effects of reduced tillage on soil surface properties affecting wind erosion in semiarid fallow...

European Journal of Agronomy 12 (2000) 191–199

Effects of reduced tillage on soil surface properties affectingwind erosion in semiarid fallow lands of Central Aragon

M.V. Lopez *, R. Gracia, J.L. ArrueDepartamento de Edafologıa, Estacion Experimental de Aula Dei, Consejo Superior de In6estigaciones Cientıficas (CSIC),

PO Box 202, 50080 Zaragoza, Spain

Received 20 July 1999; received in revised form 21 January 2000; accepted 21 January 2000

Abstract

In Central Aragon (NE Spain), where strong and dry winds are frequent all year round, fallow lands are susceptibleto wind erosion due to insufficient crop residues on the surface and loose, finely divided soils by multiple tillageoperations. Effects of conventional tillage (CT — mouldboard ploughing followed by a compacting roller) andreduced tillage (RT — chisel ploughing) on soil surface properties affecting wind erosion were studied during threeexperimental campaigns in a dryland field of Central Aragon. RT provided higher soil protection than CT througha lower wind erodible fraction of soil surface (on average, 10% less) and a significantly higher percentage of soil coverwith crop residues and clods (30% higher). Random roughness was also higher after RT than after CT (15 vs. 4%).These results indicate that RT can be an effective soil management practice for wind erosion prevention during thefallow period in semiarid Aragon. The study shows, likewise, that significant changes in soil aggregate sizedistribution associated with wind erosion processes may occur in short periods of time. Thus, temporal variability ofsoil surface properties, including crust and clods stability, needs to be considered in wind erosion research inagricultural soils. © 2000 Elsevier Science B.V. All rights reserved.

Keywords: Dryland farming; Fallowing; Conservation tillage; Soil erodibility by wind; Soil surface properties

www.elsevier.com/locate/euragr

1. Introduction

Achieving sustainable agroecosystems is a ma-jor challenge in semiarid regions. Due to particu-lar soil and climate conditions and unsuitableagronomic practices, semiarid drylands of Central

Aragon (NE Spain) are prone to land degradationby wind and water erosion. While soil loss bywater erosion has been well documented (Benitoet al., 1992; Navas, 1993; Lasanta et al., 1995),wind erosion has not been yet considered as aserious degradation problem in this region. How-ever, recent field studies, providing preliminaryresults on dust emission (Lopez et al., 1998) andsaltation transport (Sterk et al., 1999), indicatethat the risk of severe wind erosion could be highin agricultural soils of Central Aragon.

* Corresponding author. Tel.: +34-9-76576511; fax: +34-9-76575620.

E-mail address: [email protected] (M.V. Lopez)

1161-0301/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.

PII: S 1161 -0301 (00 )00046 -0

M.V. Lopez et al. / Europ. J. Agronomy 12 (2000) 191–199192

About three quarters (more than 1 millionhectares) of the rainfed arable land in Aragonreceive an average annual precipitation of 500 mmor less. The rainfall regime is characterized by theabsence of a well-defined rainy season. In anymonth there is the strong probability of havingeither an extremely low amount (B10 mm) or norain at all (McAneney and Arrue, 1993). Strongand dry WNW winds (Cierzo) are frequent andcharacteristic for Central Aragon. Wind eventswith gusts over 30 m s−1 are common, especiallyin summer (Biel and Garcıa de Pedraza, 1962).Soils are mostly alkaline (pH\8), with low or-ganic matter content (B20 g kg−1), high totalcarbonate content (\300 g kg−1 in many cases)and a dominant sandy loam to loam textural class(Montanes et al., 1991). Because of the fragilityand high susceptibility to degradation of soils inthe region, particularly the gypsiferous soils(Navas, 1990), especial attention must be paid totheir agricultural use and proper managementduring the fallow period. Long fallowing (16–17months), in the traditional cereal–fallow rotation,may enhance wind erosion due to insufficientresidues on the surface and the highly pulverisedsoils caused by multiple tillage operations. Conse-quences of wind erosion include a reduction incrop production by selective removal of the finestsoil particles, rich in nutrients and organic matter,a reduction in the water-holding capacity andincreased degradation of soil structure (Lowery etal., 1995; Larney et al., 1998).

All the above considerations were of majorconcern in planning the project Wind Erosion andLoss of Soil Nutrients in Semi-Arid Spain (WEL-SONS), initiated in 1996 in order to provide apredictive understanding of impacts driven by cli-mate and land-use changes on the degradation ofagricultural soils by wind erosion in CentralAragon (Gomes et al., 1996). According to previ-ous results on soil and crop response to conserva-tion tillage in cereal production areas of Aragon(Lopez et al., 1996; Lopez and Arrue, 1997),evaluation of reduced tillage (RT) as a soil man-agement practice for wind erosion prevention infallow lands was included among the objectives ofthat project. The purpose of this paper is to showthe effects of tillage on soil surface properties

affecting wind erosion during three intensive fieldcampaigns in an agricultural soil of CentralAragon. Soil surface conditions following mould-board ploughing and chiseling, as primary opera-tions in conventional and reduced tillage systems,respectively, are compared and discussed on thebasis of their temporal variability.

2. Materials and methods

2.1. Site characteristics

The study was carried out within an agricul-tural field in the Los Monegros area (41°36%N,0°32%W, 280 m alt.), �35 km east of Zaragoza(Fig. 1). The climate of the area is semiarid withan average annual rainfall of 380 mm and anaverage annual air temperature of 14.3°C. Theexperimental field is almost level within an undu-lating landscape. The soil is a silt loam (coarse-loamy, gypsic, thermic Gypsiorthid) (Soil SurveyStaff, 1975). Selected soil properties for twodepths in the plough layer (0–20 and 20–40 cm)are shown in Table 1.

The experimental field has been conventionallytilled with barley (Hordeum 6ulgare L.) grownunder cereal–fallow rotation for at least the previ-ous 10 years. Accordingly, half of the farm field iscropped while the other half is fallowed, alternat-ing these areas each year. Average annual cerealgrain yield in Los Monegros is less than 2 t ha−1

(Austin et al., 1998).

2.2. Experimental plan

Three intensive field campaigns were made dur-ing the summers of 1996 (July–September), 1997(June–September) and 1998 (June–July). In 1996and 1997, two adjacent plots of 135×180 m2,with a separation of 20 m, were delimited in thefallow area of the field for the application of twotillage treatments: conventional tillage (CT), con-sisting of mouldboard ploughing to 30–35 cmdepth followed by a pass of a compacting roller (atraditional practice in the study area), and re-duced tillage (RT), with only a single pass ofchisel plough to 15–20 cm depth, as an alternative

M.V. Lopez et al. / Europ. J. Agronomy 12 (2000) 191–199 193

Fig. 1. Location of the study area and layout of the experimental plots (CT, conventional tillage; RT, reduced tillage) with respectto the prevailing Cierzo wind direction.

conservation tillage system. In 1998, only the CTtreatment was considered and implemented on aplot of the same dimension. All tillage operationswere done in the prevailing wind direction(WNW). Fig. 1 shows the layout of the experi-mental plots during the study period. In 1997 and1998, tillage operations (pass of roller on the CTplot and chisel ploughing on the RT plot) wererepeated to disrupt the surface crust formed afterrainfall events, as required by ongoing simulta-neous experiments on dust emission and saltationprocesses. Repeated tillage is commonly practisedby local farmers in order to remove weeds grow-ing during the fallow period.

2.3. Sampling and measurements

Prior to preparation of the CT and RT plots,soil samples were taken from 0–20 and 20–40 cmdepths on the experimental field in order to havea first indication of soil composition (Table 1).Particle size distribution was determined by usinga light-scattering particle size analyser (COUL-TER LS 230). Organic matter content, CaCO3

content, gypsum content, electrical conductivity

(H2O, 1:5) and pH (H2O, 1:2.5) were determinedby standard methods (Page et al., 1982).

Soil surface properties affecting wind erosionwere measured immediately after the applicationof the tillage treatments. Soil samples for dryaggregate size distribution were collected from0–2.5 cm depth using a metal frame (15×15 cm2)with a cutting edge. The samples were transportedcarefully to the laboratory where they were air-

Table 1Selected physical and chemical properties of the soil at theexperimental site

Depth (cm)

20–400–20

Particle size distribution (g kg−1)Sand (2000BfB50 mm) 208 219Silt (50BfB2 mm) 623633

158159Clay (fB2 mm)

8.1 8.1pH (H2O, 1:2.5)Electrical conductivity (1:5) (dS m−1) 2.0 2.1

15.9Organic matter (g kg−1) 14.8354352CaCO3 (g kg−1)

171Gypsum (g kg−1) 177


Fig. 2. Daily precipitation and average daily wind speed at 2 m height measured at the field site during the experimental periods.

dried and sieved with an electromagnetic sieveshaker (CISA, Barcelona). The soil was separatedin fourteen size fractions: 38–12.5, 12.5–8, 8–6.3,6.3–2, 2–1, 1–0.84, 0.84–0.5, 0.5–0.4, 0.4–0.25,0.25–0.1, 0.1–0.08, 0.08–0.063, 0.063–0.04 andB0.04 mm in diameter. Aggregate size distribu-tion was obtained by using the data inversionmethod described by Gomes et al. (1990). Thisfitting procedure is based on the adjustment ofmultimodal log–normal distributions to the mea-sured values by minimizing the difference betweenthe simulated and observed populations of eachsize class. Each mode of the mass–size distribu-tion is characterized by three parameters: massmedian diameter (MMD), standard deviation(S.D.) and amplitude (% mass). In order to evalu-ate the temporal variation of soil erodibility, thedry aggregate size distributions from soil samplestaken in the CT plot immediately after tillage onJuly 8 in 1998 were compared with those collectedon July 16 following a dry and windy period. Todetect more efficiently differences between dates,soil samples from July 16 were shallower (0–1cm) than those from previous samplings (0–2.5cm). Assuming that, immediately after tillage, theaggregate size distribution of the soil in the upper2.5 cm and 1 cm is the same, direct comparisonsbetween dates are possible.

Soil surface roughness was measured in theWNW direction (random roughness) using thechain method (Saleh, 1993). Frontal and basalareas occupied by clods (aggregates \38 mm indiameter), crop residues and pebbles were esti-

mated with a 10×10 cm2 grid within a 1×1 m2

frame. Frontal area refers to the lateral surface ofthese elements exposed to dominant wind per unitof ground cover and basal area to their horizontalsurface expressed as percentage of total soil sur-face. In 1997 and 1998, soil crust was character-ized in terms of thickness, maximum crustpenetration resistance, measured by a hand sur-face penetrometer (Type IB, Eijkelkamp, TheNetherlands), and amount of loose aggregateslying on the crust by using an ordinary vacuum-cleaner.

All above determinations were made at 12points randomly selected in each plot. Hourlyvalues of meteorological parameters (precipita-tion, air temperature, wind speed and wind direc-tion, relative humidity and global solar radiation)were registered through the whole experimentalperiod with an automatic weather station(CAMPBELL Scientific, datalogger CR10) in-stalled in the experimental field.

3. Results

Total rainfall received during the experimentalperiods of 1996, 1997 and 1998 was 60, 171 and62 mm, respectively (Fig. 2). As compared withthe long-term average precipitation (1965–1995)registered for the same seasonal periods at theMonegrillo pluviometric station (�10 km distantfrom the site), the rainfall in the 1996 and 1998campaigns was 17% lower and 23% higher than


Table 2Soil surface characteristics as affected by tillage (CT, conventional tillage; RT, reduced tillage)

Tillage date Tillage Frontal areaField Wind erodible RandomSoil cover(cm2 m−2)(%)campaign roughness (%)btreatment fraction (%)a

Residues Clodsc Residues Clodsc

CT 43ad 0.80a1996 1.24aJuly 8 80a 65a 3.5aRT 33b 25.21b 24.78b 1970b 739b 13.1b

1997 June 16 CT 39a 0.18a 4.06a 6a 216a 6.4aRT 27b 13.29b 11.13b 483b 967b 14.3bCT 40a 0.09a 0.25a 2aSeptember 3 15a 2.0aRT 37a 9.33b 12.58b 323b 954b 16.7b

CT 43 3.42 4.63 103 243 2.61998 June 22CT 48 1.00 2.15July 8 16 9 1.4

a Mass of aggregates B0.84 mm in diameter (0–2.5 cm depth).b Measured in the WNW direction (292.5°).c Aggregates \38 mm in diameter.d At the same date, means followed by the same letter are not statistically different at PB0.05.

average, respectively, while in 1997 it was well-above average (61% higher). Frequent showersand rainstorms, particularly in 1997, affected to alarge extent the course of the wind erosion experi-ment due to soil surface crusting, as detailedbelow. In addition, many of the days with highwind speed were preceded by rainfall during theprevious one or two days (Fig. 2). Wind speedvaried among months and experimental cam-paigns. On the basis of the number of hours permonth with a mean wind speed equal to or higherthan 5 m s−1 at 2 m height, the month with thehighest frequency of erosive winds was July 1998(35% of time). The least windy months were Au-gust and September of 1997 with wind speedshigher than 5 m s−1 for only 4–5% of time.

Table 2 shows the soil surface conditions afterCT and RT for the different dates of tillageoperations during the experimental period. Withthe exception of the second date in 1997, the winderodible fraction (aggregates B0.84 mm in di-ameter) was significantly higher (PB0.05) afterCT (40%) than after RT (30%). In the RT plot,additional chiseling applied on September 3 in-creased the fraction of erodible aggregates to thatof CT. In all cases, the aggregate size distributionwas characterized by three populations (modes)

with MMD ranging from 12 to 17 mm for thefirst, 1.2–3.2 mm for the second and 0.17–0.31mm for the third population, respectively. As canbe seen in Fig. 3, corresponding to the aggregatedistribution after tillage on June 16 in 1997, theMMD was slightly higher after RT than after CTfor the three populations. On the other hand, interms of amplitude (% mass), the third populationwas, in general, the most affected by tillage with afraction of the smallest aggregates about 6%higher under CT than RT. However, the majordifferences between treatments concerned soil sur-face cover and roughness created by crop residuesand clods; the presence of pebbles was negligible

Fig. 3. Dry aggregate mass–size distribution (B38 mm diame-ter) of soil in the 0–2.5 cm depth after conventional tillage(CT) and reduced tillage (RT) on June 16, 1997.


Table 3Soil crust thickness, maximum crust penetration resistance and amount of loose aggregates on the crust surface under conventionaltillage (CT) and reduced tillage (RT)

Tillage Thickness Penetration resistanceDateField campaign Aggregate mass (g m−2)(mm)treatment (N cm−2)

Total B0.84 mm

1997 CTJuly 28 4.76aa 61a 84.62a 11.63aRT 5.01a 42b 15.85b 6.01b

July 71998 CT 7.40 79 54.19 17.70

a Means followed by the same letter are not statistically different at PB0.05.

Table 4Statistical parameters of the aggregate mass–size distributions (B38 mm diameter) of soil surface immediately after CT (July 8,1998) and 8 days later (July 16, 1998)

Population 2Date Population 3Population 1a

MMD S.D.c Amplitude AmplitudeMMD S.D.S.D. Amplitude MMD(%)(mm) (mm) (%)(mm)b (%)

1.50July 8 13.211.49 1.91 4.15 60.4 0.17 2.29 26.41.59July 16 27.59.06 1.79 4.03 60.7 0.11 1.48 11.8

a Each population identifies one mode of the mass–size distribution.b Mass median diameter.c Standard deviation.

in both treatments. Thus, RT provided a moreprotective soil surface than CT, with a percentageof soil cover in all cases significantly higher underRT than under CT (on average, 30% higher)(Table 2). This difference was attributed to both ahigher presence of residues and a higher numberof large clods (4–10 cm diameter). On the otherhand, the frontal surface of these roughness ele-ments can result in an effective soil protectionfactor during wind erosion events by reducingwind energy and trapping soil erodible particles.RT provided a total frontal area of nonerodiblematerial almost 20 and 7 times higher than CT,after the initial tillage operations of 1996 and1997, respectively. This difference between treat-ments was greatly increased after the repeatedtillage in September 1997 (Table 2). In 1997 and1998, an extra pass of the compacting roller overthe CT plot considerably reduced the low soilprotection provided initially by this treatment.

Table 3 shows the characteristics of the soilcrust formed after intermittent rainfalls receivedat the end of June and during July in 1997 andafter the rainfall on July 1 in 1998 (Fig. 2). In thetwo years, a thin and consolidated crust wasformed, with higher values of penetration resis-tance and mass of both total and erodible aggre-gates in the CT treatment.

The average statistical parameters of the mass–size distributions for soil aggregates collected onJuly 16 and 8 in 1998 to evaluate the temporalvariation of soil erodibility are summarised inTable 4. Although the three populations (modes)remained with time, clear differences in their con-tributions were observed. Thus, the proportion ofthe first population increased from 13 to 27%from the first to the second date and the reverseoccurred for the third population (from 26 to12%). The wind erodible fraction was significantlyreduced from 48 to 31% (PB0.05).


4. Discussion

In a previous study on a different agriculturalsoil of the Zaragoza province, Lopez et al. (1998)have observed more favourable soil surface condi-tions against wind erosion after RT than afterCT. Although further comparisons with thepresent study are difficult, because in the fieldexperiment from Lopez et al. (1998) chiseling andmouldboard ploughing were implemented consecu-tively in the same plot, in both cases the lowererodibility of the soil under RT was based onreducing the wind erodible fraction of soil surfacebut, especially, on creating a rough and protectiveground cover. Horning et al. (1998), using fieldwind tunnel data, have established a mathematicalrelationship to estimate the combined effect ofresidue surface cover and random roughness onsoil losses by wind erosion. Application of thisequation to data reported here (Table 2) indicatesthat the potential reduction of these losses, rela-tive to flat, bare soil, is 30–70% higher in RT thanin CT. These results agree with those of Lopez etal. (1998), who reported �50% less dust emissionby wind (vertical dust flux) after chiseling thanafter mouldboard ploughing for erosion episodeswith a mean wind speed ranging from 5 to 8 ms−1 at 2 m height. Likewise, saltation transportdata recorded in the experimental field of LosMonegros during the 1996 and 1997 campaigns(Sterk et al., 1999) support the general observa-tion of a lower susceptibility of soil to winderosion with RT than with CT. Although winderosion was likely limited due to exceptional wetconditions and soil crusting during the experimen-tal periods, saltation flux was registered in the CTplots during several Cierzo wind events but neverin the RT plots (Sterk et al., 1999).

A more resistant soil condition against windenergy could have been achieved by tilling perpen-dicularly to the dominant wind direction. In thiscase, the surface roughness provided by clods andresidues (random roughness) is increased in thatdue to the ridges produced by tillage (orientedroughness). This is possible for RT because chiselploughing created ridges 10 cm high at 50-cmintervals, but not for CT where the compactingroller eliminates the oriented roughness left by the

mouldboard plough. In any case, and regardlessof the kind of tillage implements commonly usedin a given area, farmers in the region do not takeinto account the tillage orientation with respect toCierzo. In fact, only 15 out of a total of 66agricultural fields surveyed in semiarid Aragon,had tillage operations perpendicular to the pre-vailing wind direction. The range in the directionof ridges in these fields was between 0° (north)and 45° (north-east) (unpublished data).

As discussed by Sterk et al. (1999), the occur-rence of saltation transport at the study site wasinfluenced strongly by the intensity and distribu-tion of rainfall during the experimental period.This was mainly due to soil surface crusting ratherthan to soil moisture because of high evaporationduring the summer season. The low presence ofloose material lying on the surface under RT, inspite of a lower crust resistance to abrasion ascompared with CT (i.e. smaller crust strength),was likely due to the protective effect of cropresidues and roughness on the underlying parti-cles. However, regardless of tillage, the initialprotection from wind erosion provided by crust-ing, as well as by cloddiness, can be temporary.The surface of both crust and clods can, withtime, become a significant source of erodible ma-terial from the abrasive action of wind-blown soilparticles (Cahill et al., 1996; Mirzamostafa et al.,1998). Soil texture has a large influence on the dryaggregate and crust stabilities, with the coarse-tex-tured soils less resistant to breakdown (Potter,1990; Skidmore and Layton, 1992). In this re-spect, the high content of CaCO3 in our soils isprobably an important factor of erodibility to addto the fragile nature of a gypsiferous soil. Whilenoncalcareous soils with a silt loam texture arenot, in general, highly erodible, the presence ofCaCO3 increases their erodibility by reducing themechanical stability of clods and producing amore disaggregated surface (Gillette, 1988; Bre-uninger et al., 1989).

The temporal variability of soil conditions mustbe also taken into account in wind erosion studieswhen the surface of agricultural soils is bare andloose. Furthermore, Lopez (1998) indicated thatsignificant changes in soil erodibility associatedwith wind erosion processes may occur in short


periods of time, it being possible to reach a situa-tion of limited supply of particles available forerosion. This temporal variation can be evaluatedfrom changes in soil aggregate size distributionafter a dry and windy period, as it occurred in ourstudy period (July 8–16, 1998). During this pe-riod, no rainfall was received and the averagedaily wind speed was near or higher than 5 m s−1

in 5 of these 8 days, ranging the maximum aver-age hourly wind speed registered each day be-tween 5.1 and 11.4 m s−1 (Fig. 2). The changesobserved in soil aggregate mass–size distribution,that is, an increase in the fraction of the greatestaggregates, at the expense of a decrease in that ofthe smallest ones, with a reduction in the winderodible fraction (Table 4), reflect a progressivedepletion of fine particles from the soil surfaceand exposure of nonerodible aggregates undersignificant Cierzo wind episodes. These resultsshow that, under the influence of the wind erosionprocess itself, soil erodibility in agricultural fieldscan substantially change in relatively short peri-ods of time.

5. Conclusion

Results from the characterisation of soil surfaceconditions after tillage showed that chiseling, asprimary tillage operation, was more effective thanmouldboard ploughing followed by a pass ofcompacting roller in both creating a protectiveground cover against wind erosion, through acombination of clods and crop residues, and re-ducing the wind erodible fraction of the soil sur-face. The lower soil erodibility by wind under RTindicates that this tillage system can be consideredas a suitable soil management practice to preventwind erosion during the fallow period in the studyarea. Reduction in the number of tillage opera-tions and the consideration of tillage orientationwith respect to the prevailing wind directionshould also be recommended.

On the other hand, since soil crusting is acommon feature of soils in the area, further re-search on crust stability is needed to assess itstemporal character as a protecting element againstwind erosion on fallow land. Similarly, clods cre-

ated by tillage should be studied as a potentialsource of erodible particles via breakdown andabrasive action of saltating particles during winderosion events. In this sense, this study shows thatsignificant changes in the soil aggregate size distri-bution associated with wind erosion processesmay occur in short periods of time. Therefore, thetemporal variability of soil surface propertiesmust be considered in wind erosion research inagricultural soils to correctly assess the extensionof the problem and to design adequate controlmeasures.

Acknowledgements

This work was supported by the WELSONSproject funded by the European Union (contractno. ENV4-CT95-0182). The first author is gratefulto the Spanish Ministry of Education and Culturefor her PNFPI contract.

References

Austin, R.B., Cantero-Martınez, C., Arrue, J.L., Playan, E.,Cano-Marcellan, P., 1998. Yield–rainfall relationships incereal cropping systems in the Ebro river valley of Spain.Eur. J. Agron. 8, 239–248.

Benito, G., Gutierrez, M., Sancho, C., 1992. Erosion rates inbadland areas of the central Ebro Basin (NE-Spain).Catena 19, 269–286.

Biel, A., Garcıa de Pedraza, L., 1962. El clima en Zaragoza yensayo climatologico para el Valle del Ebro. Ministerio delAire, Servicio Meteorologico Nacional. Publicaciones SerieA (Memorias) No. 36, Madrid, 57 p.

Breuninger, R.H., Gillette, D.A., Kihl, R., 1989. Formation ofwind-erodible aggregates for salty soils and soils with lessthan 50% sand composition in natural terrestrial environ-ments. In: Leinen, M., Sarnthein, M. (Eds.), Paleoclimatol-ogy and Paleometeorology: Modern and Past Patterns ofGlobal Atmospheric Transport. Kluwer Academic Publish-ers, Dordrecht, pp. 31–63.

Cahill, T.A., Gill, T.E., Reid, J.S., Gearhart, E.A., Gillette,D.A., 1996. Saltating particles, playa crusts and dust aero-sols at Owens (dry) Lake, California. Earth Surf. Proc.Landforms 21, 621–639.

Gillette, D.A., 1988. Threshold friction velocities for dustproduction for agricultural soils. J. Geophys. Res. 93,12645–12662.

Gomes, L., Bergametti, G., Dulac, F., Ezat, U., 1990. Assess-ing the actual size distribution of atmospheric aerosols


collected with a cascade impactor. J. Aerosol Sci. 21,47–59.

Gomes, L., Gaudichet, A., Arrue, J.L., Lopez, M.V., Sterk,G., Hollander, W., Frangi, J.P., 1996. Wind Erosion andLoss of Soil Nutrients in Semiarid Spain (WELSONS). In:Book of Abstracts. International Conference on Mediter-ranean Desertification. Research Results and Policy Impli-cations, Crete-Hellas, Greece.

Horning, L.B., Stetler, L.D., Saxton, K.E., 1998. Surfaceresidue and soil roughness for wind erosion protection.Trans. ASAE 41, 1061–1065.

Larney, F.J., Bullock, M.S., Janzen, H.H., Ellert, B.H., Olson,E.C.S., 1998. Wind erosion effects on nutrient redistribu-tion and soil productivity. J. Soil Water Cons. 53, 133–140.

Lasanta, T., Perez-Rontome, C., Garcıa-Ruiz, J.M., Machın,J., Navas, A., 1995. Hydrological problems resulting fromfarmland abandonment in semi-arid environments: theCentral Ebro Depression. Phys. Chem. Earth 20, 309–314.

Lopez, M.V., 1998. Wind erosion in agricultural soils: anexample of limited supply of particles available for erosion.Catena 33, 17–28.

Lopez, M.V., Arrue, J.L., 1997. Growth, yield and water useefficiency of winter barley in response to conservationtillage in a semiarid region of Spain. Soil Till. Res. 44,35–54.

Lopez, M.V., Arrue, J.L., Sanchez-Giron, V., 1996. A com-parison between seasonal changes in soil water storage andpenetration resistance under conventional and conserva-tion tillage systems in Aragon. Soil Till. Res. 37, 251–271.

Lopez, M.V., Sabre, M., Gracia, R., Arrue, J.L., Gomes, L.,1998. Tillage effects on soil surface conditions and dustemission by wind erosion in semiarid Aragon (NE Spain).Soil Till. Res. 45, 91–105.

Lowery, B., Swan, J., Schumacher, T., Jones, A., 1995. Physi-

cal properties of selected soils by erosion class. J. SoilWater Cons. 50, 306–311.

McAneney, K.J., Arrue, J.L., 1993. A wheat–fallow rotationin northeastern Spain: water balance–yield considerations.Agronomie 13, 481–490.

Mirzamostafa, N., Hagen, L.J., Stone, L.R., Skidmore, E.L.,1998. Soil aggregate and texture effects on suspensioncomponents from wind erosion. Soil Sci. Soc. Am. J. 62,1351–1361.

Montanes, L., Sanz, M., Heras, L., 1991. Fertilidad de lossuelos de secano de la provincia de Zaragoza. DiputacionGeneral de Aragon, Departamento de Agricultura,Ganaderıa y Montes. Serie Estudios Agrarios, Zaragoza,233 p.

Navas, A., 1990. The effect of simulated runoff on the erosionof gypsiferous soils. Land Degrad. Rehabil. 2, 117–126.

Navas, A., 1993. Soil losses under simulated rainfall in semi-arid shrublands of the Ebro Valley, Spain. Soil Use Man-age. 9, 152–157.

Page, A.L., Miller, R.H., Keeney, D.R., (Eds.), 1982. Methodsof Soil Analysis: Part 2. Chemical and MicrobiologicalProperties, 2nd edn. Agronomy No. 9. ASA, Madison, WI.

Potter, K.N., 1990. Estimating wind-erodible materials onnewly crusted soils. Soil Sci. 150, 771–776.

Saleh, A., 1993. Soil roughness measurement: chain method. J.Soil Water Cons. 48, 527–529.

Skidmore, E.L., Layton, J.B., 1992. Dry-soil aggregate stabil-ity as influenced by selected soil properties. Soil Sci. Soc.Am. J. 56, 557–561.

Soil Survey Staff, 1975. Soil taxonomy: a basic system of soilclassification for making and interpreting soil surveys.USDA-SCS Agricultural Handbook 436. US GovernmentPrinting Office, Washington, DC, USA.

Sterk, G., Lopez, M.V., Arrue, J.L., 1999. Saltation transporton a silt loam soil in north east Spain. Land Degrad. Dev.10, 545–554.

..

effects of reduced tillage on soil surface properties affecting wind erosion in semiarid fallow...

Documents