrusts (Biocrusts-RI),and ii) Biocrusts outside M. luisana-RI (Biocrusts-ORI); during dry and rainy seasons, atthe Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�an Biosphere Reserve, Puebla, Mexico.Bars with different uppercase letters (A, B, C) indicate that means are significantlydifferent (P � 0.05) among microenvironments, within a same sampling season.Different lowercase letters (a, b, c) indicate that means are significantly different(P � 0.05) between sampling seasons, within a same microenvironment.

Table 3Soil physical properties and nutrient content seasonal means (± standard error),within three microenvironments: i) M. luisana-RI þ biocrusts (Biocrusts-RI), ii)Biocrusts outsideM. luisana-RI (Biocrusts-ORI), and iii) Open areas (OA), as a control;in the dry and rainy seasons, at the Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�anBiosphere Reserve, Puebla, Mexico.

Microenvironment

Biocrusts-RI Biocrusts-ORI OA

Soil physical propertiespH (1:2 in water)Dry season 7.9 (0.03)Bb 8.1 (0.03)Ab 8.1 (0.05)Ab

Rainy season 8.5 (0.02)Ba 8.6 (0.04)Aa 8.7 (0.04)Aa

Electrical Conductivity (mS/cm)Dry season 222.7 (16.1)Aa 194.4 (15.3)Ba 245.4 (50.1)Aa

Rainy season 191.6 (10.7)Ab 180.0 (18.0)Bb 197.5 (15.8)Ab

Moisture (%)Dry season 1.61 (0.18)Ab 1.32 (0.12)Ab 1.44 (0.15)Ab

Rainy season 19.52 (1.1)Aa 17.93 (1.3)ABa 14.58 (1.2)Ba

Temperature in situ (�C)Dry season 26.7 (1.2)Aa 38.1 (3.1)BCa 41.6 (3.6)Ca

Rainy season 24.1 (0.9)Ab 27.6 (1.8)Bb 29.7 (1.9)Bb

Litter mass and soil nutrientsLitter mass (g m�2)Dry season 105.7 (12)Ab 42.9 (6)Bb 46.1 (5)Bb

Rainy season 121.2 (28)Aa 125.2 (23)Aa 77.0 (14)Bb

Total C (mg C g�1)Dry season 54.9 (9.2) 50.6 (7.8) 47.5 (8.1)Rainy season 57.7 (9.2) 53.2 (7.8) 42.9 (8.1)

Inorganic C (mg C g�1)Dry season 33.1 (5.8) 32.2 (6.8) 31.6 (7.3)Rainy season 32.1 (7.2) 37.0 (7.3) 32.6 (8.7)

Organic C (mg C g�1)Dry season 21.8 (4.2)A 18.4 (1.9)AB 16.0 (1.7)B

Rainy season 25.8 (3.5)A 17.3 (2.1)B 12.9 (2.4)C

Total N (mg N g�1)Dry season 2.8 (0.6)Ab 1.6 (0.3)Bb 1.3 (0.2)Bb

Rainy season 3.4 (0.4)Aa 1.9 (0.4)Bb 0.8 (0.1)Cb

Values followed horizontally by different uppercase letter (A, B, C) indicate thatmeans are significantly different (P� 0.05) amongmicroenvironments, within samesampling season; whereas, values followed vertically by a different lowercase letter(a, b, c) indicate that means are significantly different (P � 0.05) between samplingseasons, within same microenvironment.

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e10398

principal components analysis are shown in Table S1, highlightingthat mosses, lichens and cyanobacteria covers were positivelyrelated to microbial C and N. Moreover, mosses cover is positivelycorrelated with soil moisture, labile C, and C mineralization; whilelichens cover is positively correlated with soil temperature, littermass, labile C, and N mineralization. In addition, organic and labileC had a positive correlationwith total N, microbial C, as well as to C

and N mineralization; although, labile C was negatively correlatedto NO3

� and nitrification (Table S1). Thus, a stepwise multipleregression analysis was realized for each season, showing that, inthe dry season, C mineralization was positively related to labile C,microbial C, and NH4

þ; though, in the rainy season, C mineralizationwas positively related to microbial N and NO3

� (Table 5). N miner-alization showed a positive relation to labile C and microbial C inboth seasons. Furthermore, nitrification was positively related tolabile C in dry season and to NH4

þ, andmicrobial C and N in the rainyseason (Table 5).

4. Discussion

Mimosa luisana improves the fertility of the soil under its can-opy, creating, as well, a benign microenvironment for many otherplants and microorganisms, generating both fertile and resourceislands within the semiarid Valley of Tehuac�an-Cuicatl�an [5,7].Further, biocrusts are found in both M. luisana-RI (Biocrusts-RI) oroutside them (Biocrusts-ORI), but with different assemblages andfunctions depending on rainfall seasonality, hence modifying soilbiology under its influence.

The Biocrusts-RI microenvironment, with high humidity and afavorable temperature, promotes the establishment of moss spe-cies; whereas, lichen species are dominant at Biocrusts-ORImicroenvironment, due to their tolerance to higher temperatures,and resistance to structural and functional damages caused by UV-

Table 4Soil labile and microbial pools seasonal means (± standard error), within threemicroenvironments: i) M. luisana-RI þ biocrusts (Biocrusts-RI), ii) Biocrusts outsideM. luisana-RI (Biocrusts-ORI), and iii) Open areas (OA), as a control; in the dry andrainy seasons, at the Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�an BiosphereReserve, Puebla, Mexico.

Microenvironment

Biocrusts-RI Biocrusts-ORI OA

Labile nutrientsLabile C (mg C g�1)Dry season 100.1 (20)Aa 70.2 (20)ABa 40.2 (10)Ba

Rainy season 10.2 (1.1)Ab 1.2 (0.5)ABb 1.3 (0.1)Bb

Ammonium (mg NH4þ g�1)

Dry season 8.7 (0.9)a 7.3 (1.2)a 8.1 (0.9)a

Rainy season 1.2 (0.1)b 1.1 ((0.1)b 0.9 (0.1)b

Nitrate (mg NO3� g�1)

Dry season 4.8 (1.8)b 3.2 (0.6)b 4.2 (0.8)b

Rainy season 9.7 (1.8)a 8.1 (1.3)a 5.6 (1.2)a

Microbial nutrientsMicrobial C (mg C g�1)Dry season 909 (99)Aa 640 (75)Ba 456 (57)Ca

Rainy season 671 (92)Ab 503 (74)Bb 375 (49)Cb

Microbial N (mg N g�1)Dry season 94 (13)Ab 58 (7)Bb 37 (8)Cb

Rainy season 103 (12)Aa 83 (10)Ba 47 (9)Ca

Microbial C:N ratioDry season 9.8 (0.4)Ba 10.6 (0.8)Ba 17.2 (3.5)Aa

Rainy season 6.3 (0.5)Bb 5.9 (0.5)Bb 9.5 (1.2)Ab

Values followed horizontally by different uppercase letter (A, B, C) indicate thatmeans are significantly different (P� 0.05) amongmicroenvironments, within samesampling season; whereas, values followed vertically by a different lowercase letter(a, b, c) indicate that means are significantly different (P � 0.05) between samplingseasons, within same microenvironment.

Fig. 3. Seasonal variation of a) Potential C mineralization rate (CO2eC), b) Net Nmineralization, and c) Net nitrification, after 21-day incubation of soil samplescollected at three microenvironments: i) M. luisana-RI þ biocrusts (Biocrusts-RI), ii)Biocrusts outside M. luisana-RI (Biocrusts-ORI), and iii) Open Areas (OA); at the Valleyof Zapotitl�an Salinas, Tehuac�an-Cuicatl�an Biosphere Reserve, Puebla, Mexico. Bars withdifferent uppercase letter (A, B, C) indicate that means are significantly different(P � 0.05) among microenvironments, within a same sampling season. Differentlowercase letter (a, b, c) indicate that means are significantly different (P � 0.05) be-tween sampling seasons, within a same microenvironments.

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103 99

rays [11,12], and to changes in soil pH [22]. Though, the mostcommon component of biocrusts are the cyanobacteria, which areable to associate with both mosses and lichens [11,20,22]; conse-quently, being present in both microenvironments. Therefore,M. luisana-RI can modify biocrust species richness and their cover;even more, it has also been reported that other plant species maycause the same effect; for instance, the halfah grass (Stipa tena-cissima L.) in Spanish arid ecosystems [13,41].

Several authors [14,24,26] consider that in the soil under thebiocrusts, in a microenvironmental scale, soil C and N dynamics areclosely related to the cover, and the composition of each one of theorganisms forming the biocrusts, and to the accessibility to rainfallwater. For instance, in this study, a similar contribution ofBiocrusts-RI and Biocrusts-ORI to soil C and N mineralization wasfound, in the rainy season, that could be explained by the speciesshared between both microenvironments. However, in the dryseason, when less species are shared, Biocrusts-RI and Biocrusts-ORI showed a different contribution to soil C and N mineraliza-tion (Fig. 5). In fact, in Biocrusts-ORI, lichens are the dominantgroup, being presumably critical for nutrient transformations, sincethese organisms input large amounts of labile C, reflecting highmineralization rates in dry ecosystems [42,43].

It has already been demonstrated that M. luisana-RI accumulatehigh amounts of organic material, caused by root and litter depo-sition (i.e. leaves and twigs), producing favorable conditions formicroorganisms [6,7] and plant growth, either annuals or peren-nials, under its canopy [7,44�46]. In the case of both Biocrusts-RIand Biocrusts-ORI microenvironments, biocrusts also favor thecapture of litter, allowing the incorporation of organic material tothe soil, process that is caused by biocrusts roughness, which actsas a physical trap [12]. Furthermore, due to the fact that M. luisana,like other legumes, establish a symbiotic association with N-fixingRhizobium bacteria, its litter has a high N content [7,46]. In a study,

also done at the Valley of Zapotitl�an Salinas, Pav�on et al. [46] foundthat M. luisana produces ca. 35% of the litterfall, and reported that

Fig. 4. Principal components analysis (PCA) performed with all soil variables measured at three microenvironments: i) M. luisana-RI þ biocrusts (Biocrusts-RI), ii) Biocrusts outsideM. luisana-RI (Biocrusts-ORI), and iii) Open Areas (OA); during dry and rainy seasons, at the Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�an Biosphere Reserve, Puebla, Mexico.

Table 5Stepwise multiple-regression analyses of soil C and N microbial processes measured in laboratory incubations. Soil was collected within three microenvironments: i)M. luisana-RIþ biocrusts (Biocrusts-RI), ii) Biocrusts outsideM. luisana-RI (Biocrusts-ORI), and iii) Open areas (OA); in dry and rainy seasons, at the Valley of Zapotitl�an Salinas,Tehuac�an-Cuicatl�an Biosphere Reserve, Puebla, Mexico.

Dependent variable All factor included Significant factors in the model r R2

Dry seasonC mineralization Lab Ca

C micb, N micc, NH4þd,NO3

�eLab CC micNH4

þ

0.330.390.41

0.72***

N mineralization Lab CC mic, N mic,

Lab CC mic

0.531.04

0.60**

Nitrification Lab CC mic, N mic, NH4

þLab C 1.31 0.21**

Rainy seasonC mineralization Lab C

C mic, N mic, NH4þ, NO3

�N micNO3

�0.550.39

0.68***

N mineralization Lab CC mic, N mic, inorg N

Lab CC mic

0.431.0

0.56**

Nitrification Lab CC mic, N mic, NH4

þNH4

þ

C mic, N mic1.672.223.30

0.51*

*P � 0.05; **P � 0.01; ***P � 0.001.a Labile C.b Microbial C.c Microbial N.d Ammonium.e Nitrate; n ¼ 42.

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103100

this species had a low N resorption, suggesting that most of the Nwas incorporated to soil via litterfall. Moreover, when an extraor-dinary rainfall period takes place (over 50 mm), M. luisana mayhave a second leaf production [44,45], providing an extra contri-bution of litter that, presumably, could explain no seasonal differ-ences on the litter captured in Biocrusts-RI.

Likewise, the retention and accumulation of litter on Biocrusts-ORI microenvironment only occurred in the rainy season, fact that

may have two explanations: i) Biocrusts-ORI also have a roughmicrotopography [11,12,22], and ii) In the rainy season, biocrustsconstituents (i.e. lichens, mosses, cyanobacteria and filamentousalgae) increase their activity and growth, releasing organic poly-mers, which confer a great adhesion to the organic material [12,47].

Therefore, the contribution of both Biocrusts-RI and Biocrusts-ORI to organic C and total N input to the soil, is explained by twomechanisms: i) Litter enriched with C and N trapped in the biocrust

Fig. 5. Hypothetical model summarizing how biocrusts, inside and outside Mimosa luisana-resources islands, linked to rainfall seasonality, influence soil biological processes in atropical semiarid ecosystem in Mexico. The size of the boxes indicates relative differences in nutrient concentrations among microenvironments (M. luisana-RI þ biocrusts ¼ Biocrusts-RI, Biocrusts outsideM. luisana-RI ¼ Biocrusts-ORI, and Open Areas ¼ OA) or seasons (dry and rainy); while the thickness of the arrows shows the relativeimportance of the biological processes. Dotted lines indicate hypothetical intensity of biological processes, which were not measured in this study. SOC ¼ soil organic carbon;TN ¼ total nitrogen; Microbial C and N; NH4 ¼ ammonium; NO3 ¼ nitrate.

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103 101

surface, which can be incorporated by solubilization or decompo-sition at the beginning of the rainy season, when microbial activityis stimulated [10,7,48], and ii) Biocrusts can also incorporateorganic C and N as a product of their metabolic activity [15,42,43].For example, C fixed through photosynthesis varies between bio-crusts components: 24e80 g m�2 y�1 in bryophytes,12e37 g m�2 y�1 in lichens, and 0.4e2.3 g m�2 y�1 in cyanobac-teria, increasing soil organic C up to 300% in desert ecosystems[12,14,47].

Similarly, N inputs by mosses (46 mg m�2 y�1), lichens(38.1 mg m�2 day�1), and cyanobacteria (192 mgm�2 y�1) conforma critic pool of organic N that contribute, by mineralization, to NH4

þ

and NO3� availability in the soil under biocrusts [14�16]. Though, in

this study, no significant differences were found in soil inorganic Nbetween Biocrusts-RI and Biocrusts-ORI; similar results were re-ported by Delgado-Baquerizo et al. [16] in semiarid areas of Spain.However, net N mineralization, in the dry season, suggests thatBiocrusts-RI and Biocrusts-ORI are potential microenvironmentsfor the production of inorganic N, when compared with OA;whereas, net nitrification reveals that higher NH4 oxidation occursat Biocrusts-ORI and OA. This distinct pattern may be explained bythe fact that Biocrusts-RI and Biocrusts-ORI, with high labile C,promote the formation of NH4

þ, and a higher demand of hetero-trophic microorganisms, favoring N retention within microbialbiomass [9,49]. However, Biocrusts-RI had higher microbial N thanBiocrusts-ORI, in both seasons, suggesting that in Biocrusts-RI, Nimmobilization was dominant over mineralization (Fig. 5).

In contrast, at Biocrusts-ORI and OA, nitrifiers are able to obtainlarge amounts of NH4

þ without having competition with

heterotrophs, whose growth is limited by labile C [49]. For instance,Johnson et al. [50] reported a higher abundance of ammonifyingmicrobes than nitrifiers in biocrusts. Also, other authors [9,42,51]pointed out that the availability of organic C exert a strong con-trol on soil N dynamics, in dry tropical ecosystems. Furthermore, inrainy season, moisture input presumably stimulated nitrificationthrough a decrease in labile C for leaching or microbial use, limitingthe growth and N demand of heterotrophic microorganisms, andfavoring NH4 availability by the nitrifier ones [9,51]. Thus, in amicroscale, inorganic N produced under Biocrusts-RI and Biocrusts-ORI might diffuse to adjacent microenvironments as the openareas.

In the rainy season, at Biocrusts-ORI microenvironment, soilmicroorganisms favored C mineralization in a magnitude compa-rable to the one obtained at Biocrusts-RI, suggesting that the littertrapped on Biocrusts-ORI surface that carries through an active Cinput to the soil, results in the stimulation of the heterotrophicmicrobial communities (Fig. 5), which importantly contribute tothe soil nutrient availability [9,43]; however, this contribution maybe also regulated by C availability. Verhagen et al. [49] observedthat a low labile C, might allow to the replacement of soil microbialcommunities, probably inhibiting the activity of the heterotrophicmicroorganisms, and activating autotrophic routes that use Nforms, due to C available depletion. Thus, in tropical semiaridecosystems, this might be an important mechanism for which bothBiocrusts-RI and Biocrusts-ORI, could be able to influence soilbiology dynamics.

In this study, during the rainy season, when rainfall water is nota constraint, C mineralization and nitrification were explained by

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103102

microbial N and nitrate, supporting the assumption that microbialcommunity was limited by the available C; hence, soil microbesswitch to N as their energy source. Though, in dry season, C and Ntransformations were firstly related to labile C and microbial C.Similar changes in soil microbial communities, induced by Cavailability, have also been reported by other authors in differentecosystems of the world [43,49�53].

5. Conclusions

Mimosa luisana-RI, in addition with rainfall seasonality, alteredbiocrust species richness and their cover, affecting the soil C and Ndynamics under them. For instance, biocrust constituents have theability to provide C and N to the soil at different rates; hence, thishighlight the importance for determining the species associated toeach biocrust, as well as their metabolism, for a better under-standing of soil C and N dynamics. Nevertheless, though bothM. luisana-RI and biocrusts influence the soil under them, theirrange of influence is completely different in scale; then, no com-parison is possible. Thus, the overall results support that, in amicroscale, both Biocrusts-RI and Biocrusts-ORI may be forming“mantles of fertility” in comparison with OA. In addition, our studyprovides an experimental data set that supports a hypotheticalmodel predicting that biocrusts have a strong influence on soilbiology dynamics (Fig. 5). This model highlight that, in the rainyseason, when both, Biocrusts-RI and Biocrusts-ORI, are metaboli-cally active, they have a similar functional role on soil microbialdynamics and N transformations driven by their effect on C avail-ability, through their contribution to the retention of litter and in-puts of organic C, labile C, and total N, in the soil under them.Likewise, the model suggests that a microbial switch for labile Cand N, as an energy source, may bemodifyingmicrobial demand forN, and thus promotes N immobilization in microbial biomass orfavoring its release by nitrification (Fig. 5). Hence, biocrusts, insideand outside of M. luisana-RI, may be critical components for soilfertility maintenance, particularly in the tropical semiaridecosystem of the Valley of Zapotitl�an Salinas, Tehuac�an-Cuicatl�anBiosphere Reserve, Puebla, Mexico.

Acknowledgements

We thank Alejandrina �Avila Ortiz andMarco Antonio Hern�andezMu~noz for the taxonomic identification of the biocrusts compo-nents; Irma Reyes Jaramillo, Maribel Nava Mendoza and RodrigoVel�azquez Dur�an for their technical support in the laboratory; andEduardo Laurent Martínez Olivaresy, Claudia Sosa Izabal and CarlosAlberto Sandoval Antúnez for their help during the field work. ALSPthanks the Master in Biology, Universidad Aut�onoma Metropoli-tana-Iztapalapa, and Consejo Nacional de Ciencia y Tecnología forthe grant received (No. 224682).

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ejsobi.2016.03.006.

References

[1] A. Austin, L. Yahdjian, J.M. Stark, J. Belnap, A. Porporato, U. Norton,D.A. Ravetta, S.M. Schaeffer, Water pulses and biogeochemical cycles in aridand semiarid ecosystems, Oecologia 141 (2004) 221e235.

[2] M.R. Aguiar, O.E. Sala, Patch structure, dynamics and implications for thefunctioning of arid ecosystems, Trends Ecol. Evol. 14 (1999) 273e277.

[3] J.F. Reynolds, R.A. Virginia, P.R. Kemp, A.G. De Soyza, D.C. Tremmel, Impact ofdrought on desert shrubs effects of seasonality and degree of resource islanddevelopment, Ecol. Monogr. 69 (1999) 69e106.

[4] A. Challenger, Utilizaci�on y conservaci�on de los ecosistemas terrestres de

M�exico: pasado, presente y futuro, Agrupaci�on Sierra Madre, S.C. M�exico,1998.

[5] S.L. Camargo-Ricalde, S.S. Dhillion, R. Grether, Community structure ofendemic Mimosa species and environmental heterogeneity in a semi-aridMexican valley, J. Veg. Sci. 13 (2002) 697e704.

[6] S.L. Camargo-Ricalde, S.S. Dhillion, Endemic Mimosa species can serve asmycorrhizal “resource islands”within semiarid communities of the Tehuac�an-Cuicatl�an Valley, Mexico, Mycorrhiza 13 (2003) 129e136.

[7] S.L. Camargo-Ricalde, I. Reyes-Jaramillo, N.M. Monta~no, Forestry insularityeffect of four Mimosa L. species (Leguminosae-Mimosoideae) on soil nutrientsof a Mexican semiarid ecosystem, Agrofor. Syst. 80 (2010) 385e397.

[8] R. García-S�anchez, S.L. Camargo-Ricalde, E. García-Moya, M. Luna-Cavazos,A. Romero-Manzanares, N.M. Monta~no, Prosopis laevigata and Mimosa biun-cifera (Leguminosae) jointly influence plant diversity and soil fertility in aMexican semiarid ecosystem, Rev. Biol. Trop. 60 (2012) 87e103.

[9] Y. Perroni, C. Monta~na, F. García-Oliva, Carbon-nitrogen interactions infertility island soil from a tropical semi-arid ecosystem, Funct. Ecol. 24 (2010)233e242.

[10] A. Serrano-V�azquez, S. Rodríguez-Zaragoza, H. P�erez-Ju�arez, J. Baz�an-Cuenca,V. Rivera-Aguilar, A. Dur�an, Physical and chemical variations of the soil undertwo desert shrubs in Tehuac�an, Mexico, Soil Sci. 178 (2013) 87e103.

[11] V. Rivera-Aguilar, G. Montejano, S. Rodríguez-Zaragoza, A. Dur�an, Distributionand composition of cyanobacteria, mosses and lichens of the biological soilcrusts of the Tehuac�an Valley, Puebla, Mexico, J. Arid. Environ. 67 (2006)208e225.

[12] J. Belnap, The world at your feet: desert biological soil crusts, Front. Ecol.Environ. 1 (2003) 181e189.

[13] F.T. Maestre, M. Huesca, E. Zaady, S. Bautista, J. Cortina, Infiltration, penetra-tion resistance and microphytic crust composition in contrasted micrositeswithin a Mediterranean semi-arid steppe, Soil Biol. Biochem. 34 (2002)895e898.

[14] A.P. Castillo-Monroy, F.T. Maestre, M. Delgado-Baquerizo, A. Gallardo, Bio-logical soil crusts modulate nitrogen availability in semi-arid ecosystems:insights from a Mediterranean grassland, Plant Soil 333 (2010) 21e34.

[15] M. Delgado-Baquerizo, A.P. Castillo-Monroy, F.T. Maestre, A. Gallardo, Plantsand biological soil crusts modulate the dominance of N forms in semiaridgrassland, Soil Biol. Biochem. 42 (2010) 376e378.

[16] M. Delgado-Baquerizo, F. Covelo, F.T. Maestre, A. Gallardo, Biological soilcrusts affect small-scale spatial patterns of inorganic N in semiarid Mediter-ranean grassland, J. Arid. Environ. 91 (2013) 147e150.

[17] H. Godinez-�Alvarez, C. Morín, V. Rivera-Aguilar, Germination, survival andgrowth of three vascular plants on biological soil crusts from a Mexicantropical desert, Plant Biol. 14 (2012) 157e162.

[18] M. Pando-Moreno, V. Molina, E. Jurado, J. Flores, Effect of biological soil crustson the seed germination of three plant species under laboratory conditions,Bot. Sci. 92 (2014) 273e279.

[19] V. Rivera-Aguilar, H. Godinez-�Alvarez, I. Manuell-Cacheux, S. Rodríguez-Zar-agoza, Physical effects of biological soil crusts on seed germination of twodesert plants under laboratory conditions, J. Arid Environ. 63 (2005) 344e352.

[20] Y. Maya, A. L�opez-Cort�es, Cyanobacterial microbiotic crusts in eroded soils of atropical dry forest in the Baja California Peninsula, Mexico, Geomicrobiol. J. 19(2002) 505e518.

[21] A. L�opez-Cort�es, Y. Maya, J.Q. García-Maldonado, Diversidad filogen�etica deMicrocoleus de las costras biol�ogicas de suelo de la Península de Baja Cali-fornia, M�exico, Rev. Mex. Biodivers. 81 (2010) 1e7.

[22] V. Rivera-Aguilar, H. Godinez-�Alvarez, R. Moreno-Torres, S. Rodríguez-Zar-agoza, Soil physico-chemical properties affecting the distribution of biologicalsoil crusts along an environmental transect at Zapotitl�an drylands, Mexico,J. Arid Environ. 73 (2009) 1023e1028.

[23] A. Jim�enez, E. Huber-Sannwald, J. Belnap, D.R. Smart, T. Arredondo-Moreno,Biological soil crusts exhibit a dynamic response to seasonal rain and releasefrom grazing with implications for soil stability, J. Arid Environ. 73 (2009)1158e1169.

[24] L. Concostrina-Zubiri, E. Huber-Sannwald, I. Martínez, J. Flores-Flores, Bio-logical soil crusts greatly contribute to small-scale soil heterogeneity along agrazing gradient, Soil Biol. Biochem. 64 (2013) 28e36.

[25] D.B. Thompson, L.R. Walker, F.H. Landau, L. Stark, The influence of elevation,shrub species, and biological soil crust on fertile islands in the Mojave Desert,J. Arid Environ. 61 (2005) 609e629.

[26] F.T. Maestre, C. Escolar, M. Ladr�on de Guevara, J.J. Quero, R. L�azaro,M. Delgado-Baquerizo, V. Ochoa, M. Berdugo, B. Gonzalo, A. Gallardo, Changesin biocrust cover drive C cycle responses to climate change in drylands, Glob.Chang. Biol. 19 (2013) 3835e3847.

[27] E. García, Modificaciones al sistema de clasificaci�on clim�atica de K€oeppen, 3aed., Instituto de Geografía, UNAM, M�exico, M�exico, 1981.

[28] F. García-Oliva, Influencia de la din�amica del paisaje en la distribuci�on de lascomunidades vegetales en la Cuenca del río Zapotitl�an, Puebla. UniversidadNacional Aut�onoma de M�exico. Bol, Investig.. Geogr. 23 (1991) 53e70.

[29] P. D�avila, M.C. Arizmendi, A. Valiente-Banuet, J.L. Villase~nor, A. Casas, R. Lira,Biological diversity in the Tehuac�an-Cuicatl�an Valley, Mexico, Biol. Conserv.11 (2002) 421e442.

[30] A. Martínez-Bernal, R. Grether, Mimosa, in: A. Novelo, R. Medina-Lemus (Eds.),Flora del Valle de Tehuac�an-Cuicatl�an. Fascículo 44, Instituto de Biología,Universidad Nacional Aut�onoma de M�exico, M�exico, 2006, pp. 42e99.

[31] S.S. Dhillion, S.L. Camargo-Ricalde, The cultural and ecological roles of Mimosa

A.L. Sandoval P�erez et al. / European Journal of Soil Biology 74 (2016) 93e103 103

species in the Tehuac�an-Cuicatl�an valley, Mexico, Econ. Bot. 59 (2005)390e394.

[32] D.J. Eldridge, R. Rosentrete, Morphological groups: a framework for moni-toring microphytic crusts in arid landscapes, J. Arid Environ. 41 (1999) 11e25.

[33] A.E. Magurran, Ecological Diversity and its Measurement, Princeton UniversityPress, 2004.

[34] UIC, Operation Manual for the CM5012 CO2 Colourmeter, Joliet IL., USA, 1995.[35] J.M. Bremmer, Nitrogen-total, in: D.L. Spark, A.L. Page, M. Summer,

M. Tabatabai, P.A. Helmke (Eds.), Methods of Soil Analyses Part 3: ChemicalAnalyses, Soil Science Society American, Madison, WI, 1996, pp. 1085e1121.

[36] Technicon, Industrial System. Method No. 329-74W/B Determinations of Ni-trogen and/or Phosphorus in BD Acid Digest, Tech Industrial Sys, New York,1977.

[37] P.G. Robertson, D.C. Coleman, C.S. Bledsoe, P. Sollins, Standard Soil Methodsfor Long-term Ecological Research (LTER), Oxford University Press, 1999.

[38] R.G. Joergensen, The fumigation-extraction method to estimate soil microbialbiomass: calibration of KEC, Soil Biol. Biochem. 28 (1996) 25e31.

[39] R.G. Joergensen, T. Mueller, The fumigation-extraction method to estimate soilmicrobial biomass: calibration of the KEN-factor, Soil Biol. Biochem. 28 (1996)33e37.

[40] R.R. Sokal, F.J. Rohlf, Biometry, third ed., Freeman and Company, San Francisco,CA, 1995.

[41] F.T. Maestre, A. Escudero, I. Martínez, C. Guerrero, A. Rubio, Does spatialpattern matter to ecosystem functioning? Insights from biological soil crusts,Funct. Ecol. 19 (2005) 566e573.

[42] I. Miralles, C. Trasar-Cepeda, M.C. Leir�os, F. Gil-Stores, Labile carbon in bio-logical soil crusts in the Tabernas desert, SE Spain, Soil Biol. Biochem. 58(2013) 1e8.

[43] J. Yu, N. Glazer, Y. Steinberger, Carbon utilization, microbial biomass, andrespiration in biological soil crusts in the Negev Desert, Biol. Fert. Soils 50(2014) 285e293.

[44] N.P. Pav�on, O. Briones, Root distribution, standing crop biomass and below-ground productivity in a semidesert in Mexico, Plant Ecol. 146 (2000)131e136.

[45] N.P. Pav�on, O. Briones, Phenological patterns of perennial plants in an inter-tropical semi-arid Mexican scrub, J. Arid. Environ. 49 (2001) 65e278.

[46] N.P. Pav�on, O. Briones, J. Flores-Rivas, Litterfall production and nitrogencontent in an intertropical semi-arid Mexican scrub, J. Arid Environ. 60 (2005)1e13.

[47] D.M. Mager, A.D. Thomas, Extracellular polysaccharides from cyanobacterialsoil crusts: a review of their role in dryland soil processes, J. Arid Environ. 75(2011) 91e97.

[48] S. Rodríguez-Zaragoza, T. Gonz�alez, E. Gonz�alez-Lozano, A. Lozada-Rojas,E. Mayzlish, Y. Steinberger, Vertical distribution of microbial communitiesunder the canopy of two legume bushes in the Tehuac�an Desert, Mexico,Europ. J. Soil Biol. 44 (2008) 373e380.

[49] F.J. Verhagen, H.J. Laanbroek, Competition for ammonium between nitrifyingand heterotrophic bacteria in dual energy-limited chemostats, Appl. Environ.Microb. 57 (1991) 3255e3263.

[50] S.L. Johnson, C.B. Budinoff, J. Belnap, F. García-Pichel, Relevance of ammoniumoxidation within biological soil crust communities, Environ. Microb. 7 (2005)1e12.

[51] N.M. Monta~no, A.L. Sandoval-P�erez, M. Nava-Mendoza, J.M. S�anchez-Ya~nez,F. García-Oliva, Variaci�on espacial y estacional de grupos funcionales debacterias cultivables del suelo de un bosque tropical seco en M�exico, Rev. Biol.Trop. 61 (2013) 439e453.

[52] M.A. De Graaff, A.T. Classen, H.F. Castro, C. Schadt, Labile soil carbon inputsmediate the soil microbial community composition and plant residuedecomposition rates, New Phytol. 188 (2010) 1055e1064.

[53] L.E. Green, A. Porras-Alfaro, R.L. Sinsabaugh, Translocation of nitrogen andcarbon integrates biotic crust and grass production in desert grassland, J. Ecol.96 (2008) 1076e1085.