Structure and hydraulic properties of the boreal ... - Wiley Online Library

10 downloads 1357 Views 887KB Size Report
differently managed BZs established in a boreal Vertic Stagnic Cambisol (clay, 51%). The three management practices for vegetation were as follows: natural ...
SoilUse and Management doi: 10.1111/sum.12043

Soil Use and Management, September 2013, 29, 410–418

Structure and hydraulic properties of the boreal clay soil under differently managed buffer zones K. R A S A 1 , 3 & R. H O R N 2 1

Department of Food and Environmental Sciences, 00014 University of Helsinki, Latokartanonkaari 11, Helsinki, Finland, and Institute of Plant Nutrition and Soil Science, Christian-Albrechts-Universita¨t, Hermann-Rodewald-Str. 2, 24118 Kiel, Germany

2

Abstract Vegetated buffer zones (BZs) between arable fields and bodies of water are commonly established to reduce erosion and run-off of particle-bound nutrients. Functioning of a BZ depends on soil structure, as it is important for water infiltration. Therefore, it is vital to understand how varying management practices affect soils of BZs. We studied the structural and hydraulic properties of three differently managed BZs established in a boreal Vertic Stagnic Cambisol (clay, 51%). The three management practices for vegetation were as follows: natural with no treatment, harvested yearly and grazed by cattle. We used bulk density and macroporosity, together with a pore geometry index (air permeability per unit air-filled porosity), to describe the soil structural properties. Hydraulic properties were measured at different length scales by means of an aggregate sorptivity test, saturated hydraulic conductivity of the core samples and field-saturated hydraulic conductivity. Vegetation management markedly affected the physical properties in the top 5 cm of the soil. Properties were least favourable for infiltration at the grazed site, with the greatest bulk density, least macroporosity and hydraulic conductivity or greatest pore tortuosity. In general, spatial variation in zones with restricted and good hydraulic conductivity together with reduced aggregate sorptivity in the deeper horizons made the soil prone to preferential flow when initially dry. Prolonged wetness, on the other hand, reduced saturated hydraulic conductivity significantly, resulting in surface run-off. Harvesting was considered the best management practice due to its inherent capacity for reducing the soil nutrient content and because it has minor implications for soil physical properties.

Keywords: Buffer zone, pore geometry index, soil structure, infiltration, air permeability, hydraulic conductivity

Introduction Successful implementation of the EU Water Framework Directive requires effective adoption of mechanisms to reduce pollution of water resources. Vegetated buffer zones (BZs), established between sloping agricultural fields and bodies of water, reduce soil erosion and nutrient leaching (Hoffmann et al., 2009). Although their efficiency in reducing agricultural loading has been widely studied (e.g. Uusi-Ka¨mppa¨ et al., 2000; Syversen, 2005), the structural and hydraulic properties of BZs have not been as broadly investigated. These properties control water entry into the soil and determine whether infiltration or surface run-off Correspondence: K. Rasa, E-mail: [email protected] 3 Present address: MTT Agrifood Research Finland 31600 Jokioinen Finland Received April 2011; accepted after revision February 2013

410

occurs. To define the best management practices for BZs, attention must be focused on soil physical properties, which are fundamental for water infiltration in the BZs. The surface horizon is the most active part of the soil in terms of biological, chemical and physical processes. Cultivated soils are annually loosened to facilitate root growth, while natural aggregate formations, accessibility to exchange surfaces, as well as biological mineralization processes, are interrupted annually. The least affected parts of arable land are the BZs, where soils are under permanent vegetation. The structure of these surface soils is, however, prone to artificial stresses, such as tractor wheeling and cattle trampling. The intensity of these stresses is dependent on management practice, varying from no stress (unmanaged BZs) to intensive grazing (Pietola et al., 2005). Moreover, the abiotic stresses, including shrink–swell and freeze– thaw cycles, are most intense and frequent in the topsoil (Rasa et al., 2009). Therefore, the structural properties of

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science

Physical properties of clayey surface soil

surface horizons are of special interest when the hydraulic properties and environmental effects of soils without annual tillage are considered. The importance of macroporosity (pores >30 lm) and especially large macropores (pores >300 lm) for drainage of clayey soils in Finland is well recognized (Aura, 1983; Alakukku, 1998; Pietola et al., 2005). In well-structured soil with abundant macropores, rapid infiltration reduces surface run-off and accompanied soil erosion. On the other hand, rapid infiltration may also facilitate transport of agrochemicals, including inorganic and organic fertilizers, and particle-bound nutrients to subsurface drains (Uusitalo et al., 2001; Jarvis, 2007). Uniform infiltration, however, allows sorption processes and sedimentation of particle-bound nutrients as water flows through the soil profile. Therefore, soil structure conducive to water infiltration is considered as a prerequisite for well-functioning BZs (Hoffmann et al., 2009). Field-scale studies reveal significant seasonal variation in nutrient retention of the BZs (Syversen, 2005; Uusi-Ka¨mppa¨ & Jauhiainen, 2010). The increase in run-off water and sediment tends to take place in spring and autumn. The main reason is excess water in these periods (snowmelt water, autumn rains and little evapotranspiration). However, it is likely that moisture-related changes in soil hydraulic properties play an important role in the generation of run-off flow as well (Messing & Jarvis, 1990; Uusi-Ka¨mppa¨ & Jauhiainen, 2010). Soil management of BZs differs from practices common in fields under active production, including natural unmanaged BZs and long-lasting harvesting without resowing. The objective of the study was to compare soil physical properties under different management practices and to evaluate the effect of management on water infiltration. We studied the surface horizons (10 cm) of three differently managed BZs established on clayey soil in a boreal climate. Indirect attributes, such as bulk density and macroporosity, as well as a pore geometry index derived from air permeability and airfilled porosity (Blackwell et al., 1990), were used to describe the structural properties of soil. Small-scale water sorptivity was measured to assess aggregate contribution to water adsorption. In addition, hydraulic properties were measured in the laboratory and field, which allowed observation of these properties under different length scales and circumstances. We hypothesized that the structural and physical properties of differently managed surface soils are of importance for the effectiveness of filtering and buffering processes in clayey soils close to streams. Better understanding of these soil physical properties provides valuable input data, for example, for models accounting for efficiency of BZs in reducing nutrient loading from agricultural fields. The physical grounds for maintenance instructions of BZs have been less emphasized in the literature, and therefore, this study supports development of proper management practices for BZs in boreal climates as well as improving the understanding of their functioning.

411

Materials and methods Study site The experimental field was situated at Jokioinen, southwestern Finland (60° 48′N, 23° 28′E), on the premises of MTT Agrifood Research Finland in an area characterized by a prevailing boreal climate. The experiment on vegetated BZs was initiated in 1991 (Uusi-Ka¨mppa¨ & Jauhiainen, 2010). The soil was classified as a Vertic Stagnic Cambisol (Eutric) (FAO, 2006) or as a Typic Cryaquept (Soil Survey Staff, 2010). It consisted of 51% clay, 42% silt and 7% sand in the Ap horizon. The cation exchange capacity determined with 1 M ammonium acetate (pH 7) was 37 cmolc/kg. The content of soil organic carbon varied between 2.1 and 5.1%. Three BZs (former cultivated fields) were sampled in May 2005. These BZs had been subjected to the following management practices (names in parentheses are used later on): 1. Fourteen-year-old natural vegetation with wild hays and flowers, shrubs and trees with no management (= natural). 2. Fourteen-year-old vegetation with grass species (mainly timothy and meadow fescue), harvested once per year with a light lawnmower and residues removed from the BZ (= harvested). 3. Three-year-old vegetation with grass species (mainly timothy and meadow fescue), grazed by cattle (72 and 234 cow-grazing days/ha/yr in 2003 and 2004, respectively). Before 2003, this area had been cultivated by conventional tillage (= grazed).

Sampling Two sets of undisturbed soil samples were taken in May 2005 before substantial vegetative growth at 0–5 cm and 5–10 cm depths. Six replicates of 100 cm3 cylinders were taken to determine air permeability and small-scale sorptivity. Other 250 cm3 cylinders were sampled with 10 replicates for bulk density, porosity and saturated hydraulic conductivity measurement. The samples were stored at +4 °C until measurement. The first field infiltration measurements were carried out in May 2005 with three replicates and the second measurements in November 2005 with two replicates.

Bulk density and soil porosity The dry bulk density was calculated by dividing the weight of the soil dried at 105 °C (n = 10) by the volume of the cylinder (250 cm3). The amounts of water removed at 10 and 1 kPa (sand bed) matric potential were used to determine pores larger than 30 and 300 lm (equivalent pore diameter), respectively.

Hydraulic measurements The samples were saturated in water for 2 weeks before measurement. Saturated hydraulic conductivity (Ksat) was

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418

412 K. Rasa & R. Horn determined with the method of permanent hydraulic head, using a laboratory permeameter (Eijkelkamp). The results were calculated, using equation (1): Ksat ¼ Vw  l=ðA  t  hÞ

ð1Þ

in which Vw is the volume of water per unit cross-sectional area of soil (A) per unit time (t), l is the length of the sample and h is the hydraulic head. Hydraulic conductivity at field saturation (Kfs) was measured, using a single-ring infiltrometer (cylinder Ø 30 cm, height 30 cm) connected with a Mariotte reservoir (Ø 20 cm, height 1.2 m). The cylinder was installed to a depth of 10 cm, and the ponding height inside the cylinder was adjusted to 10 cm by the Mariotte siphon. The loss of water from the Mariotte reservoir was manually recorded at 30-s interval. Kfs was calculated from the steady-state infiltration rate (vi) according to equation (2): vi ¼ V=ðA  tÞ ¼ Kfs ðH1 þ LÞ=L

ð2Þ

where V is the volume of infiltrated water, A the cylinder area, H1 the ponding height and L the assumed length of the percolated soil horizon (10 cm, corresponding to the depth of the cylinder bottom from the soil surface). Bouwer (1986) listed several possible sources of error and uncertainty concerning the approach described above, from which lateral divergence of flow below the cylinder and lack of information of the wetting front are probably the most relevant ones in this case. Thus, we consider the field infiltration results as comparable relative values between the sites. The small-scale water sorptivity, that is, capacity of the soil to absorp water by capillarity (Philip, 1957), was measured at various soil moisture contents that corresponded to those values for the Ka measurement (see below). Infiltration of water at 20 °C was measured, using the apparatus described by Leeds-Harrison et al. (1994) and modified by Hallett & Young (1999). Infiltration was detected within 75 s at 15-s interval. The sorptivity (S) of water was calculated according to equation (3) (Leeds-Harrison et al., 1994): S ¼ ðQ  f =ð4b  rÞÞ1=2

ð3Þ

where Q is the stationary infiltration rate of water flow, f the airfilled porosity and r the radius of the infiltration tip: 1.5 mm. The value of 0.55 was used for parameter b, and pressure head was set to 2 cm. As the device samples a small volume of soil, the results are considered to represent the sorptivity of the aggregates. For further details, see the study of Rasa et al. (2007), in which these data were presented in the form of an R-index, R = 1.95*(sorptivity of ethanol/sorptivity of water).

Air permeability and pore geometry index The samples were saturated in the laboratory with water and dried stepwise to the matric potential of 3 kPa, 6 kPa,

10 kPa, 15 kPa, 30 kPa and 50 kPa on a sand bed or on a ceramic plate. The Ka measurements were carried out after each drying step. The volume of air (Va) per time (t) was determined, using the apparatus described by Horn et al. (2004), and air conductivity (Kl) was calculated, using equation (4): Kl ¼ qa  g  ðVa  lÞ=ðt  Dp  AÞ

ð4Þ

where qa is air density at 20 °C, g the gravitation, Dp the air pressure applied during measurement (1 cm), l and A are the length (5 cm) and area of the sample (20 cm2), respectively. Ka was calculated from K1, using equation (5): Ka ¼ Kl  ðga =ðqa  gÞÞ

ð5Þ

where ga is the viscosity of air. The pore geometry index (O, equation 6. ‘Organization’ in Blackwell et al., 1990) was calculated, dividing Ka by the air-filled porosity at a matric potential of 3 kPa (Groenevelt et al., 1984; Ball et al., 1988; Blackwell et al., 1990). O ¼ Ka =

ð6Þ

Statistical analyses Statistical analyses were carried out using SAS statistical software (version 9.2). With separate analyses of variance (ANOVA), we analysed whether bulk density and macroporosity (>30 and >300 lm) varied between the depths (0–5 cm and 5–10 cm), sites or according to the joint effect of depth 9 site (interaction). In the study of Ka and sorptivity of the aggregates, the soil matric potential was also included as a dependent variable. For the analysis of Kfs, management was the only dependent variable. The assumptions for ANOVA were tested from the residuals with the Kolmogorov–Smirnow test for normality and Levene’s test for heteroscedasticity. To meet the normality assumptions, the data on Ka, aggregate sorptivity and Kfs data were log-transformed. The values reported were back-transformed. The pairwise comparisons of significant interactions were conducted with Tukey-Kramer tests. The P-values of the multiple comparisons shown in the text and figures were Bonferroni-corrected (Miller, 1981), because multiple comparisons increase the level of type I error (rejecting the H0-hypothesis without justification). Ksat was a response variable with substantial amounts of zeros (some samples had very small values below the sensitivity of the device, Ksat = 0), and tests based on normality were not valid. Therefore, we first analysed whether the occurrence of zero-conductivity was dependent on the depth or management by dividing the data into binary form: values more than 0 and values of 0. These data were analysed with generalized linear models (procedure GENMOD with binomial distribution), while also allowing for testing of interactions for binomial response. Second, we

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418

Physical properties of clayey surface soil

Bulk density and porosity Bulk density was greater at depth of 5–10 cm than at the surface horizon (Table 1), but bulk density was dependent on the management as well (Figure 1a). In the deeper soil horizon, bulk density did not differ between the managements, whereas in the topsoil, it was greater at the grazed site than at the natural and harvested sites. When the effects of management between the depths were compared, we found that the bulk density was smaller in the topsoil, both at the natural and harvested sites. However, the bulk density at the grazed site was similar at both depths, indicating that grazing mostly affected the soil surface. The total porosity corresponding to the bulk density varied Table 1 Summary of the ANOVA results for bulk density, total porosity, macropores >30 lm and large macropores >300 lm, air permeability, aggregate sorptivity and field-saturated hydraulic conductivity measures in spring

Natural Harvested Bulk density (g/cm)

Results

(a) 1.4

1.2

b

Grazed

a 1.0

0.8

0–5 cm (b) 14

5–10 cm

a

Natural Harvested

12

Grazed 10

ab bc

8

bc

Macropores >30 lm Large macropores >300 lm Air permeability

Aggregate sorptivity

Field-saturated hydraulic conductivity

Depth Management Depth*Management Depth Management Depth*Management Depth Management Depth*Management Depth Management Moisture Depth*Management Depth*Moisture Management*Moisture Depth Management Moisture Depth*Management Depth*Moisture Management*Moisture Management

ndf, ddf 1, 2, 2, 1, 2, 2, 1, 2, 2, 1, 2, 5, 2, 5, 10, 1, 2, 5, 2, 5, 10, 2,

54 54 54 54 54 54 53 53 53 35.2 35.2 150 34.8 150 151 29.6 29.6 149 29.6 149 149 6

F 23.8 7.66 9.67 37.93 5.86 8.79 7.81 9.38 11.01 9.08 11.23 2.37 0.3 0.49 0.9 68.47 0.88 1.81 7.08 1.11 1.65 21.11

P 30 lm) and large macropores (>300 lm) (Figure 1b,c, Table 1). In the deeper soil horizon, there were no differences in the abundance of these pores between management

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418

414 K. Rasa & R. Horn

Flow of water and air

10 8 6 4 2 0 Natural Harvested Grazed

Figure 2 Number of conductivity (Ksat).

80 Air permeability (µ m2)

Descriptive statistics for the Ksat are presented in Table 2, in which 22 of the 60 samples showed no measurable hydraulic conductivity (Ksat = 0) in the following defined as zero value. There was a trend (Figure 2, v22 = 5.7, P = 0.07) suggesting that the number of zero values varied among the management practices. There were more zero samples at the grazed site than at the harvested site (v21 = 4.8, P = 0.028, Figure 2). The effect of management was similar in both horizons. Although there were no differences in the number of zero values between the soil horizons, Ksat in the analyses with no zero values was generally smaller in the surface soil than in the deeper horizon (F1.32 = 13.6, P = 0.0008; mean and SE for surface 12.8 + 6.1, 4.1 and for the deeper horizon 85.5 + 34.5, 24.6, respectively). Ka was dependent only on the main factors: depth, management and moisture (Table 1). The deeper horizon had a smaller Ka value than the surface soil, (19.0 lm2 +4.3, 3.4 and 43.6 lm2 +9.2, 7.6, respectively). The grazed site had smaller Ka than the natural or harvested sites (Figure 3a). The main effects of statistical analysis showed again that Ka is dependent on moisture content, but after Bonferroni corrections of pairwise comparisons, these differences faded (data not shown). However, there was a trend indicating a drop in Ka between matric potentials of 3 and 6 kPa and that air permeability increases when the soil was dried (P-values harvested > grazed site (29–42 vol.%) (Rasa et al., 2007). For the functioning of the BZ, dense vegetation may be favourable, because it protects the soil surface against raindrop impact and slows down water flow, although removal of vegetation is recommended for reducing nutrient leaching from decaying biomass (Ra¨ty et al., 2010b; Uusi-Ka¨mppa¨ & Jauhiainen, 2010). The relatively high clay content and common wetting– drying cycles between spring and autumn favour development of aggregated soil structure (Horn et al., 1994; Rasa et al., 2009). Indeed, different degrees of soil aggregation were visually observed at the time of sampling. The results of Ka supported this finding. The large interaggregate macropores became empty at low suction (3 kPa). They dominated the airflow as then and further drying of soil increased Ka only slightly. In a homogenized soil without aggregated structure, more significant and continuous increase in Ka would have been expected (Aljibury & Evans, 1965). In our structured soil, shrinkage occurs preferentially in aggregates than in the entire soil (Rasa et al., 2009). Therefore, drying was accompanied by only slight increase in Ka. Aggregates may contribute to water flow, imbibing water in the early stages of rain events (Youngs et al., 1994). The intraaggregate pores are often smaller and more tortuous than in the bulk soil (Horn et al., 1994). The pore size distributions of our soils showed that 43–52% of the pores were smaller than 30 lm. It is reasonable to assume that the majority of these pores are within the aggregates, which emphasizes their role in water storage. However, aggregates in the surface horizons imbibed water relatively rapidly, as also shown previously (Youngs et al., 1994; Lipiec et al., 2006; Ferrero et al., 2007). Wetting–drying cycles, most frequent and intense in the topsoil, may have generated pore systems favourable for sorptivity (Czarnes et al., 2000). The sorptivity decreased significantly in the deeper horizon, indicating denser, less porous aggregates than at the soil

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418

416 K. Rasa & R. Horn surface (Lipiec et al., 2006). Decreased aggregate sorptivity is conducive to bypass flow in the deeper soil horizon. Applied external pressure caused by grazing alters the pore system and functionality, especially by the dominant shear stresses, which affect the pore system in many ways. If the soil is strong enough due to a very rigid pore system, no changes in the structural functions occur. If, however, the sensitivity of the soil is greater, for example, due to greater stresses applied or in combination with a somewhat greater water content, soil pores will be diminished in diameter and in total amount, resulting in a smaller pore volume and poorer hydraulic or gaseous fluxes. Intense moulding at a large water content gives rise to homogenization of soil structure and a smaller bulk density or larger total porosity may be observed, while the hydraulic conductivity or water infiltration is reduced. The hydraulic properties may improve again only if consecutive drying causes normal shrinkage and a new cycle of initial crack formation. In the present study, soil at the grazed site was subject to individual hoof pressure of ca. 125 kPa (Ra¨ty et al., 2010a). Consequently, significantly smaller bulk density in the surface horizon of the natural and harvested sites than at the grazed site indicates soil compaction at the latter site. In the deeper horizons, differences between the management practices faded, which is in agreement with our assumption that the adverse effects of this type of stress are limited to the thin surface soil. Similar trends were observed in the results of macroporosity (> 30 lm) and large macropores (> 300 lm) as well: (i) difference in the macropores between the natural and the harvested sites compared with the grazed site, (ii) distinct macropores between the horizons in the natural and harvested sites, (iii) more homogenous profile at the grazed site and (iv) minor difference between the sites in the deeper horizon. In general, our results are in the range documented for bulk density and porosity in boreal clayey soils (Aura, 1983; Pietola et al., 2005; Rasa et al., 2009; Regina & Alakukku, 2010). It is well known that soil compaction tends to reduce macroporosity in the first place (Aura, 1983; Pagliai et al., 2003). This was the case in our study as well, because grazing significantly reduced the large macropores in the soil surface. The decrease in the large pores adversely affect soil productivity (Aura, 1983), as well as hydraulic or gaseous fluxes, as discussed above. However, little attention has been focused on the pore characteristics, as stated by Alakukku (1998). In the present study, pore geometry index O was calculated to assess pore morphology. Grazing adversely affected soil structure by reducing the macropores, as well as by altering pore morphology. The greater pore geometry index at the natural and harvested sites indicated greater continuity of the pores than at the grazed site (Iversen et al., 2001). Dense versatile vegetation and its root system at the natural site are probably some of the reasons for the highest pore geometry index, as channels made by roots are well permeable to air (Blackwell et al.,

1990). The study of soil thin sections (Rasa et al., 2012), carried out at the same study site and time as the present study, provided qualitative description and quantitative image analysis data of the pore characteristics of the soils studied. In that study, horizontally oriented macropores (>50 lm) increased from the natural site to the harvested site and further to the grazed site, showing a platy oriented soil structure, which is in accordance with the pore geometry indices detected in the present study. Deteriorated pore geometry at the grazed site weakened the soil hydraulic properties measured, together with reduced porosity, as suggested earlier by D€ orner & Horn (2006). Therefore, pore geometry indices are effective measures of macropore characteristics, as well as their functionality. Many samples did not conduct water at all, despite their abundant macroporosity. This observation can be attributed to poor connectivity or high tortuosity of the pore system. Swelling especially tends to make pores narrower or to close the connections between pores when the pore walls are accommodated (Bullock & Murphy, 1980). The shrinkage properties of the soil studied (Rasa et al., 2009) and the macropore system characterized by partially accommodated, irregular, elongated pores (Rasa et al., 2012) suggest that wetting and swelling are likely to result in decrease in hydraulic conductivity. Zero connectivity at the grazed site can be explained by the shearing and compacting effects of hoof pressure. In contrast, there were samples showing very high conductivity, which indicates occurrence of preferential flow routes. The spatial variation in zones with low and high conductivity is conducive to macropore flow or it may give rise to surface run-off (Jarvis, 2007). Our results for Kfs were in agreement with previous statements. In spring, high volume infiltration in initially dry soil indicates the occurrence of preferential flow routes. The adverse effect of grazing on soil structure was seen in the smaller infiltration compared with the other sites. In autumn, differences between management practices were less evident, but Kfs decreased remarkably compared with in spring. The seasonal variation in Kfs of clayey soils was documented by previous studies as well (Messing & Jarvis, 1990; Bagarello & Sgroi, 2004). Our findings are in agreement with the field-scale studies conducted at the Lintupaju experimental field, where grazing increased the volume of surface run-off and the runoff generally coincided with the wet soil conditions (UusiKa¨mppa¨ & Jauhiainen, 2010). As discussed above, prolonged wetness in autumn allows the soil to swell, which causes a decrease in hydraulic conductivity and increases surface runoff. However, the results of Kfs measurements address the importance of seasonal variation in soil physical properties. Ka tended to decrease slightly when the samples were dried from a matric potential of 3 kPa to 6 kPa. This interesting detail could be related to the fact that the shrinkage study in the same soil (Rasa et al., 2009) revealed that all sites (except the natural site 0–5 cm) experienced structural collapse in the

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418

Physical properties of clayey surface soil

proportional shrinkage zone (starting an app. matric potential of 6 kPa). Structural collapse results in irreversible rearrangement of the internal structure of soil and may explain the observed trend in Ka. In that study, the weakness of the soil structure to increasing hydraulic stress was explained by prolonged soil wetness, freeze–thaw cycles and formation of ice lenses between autumn and spring, which generates unstable pore systems. Subsequent shrinkage in spring would be expected to increase the number of contact points of the particles forming aggregates, as well as the contact points between the aggregates themselves (Horn & Smucker, 2005). Consequently, soil strength would be expected to increase during spring and summer.

Conclusions The soils studied were characterized by different degree of aggregation, and it was evident that two rather distinct pore systems were responsible for water flow. Sorptivity of the aggregates and bulk soil physical properties were generally less conducive to infiltration in the deeper horizon, which may enhance lateral distribution flow in the horizon above. Fast preferential water flow occurred initially in dry soil, but prolonged wetness reduced infiltration, increasing the risk of surface run-off. The structural and hydraulic properties of the soil were least favourable for water infiltration in the grazed BZ. Cattle trampling especially deteriorates the shallow top layer of the soil by reducing the amount of large pores, as well as the pore continuity. Only minor differences between the physical properties of the natural and harvested sites were observed. In considering the structural and physical properties presented in this study, as well as the inherent property of harvesting of decreasing amount of nutrients in BZs, harvesting is recommended as the best management practice.

Acknowledgements We greatly acknowledge financial support from the Finnish Ministry of Agriculture and Forestry, Maj and Tor Nessling Foundation, Finnish Cultural Foundation and Maa- ja Vesitekniikan Tuki ry. Dr. Liisa Pietola is acknowledged for the initiation and management of the SUOTO project. We thank Professor Markku Yli-Halla and M.Sc. Mari Ra¨ty for their valuable comments on the manuscript and Dr. Outi Vesakoski for advice on the statistical analyses.

References Alakukku, L. 1998. Properties of compacted fine-textured soils as affected by crop rotation and reduced tillage. Soil & Tillage Research, 47, 83–89. Aljibury, F.K. & Evans, D.D. 1965. Water permeability of saturated soils as related to air permeability at different moisture tensions. Soil Science Society of America Proceedings, 29, 366–369.

417

Aura, E. 1983. Soil compaction by the tractor in spring and its effect on soil porosity. Journal of the Scientific Agricultural Society of Finland, 55, 91–107. Bagarello, V. & Sgroi, A. 2004. Using the single-ring infiltrometer method to detect temporal changes in surface soil field-saturated hydraulic conductivity. Soil & Tillage Research, 76, 13–24. Ball, B.C., O’Sullivan, M.F. & Hunter, R. 1988. Gas diffusion, fluid flow and derived pore continuity indices in relation to vehicle traffic and tillage. Journal of Soil Science, 39, 327–339. Blackwell, P.S., Ringrose-Voase, A.J., Jaywardane, N.S., Olsson, K.A., MCKenzie, D.C. & Mason, W.K. 1990. The use of air-filled porosity and intrinsic permeability to air to characterize structure of macropore space and saturated hydraulic conductivity of clay soils. Journal of Soil Science, 41, 215–228. Bouwer, H. 1986. Intake rate: Cylinder infiltrometer. In: Methods of soil analysis. Part 1. Physical and mineralogical methods (ed. A. Klute). pp. 825–844. American Society of Agronomy, Inc. Soil Science Society of America, Inc. Madison, Wisconsin, USA. Bullock, P. & Murphy, C.P. 1980. Towards the quantification of soil structure. Journal of Microscopy, 120, 317–328. Czarnes, S., Hallett, P.D., Bengough, A.G. & Young, I.M. 2000. Root- and microbial-derived mucilages affect soil structure and water transport. European Journal of Soil Science, 51, 435–443. D€ orner, J. & Horn, R. 2006. Anisotropy of pore functions in structured Stagnic Luvisols in the Weichselian moraine region in N Germany. Journal of Plant Nutrition and Soil Science, 169, 213–220. FAO, 2006. World reference base for soil resources 2006. A framework for international classification, correlation and communication. World Soil Resources Reports 103, Rome, Italy, p. 128. Ferrero, A., Lipiec, J., Turski, M. & Nosalewics, A. 2007. Stability and Sorptivity of soil aggregates in grassed and cultivated sloping vineyards. Polish Journal of Soil Science, 40, 1–8. Groenevelt, P.H., Kay, B.D. & Grant, C.D. 1984. Physical assessment of a soil with respect to rooting potential. Geoderma, 34, 101–114. Hallett, P.D. & Young, I.M. 1999. Changes to water repellence of soil aggregates caused by substrate-induced microbial activity. European Journal of Soil Science, 50, 35–40. Hoffmann, C.C., Kjaergaard, C., Uusi-Ka¨mppa¨, J., Hansen, H.C.B. & Kronvang, B. 2009. Phosphorus retention in riparian buffers: review of their efficiency. Journal of Environmental Quality, 38, 1942–1955. Horn, R. & Smucker, A. 2005. Structure formation and its consequences for gas and water transport in unsaturated arable and forest soils. Soil & Tillage Research, 82, 5–14. Horn, R., Taubner, H., Wuttke, M. & Baumgartl, T. 1994. Soil physical properties related to soil structure. Soil & Tillage Research, 30, 187–216. Horn, R., Vossbrink, J. & Becker, S. 2004. Modern forestry vehicles and their impacts on soil physical properties. Soil & Tillage Research, 79, 207–219. Iversen, B.V., Moldrup, P., Schjønning, P. & Loll, P. 2001. Air and water permeability in differently textured soils at two measurement scale. Soil Science, 166, 643–659. Jarvis, N.J. 2007. A review of non-equilibrium water flow and solute transport in soil macropores: principles, controlling factors and consequences for water quality. European Journal of Soil Science, 58, 523–546.

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418

418 K. Rasa & R. Horn Leeds-Harrison, P.B., Youngs, E.G. & Uddin, B. 1994. A device for determining the sorptivity of soil aggregates. European Journal of Soil Science, 45, 269–272. Lipiec, J., Ku , J., Nosalewicz, A. & Turski, M. 2006. Tillage system effects on stability and sorptivity of soil aggregates. International Agrophysics, 20, 189–193. Messing, I. & Jarvis, N.J. 1990. Seasonal variation in field-saturated hydraulic conductivity in two swelling clay soils in Sweden. Journal of Soil Science, 41, 229–237. Miller Jr, R.G. . 1981. Simultaneous Statistical Inference. SpringerVerlag, New York, pp.29, 9. Pagliai, M., Marsili, A., Servadio, P., Vignozzi, N. & Pellegrini, S. 2003. Changes in some physical properties of clay soil in Central Italy following the passage of rubber tracked and wheeled tractors of medium power. Soil &Tillage Research, 73, 119–129. Philip, J.R. 1957. The theory of infiltration: 4. Sorptivity and algebraic infiltration equations. Soil Science, 84, 257–264. Pietola, L., Horn, R. & Yli-Halla, M. 2005. Effects of trampling by cattle on the hydraulic and mechanical properties of soil. Soil& Tillage Research, 82, 99–108. Rasa, K., Horn, R., Ra¨ty, M., Yli-Halla, M. & Pietola, L. 2007. Water repellency of clay, sand and organic soils in Finland. Agricultural and Food Science, 16, 267–277. Rasa, K., Horn, R., Ra¨ty, M., Yli-Halla, M. & Pietola, L. 2009. Shrinkage properties of differently managed clay soils in Finland. Soil Use and Management, 25, 175–182. Rasa, K., Eickhorst, T., Tippk€ otter, R. & Yli-Halla, M. 2012. Structure and pore system in differently managed clayey surface soil as described by micromorphology and image analysis. Geoderma, 173-174, 10–18. doi:10.1016/j.geoderma.2011.12.017

Ra¨ty, M., Horn, R., Rasa, K., Yli-Halla, M. & Pietola, L. 2010a. Compressive behaviour of the soil in buffer zones under different management practices in Finland. Agricultural and Food Science, 19, 160–172. Ra¨ty, M., Uusi-Ka¨mppa¨, J., Yli-Halla, M., Rasa, K. & Pietola, L. 2010b. Phosphorus and nitrogen cycles in the vegetation of differently managed buffer zones. Nutrient Cycling in Agroecosystems, 86, 121–132. Regina, K. & Alakukku, L. 2010. Greenhouse gas fluxes in varying soils under conventional and no-tillage practices. Soil & Tillage Research, 109, 144–152. Soil Survey Staff 2010. Keys to Soil Taxonomy. 11th edn. United States Department of Agriculture, Natural Resources Conservation Service, p. 338. Syversen, N. 2005. Effect and design of buffer zones in the Nordic climate: the influence of width, amount of surface runoff, seasonal variation and vegetation type on retention efficiency for nutrient and particle runoff. Ecological Engineering, 24, 483–490. Uusi-Ka¨mppa¨, J. & Jauhiainen, L. 2010. Long-term monitoring of buffer zone efficiency under different cultivation techniques in boreal conditions. Agricultural, Ecosystems and Environment, 137, 78–85. Uusi-Ka¨mppa¨, J., Braskerud, B., Jansson, H., Syversen, N. & Uusitalo, R. 2000. Buffer zones and constructed wetlands as filters for agricultural phosphorus. Journal of Environmental Quality, 29, 151–158. Uusitalo, R., Turtola, E., Kauppila, T. & Lilja, T. 2001. Particulate phosphorus and sediment in surface runoff and drainflow from clayey soils. Journal of Environmental Quality, 30, 589–595. Youngs, E.G., Leeds-Harrison, P.B. & Garnett, R.S. 1994. Water uptake by aggregates. European Journal of Soil Science, 45, 127–134.

© 2013 The Authors. Journal compilation © 2013 British Society of Soil Science, Soil Use and Management, 29, 410–418