Climate and relief influence on particle size distribution and chemical properties of Pseudogley soils in Croatia

Catena 127 (2015) 340–348

Contents lists available at ScienceDirect

Catena journal homepage: www.elsevier.com/locate/catena

Climate and relief influence on particle size distribution and chemical properties of Pseudogley soils in Croatia Vedran Rubinić a,⁎, Boris Lazarević a, Stjepan Husnjak a, Goran Durn b a b

University of Zagreb, Faculty of Agriculture, Svetošimunska 25, HR-10000 Zagreb, Croatia University of Zagreb, Faculty of Mining, Geology and Petroleum Engineering, Pierottijeva 6, HR-10000 Zagreb, Croatia

a r t i c l e

i n f o

Article history: Received 23 April 2014 Received in revised form 9 December 2014 Accepted 15 December 2014 Available online 21 January 2015 Keywords: Stagnosols Loess parent materials Precipitation gradient Repeated measures statistical method

a b s t r a c t Pseudogley is a soil characterized by vertical texture contrast and periodic stagnation of precipitation water. Its profile is often designated as A-Eg-Btg-Cg. It is the second most frequent soil type in Croatia, found almost exclusively on non-calcareous loess sediments in the Pannonian region of the country. The aim of this research was to determine if differences in particle size distribution and basic chemical properties exist among Pseudogleys formed along the 600–1100 mm mean annual precipitation (MAP) gradient on two different relief positions (plateau and slope) across the Pannonian region of Croatia. A total of 33 soil pits were dug in natural forests. The trends observed with soil depth for particle size distribution, organic C content, pH, Ca2+/Mg2+ ratio, and base saturation point to the predominance of top-down formation of the investigated soil profiles. Both relief and climate influenced the distribution of Pseudogleys across the Pannonian region of Croatia. The incomplete homogeneity of loess parent materials across the study region governs the variations in clay content, silt content, and cation exchange capacity among the investigated Pseudogleys. Conversely, the increase in MAP along the investigated transect caused a decrease in soil pH, base saturation, and Ca2+/Mg2+ ratio, and the increase in organic C content along the investigated profiles. Therefore, future studies of climate impact on loess-derived soils in this region should take into account only soil chemical properties that are not directly dependent on particle size distribution. The relief position on which soil pits were situated had no effect on soil characteristics. Hence, it seems that Pseudogleys should not be systemized according to their position on plateau or on slope, as it is the case in some classification systems. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Pseudogleys are soils that largely correlate with Stagnosols (IUSS Working Group WRB, 2006). Due to periodic stagnation of precipitation water on/in the poorly permeable subsurface horizon, redoximorphic features (RMF) develop in such soils. RMF comprise redox depletions (Fe-depleted and/or clay-depleted soil matrix), redox concentrations (soft Fe–Mn masses and coatings; cemented Fe–Mn concretions and nodules), and reduced matrix (soil matrix with Fe2 +-containing minerals) (see Schoeneberger et al., 2002). Pseudogley is the second most widespread soil type in Croatia, almost exclusively found in its Pannonian region (Bogunović et al., 1998) (Fig. 1, see also Appendix A). As the climax soil in most of the Pannonian region of Croatia, Pseudogley is also the most widespread soil type in this region (Bogunović et al., 1998). It is largely found in environments that are favorable for agriculture (loamy loess sediments, moderate climate, and gentle relief) (Škorić, 1986). Thereby, 55% of

⁎ Corresponding author. Tel.: +385 12394028; fax: +385 12393963. E-mail addresses: [email protected] (V. Rubinić), [email protected] (B. Lazarević), [email protected] (S. Husnjak), [email protected] (G. Durn).

http://dx.doi.org/10.1016/j.catena.2014.12.024 0341-8162/© 2014 Elsevier B.V. All rights reserved.

Croatian Pseudogleys comprise agricultural land or agro-ecosystems (Husnjak et al., 2011). Although soils with vertical texture contrasts are very often polygenetic (e.g., Phillips, 2004), Pseudogleys in Croatia and the wider southwestern Pannonian Basin were traditionally considered to form either by the erosional–sedimentational pedogenesis (primary Pseudogleys) or by the normal top-down pedogenesis (secondary Pseudogleys) (e.g., Ćirić, 1984; Škorić, 1986). Janeković (1960) even suggested that the upper two (coarser-textured) horizons originate from the noncalcareous Holocene loess that was deposited over the finer-textured Pleistocene loess. Nevertheless, micromorphology, particle size distribution, bulk/clay mineralogy, and geochemical properties of three Pseudogleys studied by Rubinić et al. (2014) in Croatia (soil profiles 1, 13, and 25 in this study—see Appendix B) point to top-down pedogenesis from initially vertically homogeneous loess deposits. Rubinić et al. (2015), who studied morphology, particle size distribution, chemical properties, and modal compositions of heavy/light mineral associations of three Pseudogleys (soil profiles 7, 19, and 28 in this study—see Appendix B), confirm that most Croatian Pseudogleys formed primarily by topdown pedogenesis. Therefore, natural Pseudogley profiles in Croatia can be generally designated O-A-Eg-Btg-Cg.

V. Rubinić et al. / Catena 127 (2015) 340–348

341

Fig. 1. Map of Pseudogley distribution in the Pannonian region of Croatia (determined according to Bogunović et al., 1998). The 11 points on the map indicate the approximate positions of the investigated locations. At each location three replicate soil pits were investigated. WZ, CZ, and EZ are mean annual precipitation zones (western zone with 900–1100 mm, central zone with 800–900 mm, and eastern zone with 600–800 mm of precipitation, respectively; simplified according to Perčec Tadić, 2008). The red line on the west represents the 900 mm isohyet, and the red line on the east represents the 800 mm isohyet. In the upper right corner of the Figure, the map with the marked position of Croatia in Europe is presented (the red rectangle roughly marks the Pannonian region of Croatia).

Numerous researchers recognized various effects of relief (e.g., Alaoui et al., 2011; Daniels et al., 1971; Griffiths et al., 2009; Venkatesh et al., 2011; Yang et al., 2012) and climate (e.g., Alexandrovskiy, 2007; Alvarez and Lavado, 1998; Dahlgren et al., 1997; Wanhong and Yao, 2006) on both formation and characteristics of soils. The aim of our study was to determine if distinct differences in particle size distribution and basic chemical properties exist among 33 Pseudogleys found on two different relief positions (plateau and slope) along the 600–1100 mm mean annual precipitation (MAP) gradient in the Pannonian region of Croatia. Therefore, spatial distribution of Pseudogleys in respect to relief and MAP was determined across the study region, and representative Pseudogley profiles in climax forests and on loess parent materials were selected for the study. Given that the increase in water balance surplus along the MAP gradient in the Pannonian region of Croatia distinctly influences soil formation (Bašić, 2013), we wanted to test the effects of MAP on Pseudogleys in this region. Furthermore, because studies of climate effects on soils are increasingly used for predicting the biogeochemical responses to the global climate change (e.g., Egli et al., 2008; Griffiths et al., 2009; Wanhong and Yao, 2006), we wanted to highlight the data relevant for this issue. Additionally, we aimed to test if Pseudogleys in Croatia (and several other countries of the region) are rightfully systemized into lower soil units (e.g., subtypes) in respect to their position in the relief (plateau or slope) (e.g., Bašić, 2013; Resulović et al., 2008; Škorić et al., 1985). Namely, within the leading soil classification systems of the World (e.g., IUSS Working Group WRB, 2006), relief and other soil-forming factors are avoided as systematization criteria, and only clearly observable and/or measurable soil properties are considered.

2. Materials and methods 2.1. Study area—Pannonian region of Croatia 2.1.1. Basic geomorphic characteristics Pannonian region of Croatia (Fig. 1) covers 46% of the country (Bašić, 2013) and represents the southwestern edge of the Pannonian Basin. The dominant geomorphic units of this region are the Holocene terraces and the Pleistocene terraces (Bašić, 2013). The Holocene terraces make the lowest part of the study area (80–120 m asl) and comprise valleys formed by Sava, Drava, and Danube rivers (and their tributaries). The Pleistocene terraces are found roughly between 100 and 200 m asl. At the east of the region, the Pleistocene terraces are flat, spacious, and elevated several meters above the Holocene terraces. Towards the west, they mostly cover the bases of hills/mountains made from the Tertiary deposits (Bašić, 2013). Consequently, from the east to the west of the Pannonian region of Croatia, the Pleistocene terraces get progressively ragged and interspersed by drainage ditches and streams. The Pleistocene terraces in the Pannonian region of Croatia largely comprise loess derivates, with brown loess and typical loess found only in the most eastern part (see Haase et al., 2007). Due to their modification by syngenetic and/or postgenetic processes, both loess derivates and brown loess are considered as non-calcareous polygenetic sediments with increased clay content compared to the typical loess (Haase et al., 2007). Brown loess developed after the deposition of aeolian material in a humid environment, which is why it is less affected by pedogenesis compared with loess derivates (Haase et al., 2007). The increased clay content in brown loess deposited in eastern Croatia is explained by Mutić (1990). Namely, the uplift of the Đakovo loess-plateau in Croatia occurred only after the beginning of the Holocene. Accordingly, this plateau was submerged under the shallow

342

V. Rubinić et al. / Catena 127 (2015) 340–348

standing water during the late Pleistocene. Consequently, the aeolian material increased its clay content geogenetically, after its deposition in the aqueous environment. 2.1.2. Pseudogley parent materials According to Rubinić et al. (2014), parent materials of Pseudogleys in the Pannonian region of Croatia can be characterized as follows. They comprise largely loess derivates, and sporadically brown loess. Vertical trends along soil depth for several weathering indices (La/Ce, Sm/Nd, Ti/Al, and CIA (Chemical Index of Alteration, proposed by Nesbitt and Young, 1982)) indicated that these parent materials were initially vertically homogeneous. Mineralogical composition of non-clay minerals across the investigated parent materials was rather uniform (mainly quartz, K-feldspar, and plagioclase). However, distinct differences in contents of clay-sized minerals (especially of smectite, chlorite, vermiculite, and chlorite-vermiculite mixed layer minerals) were determined. Moreover, Rubinić et al. (2015) stress modal mineralogy of the investigated Pseudogley parent materials indicates more than one source of loess in the Pannonian region of Croatia. This, along with the uplifts of different loess plateaus in different points of time during the Pleistocene and the Holocene (see Mutić, 1990 and Section 2.1.1), resulted in slightly different particle size distributions of Pseudogley parent materials across the region. Nevertheless, we consider these parent materials as uniform enough to allow research of climate influence on Pseudogleys across the Pannonian region of Croatia. According to the substrate classification of Pietsch and Lucke (2008), parent materials of Croatian Pseudogleys can be labeled as Pleistocene (qp) white-yellowish (w) cover loams (L). Although these parent materials are of aeolian genesis (e), post-depositional superficial denudative processes (df) were possible on slopes. Nevertheless, just as no coarse (N2 mm) fragments (xs0) exist in these substrates on plains (P), none are found on slopes (Sl). Therefore, Pseudogley parent materials should largely be designated as qp-(e)P-Lw, and sporadically (on slopes) as qp(df)Sl-Lw-xs0. 2.1.3. Age of the youngest loess deposits (time of soil formation) Infrared stimulated luminescence (IRSL) analysis of the “Gorjanović loess section” (eastern part of the Pannonian region of Croatia) gave calculated ages that can be correlated to penultimate glacial–last interglacial–last glacial (OIS6–OIS2) (Wacha and Frechen, 2011). Given that the Pannonian region of Croatia was not affected by permafrost during the last glacial (see Renssen and Vandenberghe, 2003), the age of the Pleistocene terraces in Croatia may be estimated according to FAO (2006) to lPf (late Pleistocene, no periglacial impact) or lPp (late Pleistocene, periglacial, commonly recent soil formation on preweathered materials). Accordingly, most Croatian soils developed after the last glacial (Bašić, 2013). IRSL measurements on the stratigraphically youngest loess from the “Zmajevac loess section” (eastern part of the Pannonian region of Croatia) gave age estimates from 16.7 ± 1.8 to 20.2 ± 2.1 ka, indicating increased aeolian deposition during last pleniglacial/late glacial (Galović et al., 2009). Hence, we assume that intense formation of Pseudogleys on loess derivates in the Pannonian region of Croatia began with the end of the Pleistocene. However, because pedogenesis on the Đakovo loess-plateau began only after the beginning of the Holocene (see Section 2.1.1), Pseudogleys on brown loess should be considered younger. Nevertheless, we do not consider these variations in time of soil formation to significantly affect Pseudogley characteristics. Namely, changes of loess-derived soils usually become negligibly small 2–3 ka after the beginning of soil formation, when soils reach their climax stage (see Alexandrovskiy, 2007). 2.1.4. Climate and vegetation Climate in the Pannonian region of Croatia is moderate continental, largely humid (semihumid to semiarid only in the most eastern part of the region). Mountain areas aside, mean annual air temperature in

the region is mostly around 11 °C (with mean winter values around −1 °C and mean summer values around 21 °C). Excluding the mountain areas, MAP decreases from the west (1100 mm) to the east (600 mm) of the region (Perčec Tadić, 2008). These different amounts of MAP are due to different amounts of mean precipitation during spring and autumn. Namely, in these two seasons the east of the region receives averagely less precipitation than the west. On average, precipitation is the lowest in late winter and the highest in late spring and early summer. Evapotranspiration is lower than the potential evapotranspiration only during summer. Forest community of sessile oak and hornbeam (Epimediocarpinetum betuli) is the prevailing natural vegetation cover of the Pleistocene terraces in the Pannonian region of Croatia (Bašić, 2013). It is also the climax vegetation on Pseudogley soils in the region (Škorić, 1986). 2.2. Investigated locations Total of 11 representative locations were selected across the Pleistocene terraces in the study region (Fig. 1 and Appendix A). To study soil characteristics in accordance with the state factor model (Jenny, 1994), all locations featured uniform vegetation cover (forest of sessile oak and hornbeam). Because the forests at all investigated locations were well developed, negligible human impact on soils was presumed. However, given that most locations were close to agricultural land and/or human settlements (see Appendix A), human influence in the past cannot be excluded. The locations on slopes were uniform in terms of inclination (5%) and soil profile position (middle slope, 50 m from slope summit). Each location also featured a level microrelief. To assess the influence of climate on soil characteristics, the study area was divided into three zones with different amounts of MAP. These zones were labeled as the EZ MAP zone (eastern zone, 600–800 mm MAP), the CZ MAP zone (central zone, 800–900 mm MAP), and the WZ MAP zone (western zone, 900–1100 mm MAP) (Fig. 1). Precise data on precipitation (as well as on air temperature and evapotranspiration) are not provided because many investigated locations are remote from the nearest weather stations. To evaluate the influence of relief on soil characteristics, pairs of nearby locations with one location within the pair on plateau and the other on slope were selected across the study region. Thereby, the WZ MAP zone and the CZ MAP zone featured four locations each (two on plateaus and two on slopes). Given that in the EZ MAP zone only one representative location with Pseudogley on slope was found, that particular zone featured only three locations (see Appendix A). 2.3. Soil description and soil sampling Soil description and soil sampling took place from the winter 2010 to the autumn 2012. At each investigated location, three replicate soil pits (one central pit and two lateral pits) were dug within a circle of about 50 m radius. Therefore, the total of 33 soil pits was investigated across the study region (see Appendix B). Each soil pit was dug to the depth of about 1 m (Fig. 2). Soils were described according to FAO (2006) and/or Schoeneberger et al. (2002) (Table 1). Disturbed soil samples were collected from the upper four mineral horizons (A, Eg, Btg, and Cg horizons) of each soil profile and put in plastic bags. 2.4. Laboratory methods All soil samples were air-dried. A portion of each sample was crushed and sieved through a 2 mm sieve. Soil particle size distribution was determined by pipette-method, with wet sieving and sedimentation after dispersion with sodium-pyrophosphate (Na4P2O7, c = 0.4 M). Soil pH in H2O and in KCl (c = 1 M) was determined in 1:2.5 suspensions. Soil organic carbon (SOC) content was obtained by acid-

V. Rubinić et al. / Catena 127 (2015) 340–348

343

content, and CEC were analyzed using repeated measures in Proc Mixed (Mixed Model Repeated Measures). Repeated measures is a convenient method for analyzing data obtained repeatedly over time and space (Littell et al., 1996), including those from different soil depths at the same location (Gaeghan, 2012). In the statistical model, soil pits were treated as subjects nested within fixed effects (MAP zones and relief positions), while soil horizons (A, Eg, Btg, and Cg horizons) were treated as the repeated measurements. The covariance was modeled as unstructured according to residual log likelihood and Akaike's information criterion. Multiple comparison procedure was performed and pairwise differences for fixed effects and interactions were computed using Tukey–Kramer test. Given that the ANOVA pointed to significant interaction of soil horizons and MAP zones (SH*MAP) on all analyzed soil properties except on BS (Table 2), the results for the effect of this interaction were further analyzed and presented by graphs showing mean values with double standard error of the mean (SEM) (Figs. 3 and 4). Additionally, we calculated the ratio of clay contents (RCC) for each soil profile (RCC = clay content in the Btg horizon/clay content in the Eg horizon). We analyzed this ratio using ANOVA (analysis of variance) in Proc Mixed. The results are presented by the ANOVA table with compared mean values for the significant effect of MAP zones (Table 3).

3. Results and discussion 3.1. Morphology of the investigated Pseudogley profiles and designations of soil horizons Soil morphology of profiles 1, 3, and 25 (see Appendix B) is described in detail in Rubinić et al. (2014). Additionally, morphology of soil profiles 7, 19, and 28 (see Appendix B) is described in Rubinić et al. (2015). Hence, the following text describes only basic morphological characteristics that prevail across the 33 Pseudogleys investigated in this study. The Oi-A-Eg-Btg-Cg succession of soil horizons is generally typical for the investigated soil profiles (Fig. 2 and Table 1). The Oi horizon largely comprises litter of slightly decomposed remains of leaves and twigs from forest trees. The A horizon is very rich in soil organic matter (SOM) (Fig. 2). The predominance of horizontally-growing roots in the A horizon (Fig. 2) results from frequent stagnation of water in the Eg horizon. The Eg horizon is olive-brown (moist) and features few to common redox concentrations (mostly rounded Fe–Mn masses). Due to both the loss of clay (and sesquioxides) and the low content of SOM in the Eg horizon, this soil horizon is weakly structured. The Btg horizon features increased clay content and different soil structure in

Fig. 2. Example of a typical Pseudogley profile in Croatia (photograph of the soil profile P22 at the Location No. 8).

dichromate digestion. Additionally, soil cation exchange capacity (CEC), contents of exchangeable cations (only the ratio of exchangeable Ca and Mg ions (Ca2+/Mg2+ ratio) is shown), and base saturation (BS) were determined for 11 out of 33 soil profiles (central soil pits on each investigated location, see Appendix B). Cation exchange and re-exchange processes were induced with BaCl2 (c = 0.1 M) and MgSO4 (c = 0.02 M) solutions, respectively. 2.5. Statistical analysis Data were analyzed using the SAS® 9.2 statistical package (SAS Institutes, Cary, NC). Results for soil particle size distribution, pH, SOC Table 1 Typical morphology of Pseudogleys in Croatia. Horizon designation (FAO, 2006)

Horizon lower boundary a

Soil color—dry (moist) (Munsell Color, 2000)

Texture b

Structure c primary (secondary)

Aggregate size d primary (secondary)

RMF species e

RMF quantity f

Roots abundance g

Pores N2 mm abundance g

Oi A Eg Btg

– A C–G G

– Si-SiL Si-SiL SiL-SiCL

– GR GR BL

– VF VF FI–ME

– – RC RC, RD

– N F–C C–M

– C–M F–C V

– M C F–C

Cg

–

– 10YR 2/2–3/1 (2/1) 2.5Y 6/4 (4/4) 2.5Y 7/3–7/4 (5/3) 40–80% 2.5Y 7/1 (6/1–6/2) 5–15% 10YR 6/8 (5/8) 15–40% 2.5Y 7/3–7/4 (5/3) 15–40% 2.5Y 7/1 (6/1–6/2) 15–40 % 10YR 6/8 (5/8) 15–40%

SiL-SiCL

MA-PR (BL)

– (CO–VC)

RC, RD

C–M

V–N

V–N

a b c d e f g

A—abrupt, C—clear, G—gradual (FAO, 2006). Si—silt, SiL—silt loam, SiCL—silty clay loam (FAO, 2006). GR—granular, BL—blocky, MA—massive, PR—prismatic (FAO, 2006). VF—very fine, FI—fine, ME—medium, CO—coarse, VC—very coarse (FAO, 2006). RMF = redoximorphic features; RC—redox concentrations, RD—redox depletions (Schoeneberger et al., 2002). RMF = redoximorphic features; F—few, C—common, M—many (Schoeneberger et al., 2002). M—many, C—common, F—few, V—very few, N—none (FAO, 2006).

344

V. Rubinić et al. / Catena 127 (2015) 340–348

Table 2 ANOVA for the analyzed characteristics of the 33 investigated Pseudogley profiles in Croatia. Effect

SHa RPb MAPc SH × RP SH × MAP RP × MAP SH × RP × MAP

Pr N F Num DF

Sand content (2–0.063 mm)

Silt content (0.063–0.002 mm)

Coarse/fine silt (0.02 mm limit)

Clay content (b0.002 mm)

Silt/clay

pH KCl

pH H2O

SOCd

CECe

BSf

Ca/Mgg

3 1 2 3 6 2 6

b.0001 0.8024 0.0002 0.0013 0.0399 0.6007 0.0005

b.0001 0.3449 b.0001 0.4149 b.0001 0.1701 0.2688

0.0009 0.0118 b.0001 0.5169 0.0490 0.0631 0.0604

b.0001 0.2824 b.0001 0.2338 b0.0001 0.2631 0.1082

b.0001 0.3849 b.0001 0.5211 0.0006 0.5819 0.4975

b.0001 0.1213 b.0001 0.3451 b.0001 0.0001 b.0001

b.0001 0.6399 b.0001 0.3800 0.0137 0.0091 0.0026

b.0001 0.2406 0.0029 0.6582 0.0005 0.2059 0.2560

0.0042 0.8225 0.4293 0.4583 0.0068 0.7955 0.0535

0.0048 0.9974 0.2412 0.2546 0.7706 0.9863 0.3374

0.0037 0.5395 0.0010 0.3655 0.0414 0.4007 0.1153

The bold values are statistically significant. a Soil horizon. b Mean annual precipitation zone. c Relief position. d Soil organic carbon content. e Cation exchange capacity. f Base saturation. g Ca2+/Mg2+ ratio of exchangeable cations.

comparison with the Eg horizon (Table 1). More notably, the Btg horizon is dominated by RMF (Fig. 2). Given that parent materials of Croatian Pseudogleys are usually thoroughly altered by pedogenesis, they are morphologically similar to Btg horizons (see Rubinić et al., 2014, 2015). Most notably, both Btg horizons and parent materials are usually dominated by RMF (Table 1 and Fig. 2). In both, most redox concentrations (largely Fe–Mn masses/coatings) are within soil aggregates and most redox depletions are on aggregate faces and along root channels (Fig. 2). Such distribution of RMF agrees with preferential flows of stagnant soil water. Given that in the investigated Pseudogleys most redox concentrations are soft and lack sharp boundaries, they can be largely considered contemporary (not relict). Further, the parent materials of Croatian Pseudogleys usually have the same or even slightly increased clay content compared with the Btg horizons above them (see Bašić, 2013; Rubinić et al., 2014, 2015). Given the clay coatings observed in thin sections and the depthrelated trends of the weathering indices (Ti/Al and CIA), Rubinić et al. (2014) inferred that the vertical increase in clay content from the topsoil towards the parent material in Croatian Pseudogleys is largely due to lessivage. High contents of clay (and RMF) in C horizons can be considered the result of the downward percolation of precipitation water along the cracks (zones of the redox depletions) in the parent material (see Bockheim, 1980; Shaw et al., 2004). Out of the 33 soil profiles in our study, 18 of them feature a lowermost horizon with N 1 and b1.2 times more clay than the overlying Btg horizon (see Appendix B). Hence, these lowermost horizons are designated as BCtg2 in the Appendix B. Further, five soil profiles feature a lowermost horizon with ≥1.2 times more clay than the overlying Btg horizon and are designated as Btg2 (Appendix B) and considered as argic horizons sensu IUSS Working Group WRB (2006). The lowermost horizon of one soil profile, which has as much as 1.82 times more clay than the overlying Btg horizon, represents an older sedimentary layer (2Cg horizon) found close to the soil surface due to hillslope processes (see Appendix B). The lowermost horizon in all remaining soil profiles has less clay than the overlying Btg horizon and is thereby designated as Cg (Appendix B). In order to avoid using different designations for the lowermost horizon of different soil profiles, throughout the paper we refer to this horizon uniformly as Cg (for simplicity).

3.2. Pseudogley distribution in respect to MAP zones and relief positions Based on the soil map of Croatia at a scale 1:300,000 (Bogunović et al., 1998), the total of 5406 km2 of Pseudogley soils was determined across the study area. Out of the total area covered by Pseudogleys,

49% is found in the CZ MAP zone, 36% in the WZ MAP zone, and 15% in the EZ MAP zone (see Fig. 1). Distinctly less Pseudogleys in the EZ MAP zone than in the other two MAP zones is due to geomorphic and climatic characteristics of the study region. Namely, most of the EZ MAP zone is lowland covered with alluvial materials of the Holocene terraces (on which Pseudogleys in Croatia generally do not form). In addition, due to semiaridsemihumid climate conditions in the eastern Croatia, the Pleistocene terraces in the EZ MAP zone are largely calcareous (typical loess) and do not comprise evolutionally developed soils, such as Pseudogley. It was due to the rareness of Pseudogleys in the EZ MAP zone that only one representative location with Pseudogley on slope was found in this particular zone (Fig. 1 and Appendix A). The fact that, despite the maximum MAP in the WZ MAP zone, Pseudogleys are slightly less abundant in the WZ MAP zone than in the CZ MAP zone is explained with the most prominent relief dynamics in the WZ MAP zone (see Section 2.1.1) and the resulting poor preservation of loess sediments at sites of their deposition. Accordingly, in each MAP zone except in the WZ MAP zone, Pseudogleys on slopes were less frequent than Pseudogleys on plateaus (see Fig. 1). Specifically, among all Pseudogleys in the region, 56%, 46%, and 37% of them developed on slopes in the WZ, CZ, and EZ MAP zones, respectively. Throughout the entire research area, 52% of all Pseudogleys were mapped on plateaus, and 48% on slopes. 3.3. Particle size distribution of the investigated Pseudogley profiles Significant effect for the SH*MAP interaction was determined for the contents of all soil fractions (Table 2). Given that the contents of sand in the analyzed soil profiles are negligible (Fig. 3a), only the significant differences concerning silt and clay are further discussed. Due to loess parent material, the silt fraction dominated in all analyzed horizons (Fig. 3b). The content of silt generally decreased with soil depth, ranging from 68.7% to 81.2% (WZ MAP zone), 71.9% to 83.5% (CZ MAP zone), and 63.7% to 82.4% (EZ MAP zone) (Fig. 3b). The silt content was not the lowest in the lowermost soil horizon only in the CZ MAP zone (Fig. 3b). The ratio of contents of coarse silt and fine silt (coarse/fine silt ratio, 0.02 mm size limit) ranged from 0.8 to 1.1 (WZ MAP zone), 1.4 to 1.5 (CZ MAP zone), and 1.1 to 1.2 (EZ MAP zone) (Fig. 3c). This ratio was higher in the A horizon than in other soil horizons in the WZ MAP zone, but was otherwise constant with soil depth (Fig. 3c). Thus, topdown pedogenesis in situ was presumed. However, the differences in coarse/fine silt ratio that were noted across the MAP zones (Fig. 3c) may indicate slight heterogeneity of loess parent materials across the investigated transect. The fact that no progressive increase of the

V. Rubinić et al. / Catena 127 (2015) 340–348

345

coarse/fine silt ratio was observed across the study region neither in the westerly nor in the easterly direction (Fig. 3c) was interpreted by Rubinić et al. (2015) as an indication of loess originating from multiple sources and overlapping in the central Pannonian region of Croatia. In agreement with the vertical trends for silt content, the clay content generally increased with soil depth (Fig. 3d). Namely, the content of clay increased from 13.0% to 26.6% (WZ MAP zone), 9.6% to 23.3% (CZ MAP zone), and 11.9% to 33.3% (EZ MAP zone) (Fig. 3d). Although Btg and Cg horizons of Croatian Pseudogleys often comprise similar clay contents (see Bašić, 2013), in our study this was the case only in the CZ MAP zone (Fig. 3d). The progressive increase in clay content with soil depth below the Btg horizon was discussed in the Section 3.1. The amount of precipitation is generally positively correlated with the rate of weathering of primary minerals, i.e., with the rate of clay formation (e.g., Alexandrovskiy, 2007; Alvarez and Lavado, 1998; Dahlgren et al., 1997; Jenny, 1994). Nevertheless, in the case of Croatian Pseudogleys, the influence of parent material clearly outweighs that of the MAP (see Section 2.1 for the details on parent materials and MAP gradient). Namely, in line with their silt contents (Fig. 3b), most soil horizons in the EZ MAP zone (as the zone with the lowest MAP) feature higher clay contents than the analogue soil horizons in the remaining two MAP zones (Fig. 3d). This is dominantly due to the paleo-environment in the EZ MAP zone (see Section 2.1.1). The ratio of contents of total silt and clay (silt/clay ratio, 0.002 mm size limit) varied from 2.6 to 6.3 (WZ MAP zone), 3.1 to 8.9 (CZ MAP zone), and 1.9 to 7.1 (EZ MAP zone) (Fig. 3e). Differences were noted among soil horizons in each MAP zone, except between Btg and Cg horizons in both the EZ MAP zone and the CZ MAP zone (Fig. 3e). Specifically, in accordance with the contents of silt (Fig. 3b) and clay (Fig. 3d), the silt/ clay ratio decreased with soil depth (Fig. 3e). The fact that in each MAP zone this ratio decreased with soil depth rather steadily and within rather narrow ranges (Fig. 3e) indicates top-down pedogenesis from an initially texturally uniform material (Khan et al., 2012). Therefore, although some Croatian Stagnosols may be polygenetic (Rubinić et al., 2015), most seem to be formed primarily by normal pedogenesis. The similar vertical trends of the silt/clay ratio across the three MAP zones (Fig. 3e) point to similar conditions and rates of weathering along the investigated transect (see Constantini et al., 2002; Ray, 1963). To obtain more information on vertical migration of clay, RCC ratio was calculated. In all cases it amounted to values above 1.20 (Table 3), thereby satisfying the criterion for the argic horizon (IUSS Working Group WRB, 2006) and confirming lessivage as one of key soil-forming processes. The ANOVA showed significant effect of MAP zones on RCC ratio (Table 2). Namely, RCC ratio was the highest in the CZ MAP zone (1.55) and the lowest in the WZ MAP zone (1.35) (Table 3). The reason this ratio was not the lowest in the zone with the lowest MAP and the highest in the zone with the highest MAP may be the possibly different clay mineralogy of soil profiles across the investigated transect. Namely, Rubinić et al. (2014) determined less fine clay minerals (chlorite–vermiculite mixed-layer minerals, vermiculites, and smectites) in a Pseudogley profile within the 900–1000 mm MAP zone, than in Pseudogley profiles within the zones with lower MAP. Moreover, subsoils of the profiles in the EZ MAP zone are enriched in clay not only due to lessivage, but also due to aeolian deposition in an aqueous paleo-environment (see Rubinić et al., 2014, 2015; see also Sections 2.1.1 and 2.1.2). 3.4. Chemical properties of the investigated Pseudogley profiles In line with the top-down soil formation, pHH2O generally increased with soil depth in each MAP zone (Fig. 4a). Only between A and Eg Fig. 3. Particle size distribution of the 33 investigated Pseudogley profiles in Croatia. Histograms represent mean values. Error bars associated with the histograms represent double standard error of the mean. The two vertical bars on each graph show standard error of the difference for soil horizons (1) and mean annual precipitation zones (2). WZ, CZ, and EZ are mean annual precipitation zones (western zone with 900–1100 mm, central zone with 800–900 mm, and eastern zone with 600–800 mm of precipitation, respectively).

346

V. Rubinić et al. / Catena 127 (2015) 340–348

Fig. 4. Chemical properties of the 33 investigated Pseudogley profiles in Croatia. Histograms represent mean values. Error bars associated with the histograms represent double standard error of the mean. The two vertical bars on each graph show standard error of the difference for soil horizons (1) and mean annual precipitation zones (2). WZ, CZ, and EZ are mean annual precipitation zones (western zone with 900–1100 mm, central zone with 800–900 mm, and eastern zone with 600–800 mm of precipitation, respectively).

horizons in the EZ MAP zone and the CZ MAP zone no differences in pHH2O were noted (Fig. 4a). Namely, pHH2O ranged from 4.2 to 5.4 (WZ MAP zone), 4.7 to 5.5 (CZ MAP zone), and 5.1 to 5.9 (EZ MAP Table 3 ANOVA with means for the ratio of clay contents between Btg and Eg horizons of the 33 Pseudogley profiles investigated in Croatia. Effect RP

1

MAP

2

RP*MAP

1 2 3 4 5 6 7

Num DF

Pr N F

Effect level

Mean6

SEM7

1

0.2047

2

0.0335

2

0.1708

Slope Plateau CZ3 EZ4 WZ5 CZ3 × slope CZ3 × plateau EZ4 × slope EZ4 × plateau WZ5 × slope WZ5 × plateau

1.42 1.50 1.55 a 1.48 ab 1.35 b 1.53 1.57 1.36 1.61 1.38 1.33

0.05 0.04 0.05 0.06 0.05 0.07 0.07 0.10 0.07 0.07 0.07

Relief position. Mean annual precipitation (MAP) zone. Central MAP zone (800–900 mm). Eastern MAP zone (600–800 mm). Western MAP zone (900–1100 mm). Means with the same letters are not significantly different from each other (P N 0.05). Standard error of the mean.

zone) (Fig. 4a). MAP affects weathering rates, and leaching of basic cations is more pronounced in humid than in arid conditions (see Alexandrovskiy, 2007; Dahlgren et al., 1997). Accordingly, soil horizons in the EZ MAP zone generally had the highest pHH2O, whereas soil horizons in the WZ MAP zone generally had the lowest pHH2O (Fig. 4a). The averagely highest pHH2O in the EZ MAP zone agrees with the highest RCC ratio previously established in that MAP zone (Table 3). Namely, pHH2O values lower than about 5.0 inhibit lessivage (e.g., Sauer et al., 2009). Along soil depth, pHKCl varied slightly from 3.4 (A horizon) to 3.6 (Cg horizon) in the WZ MAP zone and from 3.6 (Eg horizon) to 3.8 (Cg horizon) in the CZ MAP zone (Fig. 4b). However, because of the least pronounced acidification in the zone with the lowest MAP, in the EZ MAP zone pHKCl increased with depth from 3.6 (Eg horizon) up to 4.2 (Cg horizon) (Fig. 4b). Accordingly, A and Cg horizons in the EZ MAP zone had higher pHKCl values than the analogue soil horizons in the two remaining MAP zones (Fig. 4b). Although relief position generally had no effect on the investigated soil properties (Table 2), due to the significant interaction of relief positions and MAP zones for soil pH (Table 2), we analyzed the influence of relief on pH (data not presented graphically). We found that in the CZ MAP zone pH was higher on plateaus than on slopes. Namely, on plateaus pHH2O was 5.2 and pHKCl was 3.8, whereas on slopes pHH2O was 4.9 and pHKCl was 3.5 (SEM = 0.1 for all four means). No relief-

V. Rubinić et al. / Catena 127 (2015) 340–348

dependent differences in pH were determined in the two remaining MAP zones. Possible reasons why relief did not affect soil characteristics to a larger extent could be due to soil moisture content which usually increases from the top to the bottom of a hillslope (e.g., Venkatesh et al., 2011; Yang et al., 2012), whereas all soil pits in this study were dug at the middle slope position. Moreover, topographic factors often affect only shallow soil layers and even then only to a certain extent (Yang et al., 2012). Besides, water infiltrability and water storage capacity are usually higher in forest soils than in grassland (and arable) soils, which is why surface runoff is comparably reduced in forest soils (see Alaoui et al., 2011; Bašić, 2013). Soil BS varied from 43.9% to 60.9% (WZ MAP zone), 44.0% to 71.6% (CZ MAP zone), and 74.9% to 91.7% (EZ MAP zone). In each MAP zone, BS was the lowest in the Eg horizon and the highest in the Cg horizon, but no statistically significant differences were established (Table 2 and Fig. 4f). No significant differences among soil horizons were due to the high SEM values for BS (Fig. 4f), which resulted from the limited number of soil profiles in which BS, CEC, and Ca2+/Mg2+ ratio were analyzed (see Section 2.4). Nevertheless, the fact that BS is generally higher in the A horizon than in the Eg horizon agrees with the high content of SOC in the A horizon (Fig. 4c) (see Curtin et al., 1998). Along the study region, BS notably increased from the WZ MAP zone to the EZ MAP zone (Fig. 4f). Although the differences in BS among the MAP zones were not significant (Table 2 and Fig. 4f), they are in accordance with the MAP gradient. Soil CEC ranged (in cmol + kg−1 unit) from 6.2 to 14.9 (WZ MAP zone), 5.6 to 16.1 (CZ MAP zone), and 9.7 to 19.5 (EZ MAP zone) (Fig. 4d). CEC was the lowest in the Eg horizon and the highest in the Cg horizon (Fig. 4d). However, the difference between these two horizons was not significant in the EZ MAP zone (Fig. 4d). This was mostly due to only nine (instead of 12) soil profiles in the EZ MAP zone (see Section 2.2). Namely, because of less soil profiles, SEM values were higher in the EZ MAP zone than in the remaining two zones (Fig. 4d). Vertical trends for CEC along soil depth agree with positive correlations between CEC on one side and contents of clay and SOM on the other (e.g., Dahlgren et al., 1997; Wright and Foss, 1972). They also agree with the typically opposite trends of the contents of clay and SOM along soil depth in humid conditions (Griffiths et al., 2009). High CEC in A horizons (Fig. 4d) is explained with SOM often having more pronounced impact on CEC in respect to clay minerals (e.g., Wright and Foss, 1972). Although differences among the MAP zones were not statistically significant, the EZ MAP zone featured the averagely highest CEC (Fig. 4d). Such result corresponds to the highest clay content determined in the EZ MAP zone (Fig. 3d). Moreover, Dahlgren et al. (1997) found CEC to be influenced by clay mineralogy rather than by the gradients of MAP and mean annual temperature. Accordingly, Rubinić et al. (2014) determined clay content and clay mineralogy often have the key impact on CEC of Croatian Pseudogleys. The Ca2+/Mg2+ ratio varied between 0.6 and 1.4 (WZ MAP zone), 1.4 and 3.6 (CZ MAP zone), and 1.8 and 3.7 (EZ MAP zone) (Fig. 4e). In general, Ca2+/Mg2+ ratio decreased with soil depth (Fig. 4e). However, no significant differences were noted among soil horizons in the WZ MAP zone (Fig. 4e). In the CZ MAP zone, the A horizon had significantly higher Ca2+/Mg2+ ratio than the remaining soil horizons (Fig. 4e). In the EZ MAP zone, both A and Eg horizons had significantly higher Ca2+/Mg2+ ratios compared to Btg and Cg horizons (Fig. 4e). Such vertical trends of the Ca2+/Mg2+ ratio mainly resulted from the Ca2+ ions being more strongly adsorbed in soils than the Mg2+ ions. The preferential adsorption of Ca2+ ions is particularly pronounced in surface soil horizons that are rich in SOM and subjected to intense base recycling (see Curtin et al., 1998; Shaw et al., 2001). All soil horizons in the EZ MAP zone had higher Ca2+/Mg2+ ratios than the analogue soil horizons in the WZ MAP zone (Fig. 4e). Moreover, Eg and Cg horizons in the EZ MAP zone had higher Ca2+/Mg2+ ratios than Eg and Cg horizons in the CZ MAP zone (Fig. 4e). Such a trend for

347

the Ca2+/Mg2+ ratio along the MAP zones can be explained with the increasing acidification rates along the precipitation gradient (see Fig. 4a and b). In each MAP zone, SOC content decreased with increasing soil depth, with no significant difference between Btg and Cg horizons in the EZ MAP zone (Fig. 4c). In the A horizons, SOC content amounted to 6.1% (CZ MAP zone), 6.4% (EZ MAP zone), and 8.9% (WZ MAP zone) (Fig. 4c). High variability of SOM content in forest soils (e.g., Pernar et al., 2009) increased the SEM values for SOC content in A horizons (Fig. 4c). Hence, significant difference for the SOC content among A horizons was determined between the WZ MAP zone and the CZ MAP zone, but not between the WZ MAP zone and the EZ MAP zone (Fig. 4c). Nevertheless, our results confirm that MAP affects the amount of SOM (e.g., Alvarez and Lavado, 1998; Dahlgren et al., 1997; Wanhong and Yao, 2006). 4. Conclusions We investigated 33 Pseudogley profiles across the Pannonian region of Croatia. Each soil profile was developed on loess parent material and under climax forest vegetation. The trends observed along profile depth for particle size distribution, SOC content, pH, Ca2+/Mg2+ ratio, and BS point to predominance of top-down formation of Pseudogleys. However, to definitely discern if and to what extent the sedimentation/erosion processes affected Pseudogley formation in Croatia, additional (geochronological) research is needed. Relief and climate strongly affect the distribution of Pseudogleys across the study region. Namely, Pseudogleys are least abundant in the most eastern part of the area, which represents the lowest and the driest part of the Pannonian region of Croatia. The incomplete homogeneity of loess parent materials across the study region seems to be the main factor responsible for the variations in clay content, silt content, and CEC among the investigated profiles. On the other hand, MAP gradient clearly governed the decrease in soil pH, BS, and Ca2 +/Mg2 + ratio, and the increase in SOC content along the investigated profiles. Therefore, future studies of climate impact on loess-derived soils in this part of the Pannonian Basin should preferably relate to soil chemical properties that are not directly dependent on particle size distribution. The relief position on which soil pits were situated had no effect on Pseudogley characteristics. However, additional research on Pseudogley properties in respect to different soil positions along the slope is needed. Nevertheless, it appears that Pseudogleys should not be systemized according to their position on plateau or on slope, as it is often the case in Croatia and several surrounding countries. Acknowledgments This research was financially supported by the Ministry of Science, Education and Sports of the Republic of Croatia (project No. 1781780692-2711). We would like to thank the two anonymous reviewers for their valuable contributions to the paper. Appendix A. KML file containing the Google maps with the locations of the Pseudogleys investigated in Croatia. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.catena.2014.12.024. These data include Google map of the most important areas described in this article. Appendix B. Properties of 33 Pseudogleys investigated in Croatia. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.catena.2014.12.024.

348

V. Rubinić et al. / Catena 127 (2015) 340–348

References Alaoui, A., Caduff, U., Weingartner, R., 2011. Preferential flow effects on infiltration and runoff in grassland and forest soils. Vadose Zone J. 10, 367–377. Alexandrovskiy, A.L., 2007. Rates of soil-forming processes in three main models of pedogenesis. Rev. Mex. Cienc. Geol. 24 (2), 283–292. Alvarez, R., Lavado, R.S., 1998. Climate, organic matter and clay content relationships in the Pampa and Chaco soils, Argentina. Geoderma 83, 127–141. Bašić, F., 2013. The Soils of Croatia. World Soils Book SeriesSpringer, Dordrecht. Bockheim, J.G., 1980. Solution and use of chronofunctions in studying soil development. Geoderma 24, 71–85. Bogunović, M., Vidaček, Ž., Husnjak, S., Sraka, M., 1998. Inventory of soils in Croatia. Agric. Conspec. Sci. 63 (3), 105–112. Ćirić, M., 1984. Pedologija. Svjetlost, Sarajevo. Constantini, E.A.C., Angelone, M., Daminani, D., 2002. Physical, geochemical and mineralogical indicators of aging in Quaternary soils of central Italy. 17th WCSS proceedings: Confronting new realities in the 21st century 599, pp. 1–9 (14–21 August 2002, Bangkok). Curtin, D., Selles, F., Steppuhn, H., 1998. Estimating calcium–magnesium selectivity in smectitic soils from organic matter and texture. Soil Sci. Soc. Am. J. 62 (5), 1280–1285. Dahlgren, R.A., Boettinger, J.L., Huntington, G.L., Amundson, R.G., 1997. Soil development along an elevational transect in the western Sierra Nevada, California. Geoderma 78, 207–236. Daniels, R.B., Gamble, E.E., Cady, J.G., 1971. The relation between geomorphology and soil morphology and genesis. Adv. Agron. 23, 51–88. Egli, M., Mirabella, A., Sartori, G., 2008. The role of climate and vegetation in weathering and clay mineral formation in late Quaternary soils of the Swiss and Italian Alps. Geomorphology 102, 307–324. FAO, 2006. Guidelines for Soil Description. fourth ed. FAO, Rome. Gaeghan, J.P., 2012. Statistical techniques II. Accessed on January 13, 2013. http://www. stat.lsu.edu/faculty/geaghan/EXST7015/Spring2012/PDF/Lecture%2026.pdf. Galović, L., Frechen, M., Halamić, J., Durn, G., Romić, M., 2009. Loess chronostratigraphy in Eastern Croatia—a luminiscence dating approach. Quatern. Int. 198 (1–2), 85–97. Griffiths, R.P., Madritch, M.D., Swanson, A.K., 2009. The effects of topography on forest soil characteristics in the Oregon Cascade Mountains (USA): implications for the effects of climate change on soil properties. Forest Ecol. Manag. 257, 1–7. Haase, D., Fink, J., Haase, G., Ruske, R., Pecsi, M., Richter, H., Altermann, M., Jager, K.D., 2007. Loess in Europe — its spatial distribution based on a European Loess Map, scale 1:2.500.000. Quatern. Sci. Rev. 26, 1301–1312. Husnjak, S., Romić, M., Pernar, N., Poljak, M., 2011. Recommendations for soil management in Croatia. Agric. Conspec. Sci. 76 (1), 1–8. IUSS Working Group WRB, 2006. World reference base for soil resources 2006. World Soil Resources Reports No. 103. FAO, Rome. Janeković, Gj, 1960. Starost i geneza pseudogleja na beskarbonatnom praporu jugozapadnog oboda panonskog bazena. Habilitacijski rad. Poljoprivredni fakultet u Sarajevu, Sarajevo. Jenny, 1994. Factors of Soil Formation: A System of Quantitative Pedology. Dover Publications, Inc., NY (first published by the McGraw-Hill Book Company, Inc., 1941). Khan, Z.H., Hussain, M.S., Ottner, F., 2012. Morphogenesis of three surface–water gley soils from the Meghna floodplain of Bangladesh. Dhaka Univ. J. Biol. Sci. 21 (2), 17–27. Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1996. SAS® System for Mixed Models. SAS Institute Inc., Cary, NC. Munsell, Color, 2000. Munsell Soil Color Charts: Year 2000. Revised Washable Edition. GretagMacbeth, Grand Rapids, MI. Mutić, R., 1990. Korelacija kvartara istočne Slavonije na osnovi podataka mineraloškopetrografskih analiza. Acta Geol. 20 (2), 1–52.

Nesbitt, H.W., Young, G.M., 1982. Early Proterozoic climates and plate motions inferred from major element chemistry of lutites. Nature 299, 715–717. Perčec Tadić, M., 2008. Karta srednje godišnje količine oborine. In: Zaninović, K., GajićČapka, M., Perčec Tadić, M. (Eds.), Klimatski atlas Hrvatske 1961.–1990., 1971.– 2000./Climate Atlas of Croatia 1961–1990, 1971–2000. DHMZ, Zagreb, p. 52. Pernar, N., Vukelić, J., Bakšić, D., Baričević, D., Perković, I., Miko, S., Vrbek, B., 2009. Soil properties in beech-fir forests on Mt. Medvednica (NW Croatia). Period. Biol. 111 (4), 427–434. Phillips, J.D., 2004. Geogenesis, pedogenesis, and multiple causality in the formation of texture-contrast soils. Catena 58, 275–295. Pietsch, D., Lucke, B., 2008. Soil substrate classification and the FAO and World Reference Base (WRB) systems: examples from Yemen and Jordan. Eur. J. Soil Sci. 59, 824–834. Ray, L.L., 1963. Silt–clay ratios of weathering profiles of Peorian loess along the Ohio Valley. J. Geol. 71 (1), 38–47. Renssen, H., Vandenberghe, J., 2003. Investigation of the relationship between permafrost distribution in NW Europe and extensive winter sea-ice cover in the North Atlantic Ocean during the cold phases of the Last Glaciation. Quatern. Sci. Rev. 22, 209–223. Resulović, H., Čustović, H., Čengić, I., 2008. Sistematika tla/zemljišta: nastanak, svojstva i plodnost. Poljoprivredno-prehrambeni fakultet Univerziteta u Sarajevu, Sarajevo. Rubinić, V., Durn, G., Husnjak, S., Tadej, N., 2014. Composition, properties and formation of Pseudogley on loess along a precipitation gradient in the Pannonian region of Croatia. Catena 113, 138–149. Rubinić, V., Galović, L., Husnjak, S., Durn, G., 2015. Climate vs. parent material—which is the key of Stagnosol diversity in Croatia? Geoderma 241/242, 250–261. Sauer, D., Schulli-Maurer, I., Sperstad, R., Sorensen, R., Stahr, K., 2009. Albeluvisol development with time in loamy marine sediments of southern Norway. Quatern. Int. 209, 31–43. Schoeneberger, P.J., Wysocki, D.A., Benham, E.C., Broderson, W.D. (Eds.), 2002. Field Book for Describing and Sampling Soils, Version 2.0. National Soil Survey Center, USDANRCS, Lincoln, NE. Shaw, J.N., West, L.T., Hajek, B.F., 2001. Ca–Mg ratios for evaluating pedogenesis in the piedmont province of the southeastern United States of America. Can. J. Soil Sci. 81, 415–421. Shaw, J.N., West, L.T., Bosch, D.D., Truman, C.C., Leigh, D.S., 2004. Parent material influence on soil distribution and genesis in a Paleudult and Kandiudult complex, southeastern USA. Catena 57, 157–174. Škorić, A., 1986. Postanak, razvoj i sistematika tla. Fakultet poljoprivrednih znanosti Sveučilišta u Zagrebu, Zagreb. Škorić, A., Filipovski, G., Ćirić, M., 1985. Klasifikacija zemljišta Jugoslavije. Akademija nauka i umjetnosti Bosne i Hercegovine, Sarajevo. Venkatesh, B., Lakshman, Nandagiri, Purandara, B.K., Reddy, V.B., 2011. Analysis of observed soil moisture patterns under different land covers in Western Ghats, India. J. Hydrol. 397, 281–294. Wacha, L., Frechen, M., 2011. The chronology of the “Gorjanović loess section” in Vukovar, Croatia. Quat. Int. 240, 87–99. Wanhong, D., Yao, H., 2006. Relation of soil organic matter concentration to climate and altitude in zonal soils of China. Catena 65, 87–94. Wright, W.R., Foss, J.E., 1972. Contributions of clay and organic matter to the cation exchange capacity of Maryland soils. Soil Sci. Soc. Am. J. 36 (1), 115–118. Yang, L., Wei, W., Chen, L., Jia, F., Mo, B., 2012. Spatial variations of shallow and deep soil moisture in the semi-arid Loess Plateau, China. Hydrol. Earth Syst. Sci. 16, 3199–3217.