Ia

Ra

Rb

c

a

ARRA

KMSTSCMM

1

osstsfiK

imact

0h

Agriculture, Ecosystems and Environment 183 (2014) 86– 92

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment

jo ur nal home p age: www.elsev ier .com/ locate /agee

nfluence of land use on soil quality and stratification ratios undergro-silvo-pastoral Mediterranean management systems

osa Francavigliaa,∗, Anna Benedetti a, Luca Dorob, Salvatore Madrauc, Luigi Leddab

Consiglio per la Ricerca e la sperimentazione in Agricoltura, Centro di ricerca per lo studio delle relazioni tra pianta e suolo, Via della Navicella 2-4, 00184ome, ItalyDipartimento di Agraria, Sezione di Agronomia, Coltivazioni erbacee e Genetica, Università di Sassari, Viale Italia 39, 07100 Sassari, ItalyDipartimento di Agraria, Sezione di Ingegneria del Territorio, Università di Sassari, Viale Italia 39, 07100 Sassari, Italy

r t i c l e i n f o

rticle history:eceived 19 July 2013eceived in revised form 23 October 2013ccepted 25 October 2013

eywords:editerranean systems

oil organic carbonotal Ntratification ratio:Nicrobial biomass

a b s t r a c t

A case study from north-eastern Sardinia (Italy) in semiarid conditions is presented. Agriculture is mainlyextensive and markedly agro-silvo-pastoral, and is typical of similar areas of the Mediterranean basin.The following land uses at different levels of crop intensification were considered: tilled vineyard (TV),no-tilled grassed vineyard (GV), hay crop and pasture with sparse cork oaks (HC and PA), semi-naturalsystems (SN, former vineyards set-aside about 30 years ago), cork oak forest (Quercus suber L.) establishedin the past century (CO). Some soil quality parameters were considered: soil organic carbon (SOC) andtotal N (TN) concentrations, stocks and their stratification ratios with depth (SRs), microbial biomasscarbon (MBC) and its quotient to SOC (qmic), and C:N ratios.

Both in terms of concentrations and stocks, SOC and TN were generally higher in HC, PA, CO and SN:in these land uses SOC in the topsoil were in the range 17.0–24.3 g kg−1 and 48.9–65.4 t ha−1; TN valueswere 1.07–2.08 g kg−1 and 3.1–6.0 t ha−1.

icrobial quotient SOC and TN SRs under the CO land use were higher than 4, quite above the proposed threshold (�2),>2 in GV, and ≥2.0 in PA.

MBC in mg kg−1 and qmic in �g g−1 were higher under CO (194 and 0.89) and GV (156 and 0.97).C:N ratios had optimum or nearly optimum (9–12) values in CO, PA and the GV, in agreement with the

SRs, MBC and qmic. A positive and significant correlation was found between SOC and TN concentrationsin all the land uses.

. Introduction

Soil organic carbon (SOC) is a major component of the soilrganic fraction, positively affects many soil properties and, con-equently, soil functions. In particular, SOC maintains importantoil functions with regard to habitat, biological diversity, soil fer-ility, crop production potential, erosion control, water retention,ubstance exchange between soil, atmosphere and water, and theltering, buffering and transforming capacity (Huber et al., 2001;irchmann and Andersson, 2001).

C sequestration via agricultural soils has a potential to signif-cantly contribute to climate change mitigation. Sound cropland

anagement can play a positive role in reducing GHGs emissions,

nd carbon dioxide in particular, through a decrease of soil organicarbon losses, by increasing the organic matter input or combininghese two options. Literature data estimate about 1550 Pg SOC to

∗ Corresponding author. Tel.: +39 06 7005299; fax: +39 06 7005711.E-mail address: rosa.francaviglia@entecra.it (R. Francaviglia).

167-8809/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.agee.2013.10.026

© 2013 Elsevier B.V. All rights reserved.

1 m depth (Lal, 2008), in comparison with 4000 Gg C of fossil fuels.Vegetation (560 Pg) and atmosphere (760 Pg) store considerablyless C than soils.

When using SOC to compare “soil quality”, we should considerthat SOC varies among environments and management systems,and generally increases with higher mean annual precipitation,with lower mean annual temperature, with higher clay content,with an intermediate grazing intensity, with higher crop residueinputs and cropping intensity, with native vegetation comparedwith cultivated management, with conservation tillage comparedwith plough tillage (Jenny, 1980; Nichols, 1984; Parton et al., 1987;Burke et al., 1989; Rasmussen and Collins, 1991; Franzluebberset al., 1998; Schnabel et al., 2001). Since stratification of SOC poolsis common both in natural and agricultural cropping systems,Franzluebbers (2002) developed the concept of using a stratifi-cation ratio as an indicator of dynamic soil quality, to test the

capability of different soil properties to express the extent ofstratification, and illustrate the potential of SOC and other soilproperties to detect management-induced changes in dynamic soilquality. Stratification ratios >2 indicate a higher soil quality and

R. Francaviglia et al. / Agriculture, Ecosystems and Environment 183 (2014) 86– 92 87

cca2

wahr22Gelsct(tt

2

2

BaiMaLael

••••••

Table 1Land uses description and management.

Land use Description

Tilled vineyard (TV) TV was established in 1993, is ploughed to40 cm and harrowed in March or April andoccasionally in July. Organic fertilization(12.5% organic nitrogen, 40% organiccarbon, 70% organic matter and C/N ratio3.2) is distributed at the end of January atthe rate of 500 kg ha−1 and incorporated inthe first 20 cm of soil with a rototiller; itprovides 200 kg ha−1 of organic carbon,62.5 of N, 42 of K and 11 of P. Pruningresidues are removed from the field. Themonitoring area is 0.98 ha.

No-tilled grassed vineyard (GV) GV was established in 1990. Mineralfertilizer up to 40 kg N ha−1, 22 kg P and 42K are applied in March. Pruning is carriedout in January and June with the pruningresidues being left on the soil. Dripirrigation (up to 100 mm) is providedbetween June and July to partially restorecrop evapotranspiration. The monitoringarea is 3.4 ha.

Hay crop (HC) The HC land use is oats, Italian ryegrassand annual clovers or vetch for 5 years andintercropped by spontaneous herbaceousvegetation in the sixth year. It is ploughedto 40 cm and harrowed before seeding 5years out of six. 50 kg N ha−1 and39 kg P ha−1 are applied before seeding andgrazing is allowed with 3–4 sheep ha−1

from January until March, before being cutin May. The monitoring area is 3.5 ha.

Pasture (PA) The PA land use is 5 years of spontaneousherbaceous vegetation, and one year ofintercropping with oats, Italian ryegrassand annual clovers or vetch cultivated as ahay crop. It is tilled 1 year out of 6 and isgrazed from December until June with 6sheep ha−1. The HC and PA land uses havea complementary 6-year rotation. Themonitoring area is 11 ha.

Cork oak forest (CO) The CO land use (Quercus suber L.) wasestablished in the past century, is used forcork production and cattle grazing and theunderstory is covered by mixed herbaceousvegetation. The monitoring area is 10 ha.

Semi-natural systems (SN) The SN land use (scrublands,Mediterranean maquis and Helichrysummeadows) arise from the naturalre-vegetation of former vineyardsestablished between 1943 and 1951, whichhave been set-aside about 30 years agoprobably due to the low grape yields and

Fig. 1. Experimental site location in northeast Sardinia (Italy).

ontribution to agriculture sustainability. Total soil N (TN) is asso-iated with SOC and plays a key role in building soil fertilitynd enhancing soil productivity (Franzluebbers and Stuedemann,008).

Researches under Mediterranean conditions are in agreementith this theory, but available literature mostly deals with intensive

gricultural systems and soil tillage practices, and very few studiesave addressed only partially the agro-silvo-pastoral Mediter-anean management systems (Murillo et al., 2004; Moreno et al.,006; Hernanz et al., 2009; Peregrina et al., 2010; Nieto et al.,011; Melero et al., 2012; Corral-Fernández et al., 2013; Lozano-arcía and Parras-Alcántara, 2013). The aim of this study was tovaluate some soil quality parameters and stratification ratios inand uses at different levels of crop intensification, under agro-ilvo-pastoral Mediterranean management systems in semiaridonditions (northeastern Sardinia, Italy). Data are discussed inerms of SOC and TN concentrations and stocks, stratification ratiosSR) calculated from contents in the 0–20 cm soil layer divided byhat in the 20–50 cm, microbial biomass carbon (MBC) and its quo-ient to SOC (qmic), and C:N ratios.

. Materials and methods

.1. Site description

The site (Fig. 1) is within an area of about 1450 ha in theerchidda Municipality (40◦46′ N, 9◦10′ E, mean altitude 285 m.s.l.), characterized by extensive agro-silvo-pastoral systems, typ-cal of north-eastern Sardinia (Italy) and similar areas of the

editerranean basin (e.g. the Iberian peninsula). The area is char-cterized by the same type of soil and cork oak forest (Quercus suber.) as potential native vegetation which has been converted to man-ged land with pastures and vineyards in recent years (Lagomarsinot al., 2011; Francaviglia et al., 2012). Six land uses, with differentevels of cropping intensification were compared (Table 1, Fig. 2):

Tilled vineyards (TV);No-tilled grassed vineyards (GV);Hay crop (HC);

Pasture (PA);Cork oak forest (CO);Semi-natural systems (SN).

the high cost of modern tillage equipment.The monitoring area is 4 ha.

TV and GV vineyards are agricultural higher intensive land uses,whereas HC, PA, CO and SN are agro-silvo-pastoral lower intensiveland uses.

Both PA and HC included scattered cork-oak trees, which are keycomponents of the “Dehesa”-type landscape (grazing system withQuercus L.) typical of this area of Sardinia and other areas of south-ern Europe (e.g. Spain and Portugal). Dehesas are often convertedto more profitable land uses such as vineyards (Francaviglia et al.,2012) or olive groves (Lozano-García and Parras-Alcántara, 2013).

The local climate is typically Pluvi-seasonal oceanic low meso-Mediterranean low sub-humid (Rivas-Martinez and Rivas-Saenz,2009), with a mean annual rainfall of 623 mm (range 367–811 mm)and mean annual temperature of 15.0 ◦C (13.8–16.4 ◦C). According

to the updated Köppen-Geiger climate classification (Kottek et al.,2006) the climate is warm temperate with dry and hot summers(Csa).

88 R. Francaviglia et al. / Agriculture, Ecosystems and Environment 183 (2014) 86– 92

F vineys n (PA

2

dSvrTcdlPitgumcl(aat(pc

wptapt2g

to 20 cm of topsoil excluding the subsoil.Soils are mainly Haplic Endoleptic Cambisols, Dystric (WRB,

2006), derived from granitic rocks, topsoils (Table 2) have sub-acid

ig. 2. Land uses in the study area. From left to right clockwise: no-tilled grassedcrubs and Mediterranean maquis (SN), grazed pasture with spontaneous vegetatio

.2. Soil sampling and analyses

Soils samples were collected along the different soil horizonsuring the pedologic survey began in February 2007 within theOILSINK national project, where pits were digged with a mini exca-ator. The six land uses are common in the study area, but samplingsefer to smaller monitoring areas chosen for the project (Table 1).he monitoring sites have been chosen because of their peculiarharacteristics, which make them suitable to highlight differencesue to land use and management within similar climatic and pedo-

ogic conditions. Four replicates were collected in the TV, GV andA, three replicates in CO, five replicates in the HC, seven replicatesn SN, to consider local differences due to soil slope and/or vege-ation heterogeneity. The number of soil samples was based on aeo-statistical analysis performed in the whole area for each landse, which allowed to assess that the spatial variability of SOC wasainly due to land uses and within them by a different land cover

ondition. As consequence, a larger number of samples was col-ected in the more heterogeneous land uses, such as the vineyardswhere samples were collected along and between the rows), PAnd HC (where samples were collected in areas covered by treesnd open areas). SN was sampled in different conditions of vegeta-ion cover due to the existence of heterogeneous natural vegetationscrublands, Mediterranean maquis and meadows). Less soil sam-les were collected for the CO where the soil was homogeneouslyovered by trees and bushes.

The depth of Ap and Bw horizons of the different soil profilesere largely different even in the same land use, mainly due toresent or past tillage operations and this would have influencedhe evaluation of the data. Thus soil parameters were normalizedt 2 fixed depth intervals (0–20 and 20–50 cm) to enable the com-

arison among the different soil profiles and the calculation ofhe stratification ratios (Fig. 3). Stratification ratios (Franzluebbers,002) were thus calculated from soil organic carbon and total nitro-en concentrations at 0–20 cm divided by those at 20–50 cm values.

ard (GV), grazed hay crop under oats land cover (HC), semi-natural systems with).

SOC and TN concentrations and stocks were determined from aset of samplings performed during the pedologic soil survey inFebruary 2007, but the determination of these data and the othermain soil parameters reported in Table 2 was not designed to assessthe variability of the parameters over the year. By way of contrast,microbiological parameters have a seasonal variability mainly dueto changes of soil moisture and temperature in the field and theinput of fresh organic materials and root exudates during the crop-ping season (Campbell et al., 1999; He et al., 1997; Waldrop andFirestone, 2006). Thus MBC and its ratio to SOC (qmic) were deter-mined from three samplings in February, May and November 2007,where only SOC and MBC have been analyzed. Moreover, since themain changes in microbial parameters due to conversion betweenland uses can be expected in the upper centimeters of the soil(Conant et al., 2001), the analyses of these parameters were limited

Fig. 3. Soil organic carbon and total soil N stratification ratios (SR) and land uses.The vertical bars indicate the standard errors. The bold horizontal line is the SRthreshold.

R. Francaviglia et al. / Agriculture, Ecosystems and Environment 183 (2014) 86– 92 89

Table 2Main soil parameters (mean ± SD).

Land use and depth Sand (g kg−1) Clay (g kg−1) Silt (g kg−1) Texture

0–20 20–50 0–20 20–50 0–20 20–50 0–20 20–50

TV 830 ± 34 840 ± 29 120 ± 14 110 ± 5 50 ± 44 50 ± 31 Loamy Sand Loamy SandGV 800 ± 6 853 ± 52 115 ± 13 82 ± 27 85 ± 14 65 ± 27 Sandy Loam Sandy LoamHC 733 ± 5 741 ± 58 131 ± 9 118 ± 8 136 ± 5 141 ± 54 Sandy Loam Sandy LoamPA 732 ± 12 756 ± 24 135 ± 8 128 ± 19 133 ± 5 116 ± 14 Sandy Loam Sandy LoamCO 761 ± 29 646 ± 30 116 ± 3 186 ± 2 123 ± 26 168 ± 25 Sandy Loam Sandy LoamSN 791 ± 36 817 ± 28 101 ± 31 101 ± 30 108 ± 37 82 ± 15 Sandy Loam Loamy Sand

Land use and depth pH P (mg kg−1) C:N Bulk density (g cm−3)

0–20 20–50 0–20 20–50 0–20 20–50 0–20 20–50

TV 5.1 ± 0.1 5.4 ± 0.3 34 ± 4 19 ± 8 14.0 ± 2.7 13.4 ± 3.6 1.51 ± 0.07 1.52 ± 0.04GV 6.2 ± 0.3 6.1 ± 0.6 30 ± 14 10 ± 7 11.7 ± 4.7 10.9 ± 5.1 1.51 ± 0.04 1.53 ± 0.04HC 5.6 ± 0.4 5.9 ± 0.3 35 ± 22 14 ± 7 14.8 ± 3.3 15.4 ± 2.2 1.38 ± 0.10 1.46 ± 0.04PA 5.5 ± 0.4 5.6 ± 0.4 26 ± 8 10 ± 4 9.1 ± 2.0 10.5 ± 2.4 1.44 ± 0.03 1.52 ± 0.04CO 5.7 ± 0.2 6.1 ± 0.1 5 ± 3 2 ± 2 12.9 ± 0.1 11.5 ± 0.1 1.42 ± 0.02 1.58 ± 0.01

6 ± 2

T ork oa

tt2

<muwfiwe(bfbtSSsmottcBtwoto

w

3

3

l(rwaac

SN 6.0 ± 0.3 6.2 ± 0.3 6 ± 2

V, tilled vineyard; GV, no-tilled grassed vineyard; HC, hay crop; PA, pasture; CO, c

o acid reaction (pH 5.1–6.2), sandy-loam and loamy-sand tex-ure (sand 732–830 g kg−1), 5–6 mg available P kg−1 in CO and SN,6–35 mg kg−1 P in all the other land uses.

Soil samples were air-dried, and the analyses were made on the2 mm soil fraction after sieving. The soil reaction (pH) was deter-ined in 1:2.5 soil:water suspension by potentiometric method

sing a pH meter; particle-size analysis and soil texture with theet sieving and sedimentation procedure and the USDA classi-cation respectively; TN with the Kjeldahl method; available Pith the Olsen method; SOC with the Walkley-Black method. MBC,

xpressed in mg C kg−1, was determined according to Vance et al.1987) with extraction of C from fumigated and unfumigated soilsy 0.5 mol L−1 K2SO4. qmic, expressed in �g C g−1, was calculatedrom the ratio MBC:SOC after Anderson and Domsch (1989). Soilulk density (BD), required to convert SOC and TN concentrationso t ha−1, was calculated with the SPAW model (Soil Plant Air Water;axton and Willey, 2006) which includes the equation proposed byaxton and Rawls (2006) after model parametrization using mea-ured BD data acquired from representative soil profiles. BD waseasured by core method in 2 soil profiles in the PA and HC (in

pen and tree covered areas), and vineyards (along and betweenhe rows); in three profiles in SN (to consider the heterogeneity ofhe natural vegetation), and only in one profile in CO where landover was homogeneous. These soil data were used to estimateD with the SPAW model modifying the density adjustment factoro match the measured bulk density. For all the other soil profilesithin the same land use, BD was estimated with SPAW using the

bserved sand, clay, gravel and organic matter contents, and usinghe density adjustment factor used for the corresponding soil depthf the same land use.

Statistical analyses were performed using the Statistica 8.0 soft-are package (Statsoft, Tulsa, USA).

. Results and discussion

.1. Soil organic carbon, total N, microbial biomass and qmic

Data show that in terms of SOC concentrations of the surfaceayer (0–20 cm) the lower intensive land uses HC, CO, PA and SNTable 3) had higher but not always significantly different contentsanging in decreasing order from 24.3 to 17.0 g kg−1 in comparison

ith the two vineyards (14–11 g kg−1). Due to the organic fertilizer

dding 200 kg ha−1 year−1 of organic carbon, the tilled vineyard had higher (14.2 g kg−1) but not significantly different value of SOC inomparison to the GV vineyard. In the subsurface layer (20–50 cm)

16.5 ± 3.4 19.4 ± 2.8 1.44 ± 0.03 1.47 ± 0.05

k forest; SN, semi-natural systems (former vineyards).

SOC contents in g kg−1 were higher and significantly different in HCand SN (∼15), intermediate in TV and PA (∼10), and lower in GVand CO (5–6). As observed by Lagomarsino et al. (2011), the sup-plementary irrigation provided in GV during the summer, createsa warm and moist condition which enhances SOC mineralizationand counterbalance the positive effect of the grass cover.

The TN content in g kg−1 of the surface layer (Table 3) wasmuch higher and significantly different in PA, CO and HC (range2.08–1.73), lower in SN, TV and GV (range 1.07–0.93). The TN con-tent in g kg−1 of the subsurface layer (20–50 cm) was higher andsignificantly different in HC and PA (∼1), intermediate in TV and SN(0.75), and much lower in GV and CO (range 0.5–0.4). Both SOC andTN concentrations decreased with depth (Table 3) when consider-ing the surface (0–20 cm) and the deep layer (20–50 cm). Resultsfor the land use under CO are consistent with the findings froma set of Cambisols under a Dehesa-type Mediterranean evergreenoak woodland in southern Spain (Corral-Fernández et al., 2013).In particular, they report SOC concentrations equal to 11.42 and4.63 g kg−1 in the Ap (0–20.9 cm) and Bw (20.9–55.5 cm) horizonsrespectively, and TN concentrations equal to 1.13 and 0.54 g kg−1.

SOC stocks of the surface layer in t ha−1 (Table 3) were higherin HC, PA, CO and SN (range 65.4–48.9) in comparison with thetwo vineyards (42.2–32.5). But the differences were not always sig-nificantly different as in the case of SOC concentrations. Resultsof the land uses under grazing activity, i.e. HC, PA and CO, arein good agreement with González and Candás (2004), reportingvalues about 54 t ha−1. For vineyards, Chiti et al. (2012) report amean SOC stock in the topsoil (30 cm) equal to 41.9 ± 15.9 t ha−1

in Italy, and a lower stock in the regions with Mediterraneanclimate (39.7 ± 10.9 t ha−1), with the vineyards showing the mini-mum stock among the land uses. Rodríguez-Murillo (2001) reportssimilar stocks for vineyards in Spain (42.5 ± 28.9 t ha−1), and Martinet al. (2010) indicate a stock equal to 39.4 ± 26.5 t ha−1 in France.In the subsurface layer SOC stocks were higher in HC, SN and PA(from 66.8 to 45.4 t ha−1), and were significantly different from thestocks of the two vineyards and CO (from 37.8 to 23.3 t ha−1). SOCstocks decreased with depth in TV, GV, PA and CO, but increased inSN and were almost constant in HC. This could be ascribed to localdifferences of slope and land degradation conditions due to watererosion.

TN stocks in t ha−1 of the surface layer (Table 3) were much

higher and significantly different in PA, HC and CO (from 6.0 to 4.4),and lower in SN, TV and GV (from 3.1 to 2.8), in good agreementwith SOC stocks. TN stocks in t ha−1 of the subsurface layer werehigher and significantly different in HC, PA, TV and SN (from 4.4 to

90 R. Francaviglia et al. / Agriculture, Ecosystems and Environment 183 (2014) 86– 92

Table 3Mean soil organic carbon (SOC) and total nitrogen (TN) expressed as concentrations and stocks ±SE and land uses.

Land use and depth SOC (g kg−1) TN (g kg−1) SOC (t ha−1) TN (t ha−1)

0–20 20–50 0–20 20–50 0–20 20–50 0–20 20–50

TV 14.2 ± 2.4ab 9.9 ± 2.3ab 0.99 ± 0.07ab 0.75 ± 0.14abc 42.2 ± 6.0ab 37.8 ± 4.7a 3.0 ± 0.2a 3.6 ± 0.6aGV 11.0 ± 2.3a 6.0 ± 2.7a 0.93 ± 0.05a 0.50 ± 0.13ab 32.5 ± 6.4b 26.9 ± 11.7a 2.8 ± 0.1a 2.2 ± 0.5bHC 24.3 ± 4.5c 15.5 ± 1.6c 1.73 ± 0.45c 1.01 ± 0.08c 65.4 ± 9.8c 66.8 ± 5.9c 4.6 ± 1.0b 4.4 ± 0.3aPA 18.7 ± 1.6abc 10.2 ± 1.7ab 2.08 ± 0.15c 0.99 ± 0.15c 53.7 ± 4.1ac 45.4 ± 7.1b 6.0 ± 0.3b 4.4 ± 0.6aCO 21.8 ± 3.0bc 5.0 ± 2.8a 1.69 ± 0.22bc 0.44 ± 0.10a 57.3 ± 6.8a 23.3 ± 2.8a 4.4 ± 0.5ab 2.0 ± 0.3b

0.75

D ong la

3fF3h

reHtadtddTfr(h0rtwaMbpo(oaqta

sbdu

TMt

Dl

SN 17.0 ± 1.2ab 14.4 ± 1.3bc 1.07 ± 0.05ab

ifferent letters within each soil depth indicate significant differences (p < 0.05) am

.3), and much lower in GV and CO (from 2.2 to 2.0). Again, resultsor the land use under CO are consistent with those from Corral-ernández et al. (2013): in particular, they report SOC stocks of1.81 and 23.58 t ha−1 in the Ap (0–20.9 cm) and Bw (20.9–55.5 cm)orizons respectively, and TN stocks of 3.15 and 2.76 t ha−1.

MBC concentrations in mg kg−1 (Table 4) were in a decreasingange from 194 to 58, and much higher and significantly differ-nt in CO and GV, intermediate in PA and TV, much lower inC and SN. This is in agreement with Nunes et al. (2012) repor-

ing that MBC is higher under native vegetation including trees,nd in land under restoration (green manure) in comparison withegraded land formed as a result of deforestation and loss of vege-ation. Zhang et al. (2006) also observe that soil microbial biomassecreases from land under native vegetation (undisturbed) to moreisturbed or degraded land. The similarity of MBC between PA andV can be ascribed to the carbon inputs provided by the organicertilization applied in TV. qmic values in �g C g−1 were in theange 0.97–0.45 in the decreasing order GV > CO > TV > HC > PA > SNTable 4). The qmic largely followed the same trend of MBC withigher and significantly different values in CO and GV (0.89 and.97). These results are in agreement with Haynes et al. (2003)eporting that qmic is higher under no-tilled in comparison withilled land uses. The results from our study have shown that qmicas lower in comparison with the data reported for forests systems

nd grassland (grazed and not grazed) by Moscatelli et al. (2007) inediterranean land use systems, which are however characterized

y differences in soil texture and pH and higher rainfall and tem-erature. It is generally accepted that qmic values are an indicatorf both a larger substrate availability to the soil microorganismsAnderson and Domsch, 1989) and an increasing trend to stockrganic C in the long term in the less disturbed systems, i.e. CO,nd GV in this study (Lagomarsino et al., 2011). In fact, in our studymic was only slightly lower in TV in comparison with GV due tohe larger substrate availability provided by the organic fertilizationpplied in TV.

Both SOC and TN concentrations and stocks were higher in theoil surface layer (0–20 cm) under HC, PA and CO land uses. This can

e ascribed to several reasons: (i) to soil management, since the noisturbance or the limited tillage operations in the CO and PA landses were more conservative for C (West and Post, 2002); (ii) to the

able 4ean microbial biomass-C concentration (MBC) and microbial quotient (qmic) of

opsoils and land uses ±SE.

Land use MBC (mg kg−1) qmic (�g C g SOC−1)

TV 112 ± 16ab 0.78 ± 0.22abGV 156 ± 19b 0.97 ± 0.26bHC 79 ± 22a 0.56 ± 0.31abPA 115 ± 15ab 0.55 ± 0.16abCO 194 ± 20b 0.89 ± 0.19bSN 59 ± 3a 0.45 ± 0.04ab

ifferent letters in each column indicate significant differences (p < 0.05) amongand uses (Fisher’s LSD post hoc test).

± 0.05bc 48.9 ± 3.2ab 62.2 ± 4.9bc 3.1 ± 0.3a 3.3 ± 0.2a

nd uses (Fisher’s LSD post hoc test).

annual inputs from crop residues and roots turnover in the HC landuse (Franzluebbers et al., 1998); (iii) to the monthly inputs of ani-mal manure during the grazing period (Schnabel et al., 2001), equalto 0.025, 0.1 and 0.5 t C ha−1 in HC, PA and CO respectively. Theseinputs were previously estimated from expert judgment in relationto the animal load, the length of the grazing period and the qualityof pastured plant materials (Francaviglia et al., 2012). SOC and TNstocks were more homogeneous with depth in HC due to the effectof deep tillage operations 5 years out of six; at sampling time inFebruary 2007 the HC land use was under oats land cover, and hadbeen ploughed and sown the previous year in late October, whilePA was at the fifth year of use as natural pasture (see Table 1). Aspreviously described, both HC and PA land uses included also scat-tered cork-oak trees with a large root system, which can explainthe similarity of the two land uses with the CO system. C inputsfrom plant residues in t C ha−1 have been estimated previouslyby inverse modeling using the RothC carbon model (Francavigliaet al., 2012). These estimates were in the range 6.0–1.0 t C ha−1 inthe decreasing order HC � SN > CO > PA > GV > TV, and can be con-sidered in agreement with the magnitude of the expected residueinputs in the different management systems. In addition, for the HCcropping system the estimated average crop residues yearly incor-porated into the soil were about 5.4 t ha−1 equal to the 40.7% of thetotal biomass production (above and below ground). This amount ofcrop residues was estimated using the EPIC model (v 0509) withinthe activity carried on for a PhD thesis (Doro, 2009).

3.2. Soil organic carbon and total N stratification ratios

When considering the stratification ratio of SOC and TN (Fig. 3)CO, GV, and PA had SR ≥ 2, with values ranging in a decreasing orderfrom 4.9 to 2 for SOC, from 4.4 to 2.1 for TN. In TV, HC and SN SRsof SOC were in the range 1.61–1.22, and those of TN in the range1.72–1.48. Similar studies in central and southern Italy show thatstratification ratios of SOC and TN allowed to discriminate betweenconservative management systems and more disturbed situations,where soil organic matter mineralization is enhanced (Papini et al.,2011). Our findings confirm that SR is an indicator of dynamic soilquality, and can allow to detect the management-induced changesin agricultural systems. This was particularly evident for HC, despitethe higher levels of SOC and TN both in terms of concentrations andstocks, and the high C inputs from animal manure and crop residues,since the soil is managed by ploughing and harrowing 5 years outof 6 which results in a soil mixing up to 40 cm. In the case of the TV,SRs are influenced by the tillage which takes place every year andthe removal of pruning residues, while for SN the lower values canbe ascribed to a slower resilience after the soil disturbance in theperiod when the area was a vineyard.

Considering the stratification ratio used to describe soil quality,

the CO land use, that is the native vegetation of the area, proved tobe the less disturbed situation and showed the highest SRs of SOCand TN, i.e. a higher soil quality. The GV land use can be consideredmore sustainable than the TV, since it is no-tilled and covered with

R. Francaviglia et al. / Agriculture, Ecosystem

Fi

sarabdatCwtrnc

3

b2SS

clr(bc(eft

(a

ut(Pia

ig. 4. Regression between total N and soil organic carbon concentrations (g kg−1)n topsoil and subsoil under the six land uses.

pontaneous grass during the vegetative cycle and pruning residuesre left in the field, while TV is ploughed every year and pruningesidues are removed. Lastly, the PA land use, which is ploughednd harrowed only 1 year out of 6, had a better soil quality and cane considered more sustainable than the HC land use. The moreisturbed situations, i.e. TV and HC, had similar SRs both for SOCnd TN. The stratification ratios of SOC and TN concentrations fromhe present study were generally higher than those reported byorral-Fernández et al. (2013) under Mediterranean evergreen oakoodlands in Spain, which ranged between 1.1 and 1.3 as a func-

ion of tillage management. Lou et al. (2012) report stratificationatios as affected by tillage management in maize field crops inortheast China: for SOC the ranges are 1.2–1.3 and 1.5–1.8 underonventional and no tillage respectively, for TN 1.1–1.3 and 1.2–1.5.

.3. Carbon–nitrogen relationship and ratio

A linear regression was used to evaluate the relationshipetween SOC and TN concentration across all depths (0–20 and0–50 cm) under the different land uses (Fig. 4). The regression wasOC = 12.6534 × TN, with a correlation coefficient R2 = 0.9053 andE = 0.4980.

The regression is highly significant (p < 0.001), meaning that SOConcentrations were highly correlated with TN concentrations in alland uses, in agreement with Chen et al. (2009) that used a linearegression to evaluate this relationship in Germany. González et al.2012) found a similar positive correlation in evergreen oak standsetween SOC and nitrogen concentrations under Mediterraneanlimate conditions (R = 0.70; p < 0.001). Ganuza and Almendros2003) found a positive correlation (R = 0.94; p < 0.001) under differ-nt climates and representative vegetation types including nativeorests, coniferous plantations, scrubs, pastures and arable crops inhe Basque Country (Spain).

Regressions applied to topsoil (0–20 cm) and subsoil20–50 cm) were SOC = 12.1342 × TN (R2 = 0.8929, SE = 0.6459),nd SOC = 11.9245 × TN (R2 = 0.8444, SE = 0.8519) respectively.

The mean C:N ratios (Table 2) of the surface layer (0–20 cm)nder the different land uses ranged from 9.1 to 16.5 inhe order PA < GV < CO < TV < HC < SN. In the sub surface layer

20–50 cm) C:N ratios ranged from 10.5 to 19.4, again in the orderA < GV < CO < TV < HC < SN. Chen et al. (2009) report no differencen C:N among plow tillage/reduced tillage (including arable crops)nd grassland (i.e. pastures, grasslands, and fallows).

s and Environment 183 (2014) 86– 92 91

C:N decreased with depth in the two vineyards and in theCO land use, thus confirming that SOC decomposition generallydecreases toward the surface (Lou et al., 2012), but increased in HCand PA, where soils are more disturbed due to the tillage operations,and the SN land use which were former vineyards now abandoned.

Soils with a C:N between 9 and 12 have optimum conditions, i.e.SOM immobilisation-mineralization processes are in equilibrium.Where C:N is higher than 12, soil nitrogen is insufficient to ensurea good and acceptable humification process (USDA-NRCS, 1995;Thomsen et al., 2008), as in the case of TV, HC and SN, both in topsoiland the subsurface layer. Rodeghiero et al. (2011) also show thenegative effect of tillage leading to an increase of the C:N ratio. Incontrast, CO, PA and GV had optimum or nearly optimum values inboth layers, in agreement with the SRs for SOC and TN, MBC andqmic.

4. Conclusions

Results from the present study indicate that a simple indicatorbased on a soil parameter (e.g. soil organic carbon or total nitrogen)is not effective in defining overall soil quality. In fact, the hay cropwhich is used for pasture had the highest concentrations and stocksof SOC and TN among the six land uses considered, but their strat-ification ratios SRs were below the proposed threshold of 2 thatdistinguish soils with improved quality from degraded or inten-sively managed soils. Additional soil parameters based on simplebiological indicators, such as the microbial biomass carbon MBCand its quotient to SOC qmic have confirmed the results obtainedwith the SRs.

A better soil quality was found under unmanaged situations, i.e.the CO land use, or more conservative management systems, i.e.the GV and PA land uses, which showed higher SRs in the orderCO � GV > PA for SOC, CO � PA > GV for TN, together with high tomedium values of MBC and qmic.

SOC and TN concentrations were highly correlated in all landuses. Moreover, CO, GV and PA land uses had C:N ratios in the range9–12 which can be considered an optimum condition at equilib-rium, thus confirming the importance of nitrogen as an indicator ofthe quantity and quality of soil organic matter.

Acknowledgements

The research is part of the Italian research project “SOILSINK”,Climate change and agro-forestry systems: impacts on soil carbonsink and microbial diversity, funded by the Integrated Special Fundfor Research (FISR) of the Italian Ministry of University and Research(D.D. 286, February 20 2006).

References

Anderson, T.H., Domsch, K.H., 1989. Ratios of microbial biomass carbon to totalorganic-C in arable soils. Soil Biol. Biochem. 21, 471–479.

Burke, I.C., Yonker, C.M., Parton, W.J., Cole, C.V., Flach, K., Schimel, D.S., 1989. Texture,climate, and cultivation effects on soil organic matter content in US grasslandsoils. Soil Sci. Soc. Am. J. 53, 800–805.

Campbell, C.A., Lafond, G.P., Biederbeck, V.O., Wen, G., Schoenau, J., Hahn, D., 1999.Seasonal trends in soil biochemical attributes: effects of crop management on aBlack Chernozem. Can. J. Soil Sci. 79, 85–97.

Chen, H., Marhan, S., Billen, N., Stahr, K., 2009. Soil organic-carbon and total nitro-gen stocks as affected by different land uses in Baden-Württemberg (southwestGermany). J. Plant Nutr. Soil Sci. 172, 32–42.

Chiti, T., Gardin, L., Perugini, L., Quaratino, R., Vaccari, F.P., Miglietta, F., Valentini, R.,2012. Soil organic carbon stock assessment for the different cropland land usesin Italy. Biol. Fertil. Soils 48, 9–17.

Conant, R.T., Paustian, K., Elliott, E.T., 2001. Grassland management and conversioninto grassland: effects on soil carbon. Ecol. Appl. 11, 343–355.

Corral-Fernández, R., Parras-Alcantára, L., Lozano-García, B., 2013. Stratificationratio of soil organic C, N and C:N in Mediterranean evergreen oak woodlandwith conventional and organic tillage. Agric. Ecosyst. Environ. 164, 252–259.

9 system

D

F

F

F

F

G

G

G

H

H

H

H

J

K

K

LL

L

L

M

M

M

2 R. Francaviglia et al. / Agriculture, Eco

oro, L., 2009. Sistemi colturali mediterranei e cambiamenti climatici: impatto sulleriserve di carbonio nel suolo. Ph.D. thesis in Agrometeorologia ed Ecofisiologiadei Sistemi Agrari e Forestali. Università degli Studi di Sassari, Facoltà di Agraria,Available at http://eprints.uniss.it/1081/ (accessed October 2013).

rancaviglia, R., Coleman, K., Whitmore, A.P., Doro, L., Urracci, G., Rubino, M., Ledda,L., 2012. Changes in soil organic carbon and climate change – application ofthe RothC model in agro-silvo-pastoral Mediterranean systems. Agric. Syst. 112,48–54.

ranzluebbers, A.J., 2002. Soil organic matter stratification ratio as an indicator ofsoil quality. Soil Till. Res. 66, 95–106.

ranzluebbers, A.J., Hons, F.M., Zuberer, D.A., 1998. In situ and potential CO2 evolu-tion from a Fluventic Ustochrept in southcentral Texas as affected by tillage andcropping intensity. Soil Till. Res. 47, 303–308.

ranzluebbers, A.J., Stuedemann, J.A., 2008. Soil-profile organic carbon and totalnitrogen during 12 years of pasture management in the Southern PiedmontUSA. Agric. Ecosyst. Environ. 129, 28–36.

anuza, A., Almendros, G., 2003. Organic carbon storage in soils of the Basque Coun-try (Spain): the effect of climate, vegetation type and edaphic variables. Biol.Fertil. Soils 37, 154–162.

onzález, J., Candás, M., 2004. Materia orgánica de suelos bajo encinas: miner-alización de carbono y nitrógeno. Invest. Agrar. Sist. Recur. Fuera de serie 3,75–83.

onzález, I., Grau Corbí, J.M., Fernández Cancio, A., Jiménez Ballesta, R., GonzálezCascón, M.R., 2012. Soil carbon stocks and soil solution chemistry in Quercusilex stands in Mainland Spain. Eur. J. For. Res. 131, 1653–1667.

aynes, R.J., Dominy, C.S., Graham, M.H., 2003. Effect of agricultural land use onsoil organic matter status and the composition of earthworm communities inKwaZulu-Natal, South Africa. Agric. Ecosyst. Environ. 95, 453–464.

e, Z.L., Wu, J., O’Donnell, A.G., Syers, J.K., 1997. Seasonal responses in microbialbiomass carbon, phosphorus and sulphur in soils under pasture. Biol. Fertil. Soils24, 421–428.

ernanz, J.L., Sánchez-Girón, V., Navarrete, L., 2009. Soil carbon sequestration andstratification in a cereal/leguminous crop rotation with three tillage systems insemiarid conditions. Agric. Ecosyst. Environ. 133, 114–122.

uber, S., Syed, B., Freudenschuß, A., Ernstsen, V., Loveland, P., 2001. Proposal for aEuropean Soil Monitoring and Assessment Framework. European EnvironmentAgency, Copenhagen, Technical report 61, 58 pp.

enny, H., 1980. The Soil Resource: Origin and behavior. Ecological Studies, vol. 37.Springer, New York, 377 pp.

irchmann, H., Andersson, R., 2001. The Swedish system for quality assessment ofagricultural soils. Environ. Monit. Assess. 72, 129–139.

ottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World map of the Köppen-Geiger climate classification updated. Meteorol. Z. 15, 259–263.

al, R., 2008. Carbon sequestration. Philos. Trans. R. Soc. B 363, 815–830.agomarsino, A., Benedetti, A., Marinari, S., Pompili, L., Moscatelli, M.C., Roggero,

P.P., Lai, R., Ledda, L., Grego, S., 2011. Soil organic C variability and micro-bial functions in a Mediterranean agro-forest ecosystem. Biol. Fertil. Soils 47,283–291.

ou, Y., Xu, M., Chen, X., He, X., Zhao, K., 2012. Stratification of soil organic C, N andC:N ratio as affected by conservation tillage in two maize fields of China. Catena95, 124–130.

ozano-García, B., Parras-Alcántara, L., 2013. Land use and management effects oncarbon and nitrogen in Mediterranean Cambisols. Agric. Ecosyst. Environ. 179,208–214.

artin, M.P., Watterbach, M., Smith, P., Meersmans, J., Jolivet, C., Boulonne, L.,Arrouays, D., 2010. Spatial distribution of soil organic carbon stocks in France.Biogeosci. Discuss. 7, 8409–8443.

elero, S., López-Bellido, R.J., Luis López-Bellido, L., Munoz-Romero, V., Moreno, F.,Murillo, J.M., Franzluebbers, A.J., 2012. Stratification ratios in a rainfed Mediter-

ranean Vertisol in wheat under different tillage, rotation and N fertilisation rates.Soil Till. Res. 119, 7–12.

oreno, F., Murillo, J.M., Pelegrín, F., Girón, I.F., 2006. Long-term impact of conser-vation tillage on stratification ratio of soil organic carbon and loss of total andactive CaCO3. Soil Till. Res. 85, 86–93.

s and Environment 183 (2014) 86– 92

Moscatelli, M.C., Di Tizio, A., Marinari, S., Grego, S., 2007. Microbial indicators relatedto soil carbon in Mediterranean land use systems. Soil Till. Res. 97, 51–59.

Murillo, J.M., Moreno, F., Girón, I.F., Oblitas, M.I., 2004. Conservation tillage: longterm effect on soil and crops under rainfed conditions in south-west Spain(Western Andalusia). Span. J. Agric. Res. 2 (1), 35–43.

Nichols, J.D., 1984. Relation of organic carbon to soil properties and climate in thesouthern Great Plains. Soil Sci. Soc. Am. J. 48, 1382–1384.

Nieto, O.M., Castro, J., Fernández, E., 2011. Long-term effects of residue managementon soil fertility in Mediterranean olive grove: simulating carbon sequestrationwith RothC model. In: Ozkaraova Gungor, B.E. (Ed.), Principles, Application andAssessment in Soil Science. InTech/Rijeka, New York/Shanghai, pp. 129–150.

Nunes, J.S., Araujo, A.S.F., Nunes, L.A.P.L., Lima, L.M., Carneiro, R.F., Salviano, A.A.C.,Tsai, S.M., 2012. Impact of land degradation on soil microbial biomass and activ-ity in Northeast Brazil. Pedosphere 22, 88–95.

Papini, R., Valboa, G., Favilli, F., L’Abate, G., 2011. Influence of land use on organiccarbon pool and chemical properties of vertic cambisols in central and southernItaly. Agric. Ecosyst. Environ. 140, 68–79.

Parton, W.J., Schimel, D.S., Cole, C.V., Ojima, D.S., 1987. Analysis of factors controllingsoil organic matter levels on grasslands. Soil Sci. Soc. Am. J. 51, 1173–1179.

Peregrina, F., Larrieta, C., Ibánez, S., García-Escudero, E., 2010. Labile organic matter,aggregates, and stratification ratios in a semiarid vineyard with cover crops. SoilSci. Soc. Am. J. 74, 2120–2130.

Rasmussen, P.E., Collins, H.P., 1991. Long-term impacts of tillage, fertilizer, and cropresidue on soil organic matter in temperate semiarid regions. Adv. Agron. 45,93–134.

Rivas-Martinez, S., Rivas-Saenz, S., 2009. Worldwide Bioclimatic Classi-fication System, 1996–2009. Phytosociological Research Center, Spainhttp://www.globalbioclimatics.org (accessed December 2012).

Rodeghiero, M., Rubio, A., Díaz-Pinés, E., Romanyà, J., Maranón-Jiménez, S., Levy, G.J.,Fernandez-Getino, A.P., Sebastià, M.T., Karyotis, T., Chiti, T., Sirca, C., Martins, A.,Manuel Madeira, M., Zhiyanski, M., Gristina, L., La Mantia, T., 2011. Soil carbonin Mediterranean ecosystems and related management problems. In: Jandl, R.,Rodeghiero, M., Olsson, M. (Eds.), Soil Carbon in Sensitive European Ecosystems:From Science to Land Management. John Wiley & Sons, Ltd., Chichester, UK, pp.175–218.

Rodríguez-Murillo, J.C., 2001. Organic carbon content under different types of landuse and soil in peninsular Spain. Biol. Fertil. Soils 33, 53–61.

Saxton, K.E., Rawls, W.J., 2006. Soil water characteristic estimates by texture andorganic matter for hydrologic solutions. Soil Sci. Soc. Am. J. 70 (5), 1569–1578.

Saxton, K.E., Willey, P.H., 2006. The SPAW model for agricultural field and pondhydrologic simulation. In: Singh, V.P., Frevert, D.K. (Eds.), Watershed Models.CRC Press, Boca Raton, FL, pp. 401–435.

Schnabel, R.R., Franzluebbers, A.J., Stout, W.L., Sanderson, M.A., Stuedemann, J.A.,2001. The effects of pasture management practices. In: Follett, R.F., Kimble, J.M.,Lal, R. (Eds.), The Potential of US Grazing Lands to Sequester Carbon and Mitigatethe Greenhouse Effect. Lewis Publishers, Boca Raton, FL, pp. 291–322.

Thomsen, I.K., Petersen, B.M., Bruun, S., Jensen, L.S., Christensen, B.T., 2008. Estimat-ing soil C loss potentials from the C to N ratio. Soil Biol. Biochem. 40, 849–852.

USDA-NRCS (United States Department of Agriculture-Natural Resources Conser-vation Service), 1995. Soil Survey Laboratory Information Manual. Soil SurveyInvestigations Report No. 45. National Soil Survey Center, Soil Survey Laboratory,Lincoln.

Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuringsoil microbial biomass C. Soil Biol. Biochem. 19, 703–707.

Waldrop, M.P., Firestone, M.K., 2006. Seasonal dynamics of microbial communitycomposition and function in oak canopy and open grassland soils. MicrobialEcol. 52, 470–479.

West, T.O., Post, W.M., 2002. Soil organic carbon sequestration rates by tillage andcrop rotations: a global data analysis. Soil Sci. Soc. Am. J. 66, 1930–1946.

WRB, 2006. World Reference Base for Soil Resources World Soil Resources Report103. IUSS, ISRIC, FAO, Rome, 145 pp.

Zhang, P.J., Li, L.Q., Pan, G.X., Ren, J.C., 2006. Soil quality changes in land degradationas indicated by soil chemical, biochemical and microbiological properties in akarst area of southwest Guizhou, China. Environ. Geol. 51, 609–619.