Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy

Quaternary International xxx (2015) 1e13

Contents lists available at ScienceDirect

Quaternary International journal homepage: www.elsevier.com/locate/quaint

Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy Claudio Colombo a, *, Giuseppe Palumbo a, Erika Di Iorio a, Filippo Russo b, Fabio Terribile c, Zhaoxia Jiang d, Qingsong Liu d a

Dipartimento Agricoltura Ambiente Alimenti, DIAAA, University of Molise, v. De Sanctis, I-86100 Campobasso, CB, Italy
degli Studi del Sannio, Via dei Mulini 59/A, 82100 Benevento, Italy Dipartimento di Scienze e Tecnologie, Universita
degli Studi di Napoli “Federico II”-Via Universita
, 100-80055 Portici, NA, Italy Dipartimento di Agraria, Universita d State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China b c

a r t i c l e i n f o

a b s t r a c t

Article history: Available online xxx

The aim of this paper is to enrich our knowledge of an important paleosol succession located in Bojano basin of the southern Apennines (Italy), with new pedological, geochemical and magnetic data. The studied area consists of alluvial and fluvial-lacustrine sequences (>160 m) dating from the Middle Pleistocene (0.5 Ma). The study area shows the presence of recent soil consisting of well-developed Andosols (RS), and several clastic sedimentary levels alternating with four layers (Solum I, II, III and IV) of paleosols. Soil and paleosols were analyzed by laser grain size distribution (GSD), diffuse reflectance spectroscopy (DRS), trace elements, and magnetic properties in order to evaluate the relative contributions of pedogenic and detrital components. Results showed that the finest pedogenic ferrimagnetic grains exhibit two trends with respect to the degree of pedogenesis indicate two different pedoclimate formations. The paleosol sequence consists of highly-weathered Vertisols (Solum I and IV) and of less weathered Entisols (Solum II, III). The recent soil (Andosol) has a strong bimodal distribution formed mostly by coarse silt-size particles related to the volcanic parent material. Solum I showed a sharp unimodal clay GSD while Solum III and IV were composed of bimodal GSD with high percentages of fine silt-size particles. On the basis of the trace element content and Gt/cfd ratio, all Solum (I, II, III, IV) exhibited low weathering pedogenesis compared with RS and negligible contribution to the magnetic properties of the coarse fractions. This occurs in Vertisols which developed under humid temperate climates (Solum I and SIV) and formed below the layer of Neapolitan Yellow Tuff, developed after 12 e15 ka BP. In Solum, II, III the finest sedimentary levels, the low rate of pedogenesis could have developed under more cold climatic conditions after the last eruption (Campanian Ignimbrite, 39 ka) in the Late Pleistocene. © 2015 Elsevier Ltd and INQUA. All rights reserved.

Keywords: Fluvial-lacustrine sediments Paleopedology Magnetic enhancement Iron oxides Southern Italy

1. Introduction The spatial distribution of volcanic soils in southern Apennines and in other mountain ecosystems is especially important in Italy (Iamarino and Terribile, 2008; Vacca et al., 2009). Volcanic activity in Central-Southern Italy during the late Pleistocene and the Holocene produced large amounts of pyroclastic deposits that covered wide areas of the Central and Southern Apennines (Italy) (Scandone et al., 1991; Scarpati et al., 1993; Vacca et al., 2003; Magliulo et al., 2006; Colombo et al., 2007), including the Matese Massif where an

* Corresponding author. E-mail address: [email protected] (C. Colombo).

intermontane basin was located (Colombo et al., 2014). The Phlegraean Fields have produced two important eruptions since ~39 ka. These eruptions produced the Campanian Ignimbrite and the Neapolitan Yellow Tuff (Rolandi et al., 2003). The eruption which produced the Campanian Ignimbrite occurred as a large-volume pyroclastic flow associated with a pyroclastic fall deposit and resulted in the formation of a large amount of volcanic material throughout the central-southern Apennines (Lulli, 2007). The last eruption occurred 12e15 ka and produced and the locally-named Tufo Giallo Napoletano (Neapolitan Yellow Tuff, Di Vito et al., 1999). Despite the importance of these pyroclastic deposits (Santacroce et al., 2008), not much is known about their importance in terms of soil development and soil properties in the central-southern Apennines (Lulli and Bidini, 1980; Quantin et al.,

http://dx.doi.org/10.1016/j.quaint.2015.11.004 1040-6182/© 2015 Elsevier Ltd and INQUA. All rights reserved.

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

2

C. Colombo et al. / Quaternary International xxx (2015) 1e13

1988; Quantin and Lorenzoni, 1992). Moreover, little is known in terms of the association between specific volcanic soils and the source of their volcanic ash parent material (Mileti et al., 2013). This lack of knowledge is related to the specific difficulties dealing with these soils. Among these are rapid weathering and erosion, typical of andic soils, which hinder their preservation and their study over the landscape (Wada, 1985; Kimble et al., 2000). This is unfortunate because a detailed understanding of the relationships between volcanic parent material, soil processes, and the distribution of volcanic soil and buried soil would also be very important in terms of paleoenvironmental landscape reconstruction (Malucelli et al., 1999). The reasons for the repeated alternation of stages dominated by alluviation or pedogenesis are of particular interest. The presence of buried soil sequences poses a problem of the relationship between syngenetic and epigenetic soil formation related with the weathering of the pyroclastic parent material. In this sense, due to its conservative geomorphic setting, the Bojano intermountain basin offers a great opportunity to study these relationships. Therefore the aim of this paper is to enrich our knowledge of the outcropping fluvial succession Bojano intermontane basin, located in the southern Apennines (Italy), with new pedological, geochemical and magnetic data. This is in order to improve our understanding of the Quaternary paleosol evolution of the colluvial layers present in this part of the sequence. Russo and Terribile (1995) studied the paleosol succession (3 m thick) in the steep cliffs of the Bojano basin and attributed it to the Upper PleistoceneHolocene. This research aim to complete this observation with more information on the relationship between debris-flow source and the age of the involved pyroclastic deposits, as is the possible correlation for paleoclimatic reconstructions and soil development occurred in the southern Apennines (Italy). 2. Geological settings and previous studies The Bojano intermontane basin (about 500 m a.s.l.) is located in the Molise sector of the Apennines (Southern Italy). It is a large Quaternary morphotectonic depression, approximately 22 km NWeSE long and 4 km wide which is formed of tectonized Meso-Cenozoic limestones and terrigenous sediments about 2e3 km thick (Fig. 1). The Biferno River drains (towards Adriatic Sea) the intermontane basin, and, for a few of meters, partially cut its Pleistocene fluvial-lacustrine filling (Rosskopf and Scorpio, 2013). The basin was probably generated in the Lower and Middle Pleistocene due to strike-slip and extensional tectonics (still active), following the Mio-Pliocene compressive deformations of the south-Apennine orogen (Corrado et al., 1997, 2000; Blumetti et al., 2000; Galli et al., 2002; Di Bucci et al., 2005; Amato et al., 2010, 2012). The Bojano basin is between the northern edge of the Matese Mountains and the southern edge of the Sannio Mountains (Southern Italy) (Fig. 2). It is a large morphostructural depression produced by tectonic movements which occurred during the Pleistocene. Its original morphology of gentle sloping topographical surfaces is almost intact. The infilling sediments, about 20 m thick, can be observed in the southeastern portion of the basin (at Campochiaro limestone quarries) and consists of coarse alluvial fan and silt lake sediments (Aucelli et al., 2011b, 2013). Basically thin pedogenic, pyroclastic and epiclastic levels are interbedded with the alluvial sediments. The sediments are interbedded with levels of paleosols with different types of pedogenic features (Russo and Terribile, 1995; Guerrieri et al., 1999). They apparently show no signs of tectonic deformation, although synsedimentary and active tectonic movements have been documented (Galli et al., 2002; Amato et al., 2010, 2012), which are connected to the bordering slopes. Russo and Terribile (1995) studied the exposed part of the paleosol succession in the Bojano basin and attributed it to the

Fig. 1. Map of the study area in Southern Italy. General overview of the study area showing the Matese Massif in relation to volcanic centres (Roccamonfina, Phlegrean Fields and Somma Vesuvius) to the southwest. Rectangle indicates the location of the Bojiano Basin.

Upper Pleistocene-Holocene. These authors attributed the remaining part (160 m thick), to the Middle Pleistocene. This is coherent with observations (lake sediments at Basin of San Massimo) of Brancaccio et al. (1979a) attributing the lacustrine sediments of the San Massimo to the Lower Pleistocene (about 0.97/ 1.13 Ma). Di Bucci et al. (2005) have recently dated the same sediments through the 40Ar/39Ar method to Middle Pleistocene (0.6 Ma) origin. In this framework, Amato et al. (2010, 2011, 2012) and Aucelli et al. (2011a; 2011b) provide further and more detailed chrono-stratigraphic knowledge, unfortunately without reaching the pre-lacustrine bedrock. The drill core in the Bojano territory reaching the depth of 160 m allowed the authors to reconstruct a detailed chrono-stratigraphic and paleoenvironmental succession of the basin (Fig. 3). Overall, this succession consists of three litho-stratigraphic units; the first two (UQS1 and UQS2), the oldest, are buried below the present basin surface and the, more recent, third (UQS3) is partly cropping out. These three units are separated by a clear erosional surface (Es). The oldest unit (UQS1), which is approximately 80 m thick, consists of layers of clay, silty/clay and carbonaceous clay with reworked volcanoclastic levels. The examination of lithofacies reveals a clear sedimentation of limno-marsh environments with infrequent flood episodes. Two pyroclastic layers in this unit, from the Roccamonfina Volcano, have been dated, using the 40Ar/39Ar method, to 426 ± 5.5 ka and 437.9 ± 5.5 ka. Bio-stratigraphic and pollen data of this unit suggest that has been deposited between 500 and 400 ka during the development of an Interglacial-Glacial-Interglacial cycle lasting about 100 ky. The intermediate unit (UQS2), about 40 m thick, consists of sands and silty limno-marshy layers stratified with frequent peats. The upper part is almost entirely made up of alluvial fan sand and gravel layers. In the UQS2 unit, from the Roccamonfina Volcano, has been dated to 331 ± ka, using the 40Ar/39Ar method. The exposed (partially), youngest unit (UQS3) is approximately 30 m thick and consists of alternating layers of limno-marshy sandy, silty and carbonate-rich clay as it passes upwards to more frequent levels of alluvial fan sands and gravels with reworked volcanoclastic and pedogenic levels. The UQS3 contains a paleosols

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

C. Colombo et al. / Quaternary International xxx (2015) 1e13

3

Fig. 2. A) Ubication of the study area (red rectangle) in the Bojano Basin. B) Geological sketch of the study area (From: Guerrieri et al., 1999, mod.). Legend: 1. Colluvial deposits (Holocene); 2. Recent alluvial deposits (Holocene); 3. Lacustrine and marsh deposits of the Bojano Basin (Upper Pleistocene); 4. Alluvial fan deposits (Upper Pleistocene); 5. Lacustrine deposits of the S. Massimo Basin (Lower (?) e Middle Pleistocene); 6. Meso-Cenozoic sandstones and shales purple (Varicolori Shales Form.); 7. Meso-Cenozoic calcarenites and pelagic marls (Molisano/Lagonegrese Basin); 8. Meso-Cenozoic shelf limestones and dolostones (“Abruzzese-Campana Platform”); 9. Old erosional surface. (the “Paleosuperficie Auct.”, Pliocene e Lower Pleistocene ?); 10. Erosional surface related to the top of the S. Massimo basin (Middle Pleistocene); 11. Erosional surface related to the top of the Bojano basin (Upper Pleistocene-Holocene); 12. Active norma1 faults; 13. Elevations (m a.s.l.). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

sequence consisting of recent soil (namely as RS) and several clastic sedimentary levels alternating with four layers of paleosols (Solum I, II, III and IV) located in the present basinal surface (UQS3, Fig. 4). Dating information is provided by the pyroclastic layers, from the Phlegrean Fields, observed in the upper part, about 2 m thick, of this unit (below the RS), dated to 12 and 15 ka, using the 40Ar/39Ar method (Deino et al., 2004). In addition, the downcutting of such deposits and the subsequent alluvial and volcanoclastic Phlegrean filling of the channels are attributed to the Late Glacial-Holocene (Narcisi, 1996; Ramrath et al., 1999).

3. Material and methods 3.1. Field observations, location, and description of sites The location for this study is in the SE of Bojano basin (41,4691N; 14,5159E) in the Campochiaro village (CB), southern Apennines (Italy). It is a large, flat and fan-shaped area with a weakly concave longitudinal profile corresponding to the surface of an old, incised and inactive alluvial fan which extends across the plain for more than 3 km. The steep cliffs of active and abandoned gravel quarries on the surface of the fan allow analysis of numerous stratigraphic sections which show a fairly complete and homogeneous, succession of the last episodes of alluvial depositional bodies infilling the Bojano basin (Fig. 4). The eroded surface of the fan is covered by a thick (1e2 m) discontinuous blanket of water-laid and pyroclastic deposits (essentially fine pumice and ash fall sediments whose age and provenance are unknown) deeply pedogenized as Andosols. The thick covering Andosols with associated water-laid deposits reflect a prolonged biostatic regime which still persists in the present Biferno alluvial Plain. Recent soil (RS) is the result of in situ pedogenic alteration (A, Bw1, Bw2 horizons) on calcaric parent materials (C1 and C2) of mainly plinian ash fall pyroclastics from the Campanian volcanic districts (Phlegrean Fields and Somma-Vesuvius) and relates to the major plinian eruptions which occurred during the Holocene (De Vivo et al., 2002; Rolandi et al., 2003; Colombo et al., 2014).

In steep cliffs, it can be observed that the current topographic surface is the expression of a younger alluvial, maximum of 20 m in thickness, conglomerate (B in Fig. 4) which consists almost entirely of massive and disorganized limestone gravels. This first part of the sedimentary sequence, mainly Würmian and Holocene, rests erosionally on a different, underlying, sedimentary body (B in Figs. 4 and 5) which consists homogeneously of coarse alluvial gravels made by debris flows showing the typical fan-shape with the geomorphological characteristics of the Apennine alluvial fans (Frezzotti and Giraudi, 1992; Frezzoti and Narcisi, 1996; Russo Ermolli et al., 2010). These gravel deposits are denoted by their whitish color, typical of local limestone, and their variable thickness, from 1 to about 10 m, in both transverse and longitudinal profiles (Fig. 4). These alluvial deposits pass upstream to cryoclastic breccias which relate to the physical degradation of the local mountain fault slopes (Coltorti et al., 1983), and were modelled in Mesozoic limestone and dolomitic-limestone through the mechanism of slope replacement during the Würmian (Brancaccio et al., 1979b; Russo and Terrible, 1995). A second conglomerate (C in Fig. 4) consists of calcareous gravels with a characteristic whitish color which renders clearly visible their basal erosive contact on a lower and older yellowish alluvial succession, exposed for just a few meters. This latter alluvial succession is also made up of limestone gravels, but with an abundant fine polygenic matrix (perhaps a result of the dismantling of Meso-Cenozoic terrigenous, siliciclastic terrains outcropping in the surrounding mountains). The difference in color allows immediate, in-the-field identification of these two overlapped conglomerate units (B and C, Fig. 4). The sedimentological, geometrical and textural characteristics of the levels in both of the conglomeratic sequences are typical of clastic bodies deposited in alluvial fan environments (Eyles and Kocsis, 1988). In terms of the age of formation of these two alluvial bodies, some regional geomorphological consideration suggests that the Upper Pleistocene is highly probable. In particular, the deposits in the younger alluvial fan (B, Fig. 4) can be traced upstream to a final Glacial

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

4

C. Colombo et al. / Quaternary International xxx (2015) 1e13

Fig. 3. Stratigraphic, chronological, climatic (Oxygen Isotopic Stages e O.I.S. e curve) and tephra dataset regarding the known fluvial-lacustrine infilling of the Bojano basin (From: Aucelli et al., 2011, mod.).

Fig. 4. Panoramic view of the steep cliff of an abandoned quarry near the Campochiaro village (CB). Two alluvial fan conglomeratic bodies (B and C) are superimposed and separated by a pedogenic band (Fp1 and Fp2) marking an important erosional surface (Es1 and Es2).

talus debris which is still perfectly “attached” to the slope. The age of the alluvial fan may be Late Würmian, in agreement with the radiometric data of Guerrieri et al. (1999) and Aucelli et al. (2011a, b). As shown in Fig. 4, this alluvial body contains two clear erosional unconformities: Es1 that correspond to the C1 and C2 horizons of the RS and Es2 on a third polygenic conglomeratic body (C) which is richer in sandy-silty yellowish matrix. The clasts of the latter body are heterogeneous, mainly calcareous (although there are also a few of a bauxitic, sandstone and marl-clayey nature) and are essentially etherometric, with dimensions ranging from fine gravels to boulders. The channeled texture and the high levels of degradation and rounding are indicative of fluid transport (Es1, Fig. 4). This conglomeratic body, as a whole, appears crudely stratified and presents clear geometries of very wide and shallow channels that are typical of runoff (stream flows) running across the surface of the alluvial fans (Rachocki, 1981). Some channels are highlighted by large incisions made in brownish well-classed sandy levels that are as much as 0.5 m in thickness. These very laterally discontinuous layers are associated, in places, with thin layers of clay, and both testify to the presence of typical overbank areas in the alluvial fan inter-channel environment. A key feature of the two alluvial sequences is the presence, constant in all outcrops, of two pedogenic bands (Fp1 and Fp2 in Fig. 4) thick (<1 m) in some places and more or less continuous in the upper part of the lower conglomeratic sequence (C in Fig. 4).

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

C. Colombo et al. / Quaternary International xxx (2015) 1e13

5

H2O. Total organic carbon (TOC) was determined by wet digestion with a WalkleyeBlack procedure. Cation exchange capacity (CEC) and exchangeable-base cations were extracted with 0.5 N BaCl2-TEA at pH 8.2. Available Phosphorus (Olsen Phosphorus) was determined extraction using alkaline sodium bicarbonate (pH 8.5) solution and determining the P with solfo molybdic complex. Total Fe, Mn, S, Cu, Ni, Zn, As, Be, B, Cd, Cr, Pb, Sr, P and V were extracted with nitric acid and hydrochloric acid, 1:3 v/v in a hot bath (140
C) for 60 min, the solutions were analysed by Inductively Coupled Plasma (ICP) spectroscopy. The granulometry analysis GSA was obtained using laser diffractometry (LD) Malvern Mastersizer 2000, in the particle size range of 0.1e2000 mm with the some pretreatment described for the pipette method. Granulometry was obtained using the following settings: pump speed: 2500e3000 rpm; number of measurements 6e10; time of measurements: 90 s (1000 measures/ s); refraction index: 1.52; absorption index: 0.1. 3.3. Spectral data collection Fig. 5. Detail of the cliff in an abandoned quarry near Campochiaro (CB): i) Recent soil (RS), classificated as Andosol and water-laid gravels Tardiglacial e Holocene; ii) whitish gravelly alluvial fan sediments Late (conglomeratic bodies B), Upper Pleistocene (Last Pleniglacial); Es1 e Es2) erosional surfaces; iii) yellowish gravelly alluvial fan sediments (conglomeratic bodies C) and sandy-clayey colluvial (Würm II-III). The four different solum paleoenvironments (SI, SII, SIII and SIV) show evident colour differences.

The surface on which each buried horizon rests is almost always abrupt, while the summit of the horizon clearly appears to be truncated by erosion, showing that they are the remnants of thick soils that have been truncated or eroded by coarse alluvial fan gravels (Solum II; Figs. 4 and 5). Where erosion has been particularly active and/or soil thickness is very thin, we found a mixture of small pedogenic and clastic sedimentary levels (Solum III and IV in Fig. 5). The study of Bojano soils led to the identification of two main pedoenvironments: the recent soil and the four different Solum, 1 to 4. The parent material of these pedogenetic levels shows a range of variation including debris or stream flows sediment eroded from bordering slopes, perhaps after denudation processes which occurred during a biostatic phase of the last interstadial stage (Würm II e Würm III ?). These sediments vary from lacustrine, conglomeritic and, occasionally, cross and graded bedded fluvial channel sand deposits of variable sizes (RS C1 and RS C2 in Fig. 5). In particular in the sequence reported in Figs. 4 and 5, it is possible to summarize the following water-laid gravels (Tardiglacial e Holocene); the whitish gravelly alluvial fan sediments Late, Upper Pleistocene (Last Pleniglacial); (Es1 e Es2) erosional surfaces; the yellowish gravelly alluvial fan sediments (Würm II-III) and sandy-clayey colluvial and pedogenic layers (Fp1 and Fp2, Fig. 4). 3.2. Laboratory analyses and spectral data collection After sampling, the soil samples were air dried, sieved to 2 mm and then split into three sub-samples. Two samples were used for the spectroscopic and magnetic measurements, and the other samples for chemical analysis. Physical and chemical analyses were performed in the laboratory in accordance with official Italian procedures (MiPAF, 2000). Coarse sand (2.0e0.2 mm), fine sand (0.2e0.05 mm), silt (0.05e0.002 mm) and clay (<0.002 mm) fractions were separated by pipette and wet sieving following pre-treatment: H2O2 to remove organic matter, of dithioniteecitrateebicarbonate (DCB) for free iron oxides, the soil with carbonate were treated with sodium acetate buffered at pH 5 with acetic acid. The soil samples were dispersed with sodium hexametaphosphate after 14e16 h shaking. Soil pH was measured potentiometrically in soil:solution suspensions of 1:2.5

Diffuse Reflectance spectra were recorded from 350 to 2500 nm in 2 nm steps at a scan speed rate of 30 nm/min by using a JASCO V570 UV-VIS-N IR spectrophotometer with double beam recording and a single monochromator (JASCO Inc. Easton, MD). The spectrophotometer was equipped with a BaS04-coated integrating sphere of 73-mm diameter (JASCO ISN 470 integrating sphere system). Air-dried soils were gently ground in an agate mortar for at least 10 min in order to exclude the influence of particle size and surface roughness (Stoner and Baumgardner, 1981; Sellitto et al., 2009). Soil samples were softly pressed by hand, to avoid undesired particle orientation in the 8- by 17 mm rectangular glass holes, and placed into the sphere. Sphere geometry permitted the simultaneous measurement of the reflectance of soil sample and BaSO4 blank. Then the reflectance values were transformed into the KubelkaeMunk (KeM) remission function and the band of the second derivative of the KeM function spectrum were used to quantitatively estimate the relative mass concentration of hematite and goethite, as described in n (1993). detail by Torrent and Barro 3.4. Magnetic methods Magnetic susceptibility (mass-specific) was measured using the AGICO MFK1-FA Multi-function Kappa bridge control unit at a dual frequency of 976 Hz (low frequency, cL) and 15,616 Hz (high frequency, cH). The sensitivity is 1011 m3 kg1 and the accuracy is 0.1% (Hrouda, 2011). The corresponding values refer to cL and cH, respectively. The frequency-dependent susceptibility was defined as cfd ¼ (cL  cH), which is proportional to the concentration of the viscous SP particles (Worm, 1998). The relative frequency-dependent susceptibility (cfd% ¼ cfd/cL) is defined as cfd% ¼ cfd/cL. Anhysteretic remanent magnetization (ARM) was imparted in an alternating field of 100 mT with a superimposed 50 mT bias field, and was then expressed by ARM susceptibility (cARM). 4. Results 4.1. The soil and paleosols Results about soil and paleosol features enable some considerations about soil genesis and evolution and their climatic significance. It is evident that most of the Bojano catchment area is covered by volcanic material, up to 2 m in thickness and that lacustrine and clay sediments along with paleosols can be found

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

6

C. Colombo et al. / Quaternary International xxx (2015) 1e13

as relict paleofeatures in channels and gullies (Figs. 5 and 6). The best outcrops are present in a river cut on the western margin of the Bojano basin where a section of about 20 m shows the presence of successive units of light brown, fine-grained clay beds (0.5e2 m thick), intercalation of successive channel deposits of conglomeratic sand beds (~3e4 m thick) and eventually a sequence of paleosols. Palesosols can be grouped into four different solum paleoenvironments, namely SI, SII, SIII and SIV. Basic chemical soil properties are given in Table 1 where differences of chemical and physical solum properties are evident. Moreover some solum (SI-1 and SIV-3) show vertic property with slickensides and plastic consistence and other solum (SII and SIII) with carbonate illuviation features (carbonate coatings and concretions). The texture of the paleosol samples ranges from sandy silt in the recent soils to silty clay and clayey in buried paleosols.

clayey solums (Solum I-1, I-2, IV-2) are also those having the lowest carbonate content (from null to 2.3%). This finding may suggest that these soils are the most developed of the whole sequence and that carbonate leaching played an important role. The XRD analysis of the Mg and K saturated specimens indicates the presence of three strong basal reflections at 0.72 and 1.01 nm in Solum I-1 and I-2, which are indicative of a mixture of illite and kaolinite minerals. The presence of Kaolinite is coherent with the low CEC (14 and 19 mmol/kg respectively for SI-1 and SI-2), with high content of clay (>45%) indicating marked leaching conditions. Along with standard texture data (by pipette method reported in Table 1) we have performed laser granulometry in order achieve a better understanding of soil genesis by analysing the entire grain size distribution (GSD) curve. Laser GSD (based on volumetric rules) and pipette methods (based on sedimentology rules) are

Table 1 Descriptive statistics (mean in black; standard deviation in grey) of the main chemical features and pedogenetic indexes of the buried paleosols in the Bojano alluvial sequence. Sample RS A

Depth (cm) 0e20

RS Bw1

20e40

RS Bw2

40e80

RS C1

80e100

RS C2

100e150

Solum I-1

280e300

Solum I-2

320e340

Solum II-1

460e480

Solum II-2

460e520

Solum III-1

840e860

Solum III-2

900e920

Solum III-3

920e940

Solum IV-1

1000e1020

Solum IV-2

1020e1040

Solum IV-3

1040e1060

Solum IV-4

1060e1080

Sand %

Fine sand %

Silt %

Clay %

pH

EC mS/cm

CaCO3 %

TOC g/kg

N tot g/Kg

C/N

CEC Cmol(+)/kg

Polsen g/kg

10 0.2 7 0.1 9 0.1 9 0.2 9 0.1 5 0.1 5 0.1 0

61 5 53 4 60 3 60 4 52 3 24 1 27 3 45 4

12 0.6 17 0.8 14 0.7 20 0.6 20 0.6 24 0.8 23 0.7 28 0.7

17 0.1 23 0.3 16 0.2 11 0.1 19 0.2 47 1.8 45 1.2 18 0.2

6.82 0.2 7.08 0.2 6.95 0.2 7.91 0.2 7.92 0.2 7.22 0.2 7.10 0.2 7.16 0.2

147 5 73 3 73 5 78 4 63 4 69 1 83 3 81 5

4.0 2 1.0 0.5 0.5 0.2 70.0 3.5 77.0 3.9 0.0 e 0.0 e 0.0 e

39.0 6.0 23.4 2.2 23.4 1.6 0.6 0.1 0.7 0.1 4.2 0.1 5.2 0.1 2.6 0.1

4.4 0.1 2.7 0.1 2.9 0.01 0.4 0.01 0.4 0.01 0.9 0.01 0.9 0.01 0.3 0.01

8.8 e 8.8 e 8.0 e 1.5 e 1.8 e 4.6 e 5.7 e 8.7 e

25.3 3 24.3 3 24.1 3 8.2 1 5.5 1 13.9 1 17.6 3 14.2 2

13.0 0.5 8.0 0.3 18.0 1.0 9.0 1.0 13.0 1.0 28.0 1.3 30.0 1.0 11.0 0.3

42 3 40 3 33 2 42 3 28 1 23 1 26 1 24 2

27 0.8 27 0.7 41 1.7 31 0.8 38 1.8 35 0.9 39 1.5 43 1.2

23 0.8 27 1.1 25 1.2 25 0.9 28 1.1 38 1.4 26 1.2 26 1.3

7.93 0.2 7.91 0.2 7.87 0.2 7.90 0.2 7.95 0.2 7.98 0.2 7.94 0.2 7.94 0.2

130 15 78 3 88 3 102 12 79 9 81 12 88 3 102 11

24.5 2 22.5 2 10.0 1 23.0 2 20.6 2 20.6 3 21.6 2 4.8 0.1

2.5 0.1 2.4 0.1 2.2 0.1 4.0 0.1 3.8 0.1 2.3 0.1 3.4 0.1 5.6 0.1

0.2 0.01 0.8 0.01 0.7 0.01 0.9 0.01 0.5 0.01 0.1 0.01 0.4 0.01 0.6 0.01

12.5 e 3.0 e 3.1 e 4.4 e 7.8 e 16.4 e 9.7 e 10.0 e

17.6 3 15.9 2 19.6 3 16.9 3 17.2 3 16.9 3 17.9 3 15.6 2

21.0 0.8 27.0 1.1 24.0 1.2 25.0 0.8 23.0 0.3 24.0 0.5 24.0 0.2 26.0 0.7

e 8 0.1 6 0.05 1 0.01 2 0.01 6 0.03 3 0.01 8 0.05 6 0.04

TOC ¼ Total organic carbon; CEC ¼ cation exchange capacity; Ntot ¼ total nitrogen; EC ¼ electrical conductivity; Polsen ¼ phosphate extracted with Olsen method.

There are considerable pedologic variations among the four groups of paleosols (Solum I to IV) and within the groups themselves, but deeper paleosols (Solum III and IV) shows similar texture, possibly because of similar pedogenesis. The CEC is relatively high in the RS with values close to 24 mmol/kg. CEC decrease in the range of 14e19 mmol/kg in all paleosols. The pH is neutral in the RS and weakly alkaline in some groups of paleosols related with the carbonate content. Solum I-1, I-2, II-1 are carbonate free while the remnant Solum (II, III and IV) the carbonate content ranged from 20 to 24%. This result can be explained by secondary CaCO3 precipitation in the deeper paleosols, which is clearly derived from the pedogenetic calcareus dissolution and precipitation as confirmed by the granulometry analysis. The most

known to produce slightly different results at the boundary claysilt, the results of these two techniques are not equivalent, but this issue is not addressed in this paper. The granulometry analysis (laser GSD) showed that all soils can be differentiated into four broad groups into different sub units (Solum) from top to bottom (Fig. 7): (i) bimodal GSD of RS (Andosols), possibly to be related to a composite volcanic parent material of the A, Bw1 and Bw2 horizons, two kinds of silt mode occur ranging from between 3-4 mm and 25e32 mm; (ii) a broad GSD curve of residual fluvial sediments (with 70 and 77% of carbonate content) of the C1 and C2 horizons after carbonate removal respectively. GSD of untreated samples show a peak towards 500e700 mm which disappears in carbonate-free samples; this

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

C. Colombo et al. / Quaternary International xxx (2015) 1e13

7

Fig. 6. Grain-size distributions (GSD) of the paleosols in the Bojano alluvial sequence using the Malvern Mastersizer 2000. The curve is a mean of 6e10 measurements. The laser GSD of the SIII and SIV buried paleosols were done on carbonate free treated sample.

Fig. 7. Some correlation of trace element content in the paleosols sequence. Positive correlation with iron oxides indicates that they probably derived from weathering processes, negative correlation with sulphate indicates preferential accumulation of these elements in more weathered paleosols.

confirms the aggregation exerted by calcium carbonate. This GSD shows the heterogeneity of the alluvial material to be connected to the variability of the alluvial deposits; (iii) The SI-1 and SI-2 clay paleosols which have a rather sharp unimodal very fine silt-clay GSD at 4 mm. Here, there is a very limited sand fraction with a mode at 350 mm and, finally, (iv) a multimodal GSD of all other paleosols (SII to SIV) with (i) a strong very fine silt mode at 4 mm, (ii) a fine silt mode at about 20e25 mm, (iii) a medium silt mode at about 35 mm and a medium sand mode at 300e400 mm. It is evident

that the complex GSD of these soils is possibly connected to an assemblage of parent materials. These data can be interpreted as follows: The upper part (RS, Andosols) exhibits a strong silt component as could be expected by aeolian sediments such as distal volcanic ash. For instance, similar results (but with a 13 mm mode) were found by Mileti et al. (2013) in Aluandic and Silandic Andosols. These authors find that volcanic-type aeolian sediments constitute about 90% of the parent material in most andic soils formed in Apennine

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

8

C. Colombo et al. / Quaternary International xxx (2015) 1e13

Mountain and in particular in Silandic Andosols. Moreover, in the case of the Boiano recent soil, the occurrence of a multimodal medium sand component (250, 500 mm contributions) indicates that the parent material is more complex and that the fine silt volcanic ash is the major component of a much more complex assemblage. Moving downwards, it is interesting to note the importance in all Solum of the very fine silt-coarse clay component with a 4 mm mode (in SI it is unimodal). This evidence is coherent with either or both a volcanic ash of extremely fine silt (this can be found elsewhere in the world but it has not been reported in Italy) or coarse clay occurring in the alluvium. Possibly, this clay size is too coarse to be produced by pedogenesis (generally producing fine clay). It is also important to emphasize that SI-1, SI-2 and S4-4 e having a marked unimodal GSD at 4 mm e show a simpler parent material and are also the most developed soils. All other solums show much more complex GSD, coherent with a more heterogeneous parent material (multiple sources). The enhancement of silt fraction content (20e35 mm) can probably be attributed to an increase in the local input of coarser

Campanian district and medium sand mainly attributed to alluvial processes. Data for the iron oxides and major oxides analyzed (Fe2O3, MnO, etc.) are listed in Table 2. The total iron concentrations and some macro elements (P2O5 and S2O5) increase in the recent soil and diminish slightly towards the bottom of the buried paleosols in the alluvial sequence. Soil colour generally reflect the changing in iron and manganese content ranging from brown (10 YR5/ 3) to pale brown (10 YR 5/4) in the RS. Solum I and II, showed very pale yellow colour ranging from 9 YR 5/4 to gray 0Y 7/4 whereas Solum III was brown (10 YR 7/4) and Solum IV ranged from gray 0Y 7/4 to pale brown (9 YR 6/4). Soil genesis interpretation based on data on soil colour must be taken with caution, considering that goethite (yellow color) possess lower pigmenting power between the crystalline Fe oxides, and the dark pigmenting behavior of organic matter in soils, even at very low organic matter content, can significantly cover the goethite colour. Therefore, the amount of goethite cannot be easy correlated with soil color (Dudas and Pawluk, 1969; Schwertmann, 1971; Torrent and Barron, 1993).

Table 2 Descriptive statistics (mean in black; standard deviation in grey) of the main some chemical properties of the buried paleosols in the Bojano alluvial sequence. Sample

Depth (cm)

Hue

Value

Chroma

Fe2O3

Hm/(Hm+Gt)

MnO

P2O5

S2O5

RS A

0e20

9.6 YR

3.6 e

33.1 0.2

0.07 e

1.0 0.01

4.0 0.02

2.0 0.02

RS Bw1

20e40

e 9.6 YR

5.5 e 5.4 e

3.5 e

37.2 0.3

0.10 e

1.5 0.01

4.1 0.02

1.6 0.01

40e80

e 9.8 YR

5.4 e

3.2 e

37.3 0.2

0.09 e

1.7 0.01

5.2 0.04

1.6 0.02

80e100

e 9.3 YR

7.0 e

3.7 e

14.6 0.07

0.09 e

0.4 0.05

3.1 0.03

5.8 0.08

100e150

e 9.7 YR

7.7 e

3.2 e

9.9 0.02

0.08 e

0.9 0.01

2.4 0.01

5.4 0.06

280e300

e 8.9 YR

5.6 e

4.0 e

29.2 0.2

0.08 e

1.1 0.01

2.7 0.02

2.2 0.06

320e340

e 8.7 YR

5.4 e

4.1 e

40.4 0.3

0.09 e

1.6 0.01

2.8 0.01

1.0 0.01

460e480

e 0.3 Y

6.9 e

3.8 e

27.4 0.2

0.00 e

1.4 0.01

2.6 0.01

2.5 0.03

460e520

e 0.0 Y

6.9 e

3.9 e

30.1 0.2

0.02 e

1.1 0.01

2.8 0.01

2.2 0.05

840e860

e 9.5 YR

6.5 e

4.0 e

31.7 0.3

0.04 e

1.6 0.01

3.1 0.01

2.1 0.02

900e920

e 9.8 YR

6.8 e

4.0 e

34.8 0.05

0.02 e

1.4 0.01

3.4 0.01

1.5 0.04

920e940

e 9.8 YR

6.8 e

4.0 e

30.7 0.2

0.02 e

1.2 0.01

3.1 0.01

2.0 0.06

1000e1020

e 0.1 Y

6.7 e

4.1 e

32.0 0.3

0.01 e

1.2 0.01

2.8 0.01

1.9 0.04

1020e1040

e 0.0 Y

6.9 e

4.0 e

31.4 0.2

0.03 e

1.1 0.01

2.8 0.01

1.7 0.03

1040e1060

e 9.6 YR

6.6 e

3.8 e

32.2 0.3

0.03 e

1.1 0.01

2.8 0.01

2.0 0.02

1060e1080

e 9.4 YR

6.3 e

3.9 e

38.9 0.4

0.06 e

1.6 0.01

3.1 0.01

1.3 0.06

RS Bw2 RS C1 RS C2 Solum I-1 Solum I-2 Solum II-1 Solum II-2 Solum III-1 Solum III-2 Solum III-3 Solum IV-1 Solum IV-2 Solum IV-3 Solum IV-4

e

Munsell color parameter and Hm/(Hm + Gt from diffuse reflectance measurements. Fe2O3, MnO, P2O5, S2O5 ¼ Total iron, manganous, phosphate and sulfur from with acid dissolution extraction (g/kg).

materials caused by relatively low active pedogenesis during the accumulation of fluvial deposits, which suggests a colder climate. In general terms, GSD describes a more complex assemblage of parent material ranging from very fine and fine silt connected with volcanic ash produced by the numerous Plinian eruptions in the

Clay content, Fe2O3, and MnO have similar trends with depth in all sequences and the expected significant correlation between the different layers does not emerge. Fe2O3, and MnO reaches significant concentrations in the central zone (340 cm, paleosols S I-2) of the sequence (40.4 g/kg of Fe2O3, and 1.6 g/kg of MnO), with the

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

C. Colombo et al. / Quaternary International xxx (2015) 1e13

second highest values at 1060 cm (paleosols S IV-4) of 38.9 g/kg of Fe2O3, and 1.6 g/kg of MnO. The Hm/(Gt þ Hm) ratio determined with diffuse reflectance spectroscopy (DRS) indicated that goethite is one order of magnitude of hematite in the recent soil (RS) and Solum I, increasing in Solum II, III, and IV (Table 2). Hm/(Gt þ Hm) low ratio (ranging between 0.03 and 0.07) indicated also that the weathering of Fe-bearing minerals induces the formation of poorly crystalline goethite along with minor amounts of ferrihydrite (Schwertmann, 1985). The Hm/(Gt þ Hm) ratio is strictly correlated with pedoclimatic conditions (Kampf and Schwertmann, 1983). Goethite forms under acid weathering conditions in soil with poor drainage, while hematite formation needs neutral pH and good drainage conditions (Schwertmann, 1985, 1987). The warmer dry climate is more favorable for the formation and preservation of hematite than for goethite. The Fe mineralogy, determined with DRS in all the paleosols sequence (Solum II, III, IV), were essentially goethitic indicating that these soils were formed in a weathering pedoenvironment different from RS and SI. Nevertheless the Hm/(Gt þ Hm) ratio of the paleosols sequence are used with caution here, only to understand the Fe minerals evolution as important pedogenic indicators that could be related with the magnetic properties. This finding is coherent with a general view of a paleosol environment with marked soil leaching processes. Moreover, in the whole paleosol sequence the highest Hm/(Gt þ Hm) values occur only in the most developed Solum I-1, I-2, (partially IV-4). This may be coherent with more seasonal soil condition, favorable to the hematite formation, in the vertic paleosols.

9

Chemical data seems to indicate that Fe2O3 and MnO in the RS are linked to the clay fraction content and that they mainly occur as mineral compounds that strongly retain many trace elements. Higher content of Fe2O3 and MnO is observed in the buried paleosol SI-2 at 300e340 cm, whereas the same trend is not observed at the bottom of the sequence (S IV-4). 4.2. Magnetic and geochemical data The buried paleosols in the studied areas were evaluated by the trend distribution of the trace elements with depth, taken as the concentration of a given element relative to a reference (stable) element in the recent soils. Mean and standard deviation of the measured total element content are listed according to paleosol sequence (Table 3). In the soil system, some of these elements (e.g. P and S, Table 2) are considered be rather mobile because the solubility of their minerals and soil leaching, whereas other elements (the majority of trace elements, Fe and Al) are typically less mobile (Yang et al., 2010). In general terms, geochemistry shows for some of the trace elements a rather high variability in the studied sequence that depends on the depth of the paleosol. This indicates the complex environment of the investigated soils and could provide some interesting information of soil pedogenesis. For instance, RS seem to have a different geochemical behavior from the paleosols sequence. Data in Table 3 show that the As, Cd, Be and Pb generally increase in topsoil horizons and decrease with the depth of the paleosol sequence but with some differences that are indicative of different mechanisms of accumulation (Table 3). For example, As

Fig. 8. Magnetic susceptibility (c) (m3kg1), frequency-dependent magnetic susceptibility (cfd) (m3kg1), anhysteretic remanent magnetization (ARM) (m2kg1), Goethite Hematite ratio (Gt/Hm) of the Bojano fluvio-lacustrine sequence.

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

10

C. Colombo et al. / Quaternary International xxx (2015) 1e13

in soil and paleosol samples ranges between 4 and 16 mg/kg increasing in the first 140 cm (RS). In the deeper paleosols, Arsenic reaches a maximum value of 16 mg/kg (SII-2) at 340e400 cm and decreases to 9e12 mg/kg at 1080 cm, indicating that As is not preferentially bound to the clay illuviation but to Fe2O3 minerals (r ¼ 0.88, p < 0.01) and MnO (r ¼ 0.67, p < 0.01) minerals (Zhang and Selim, 2008). It is generally thought that the mobility of As in soil and sediment is strongly controlled by soil organic matter, driving the reductive dissolution of Fe and Mn(hydr)oxides, and thereby causing the redox transformation and precipitation of arsenate. Because the mobility of trace elements including heavy metals (V, Cr, Pb, and Zn) during the weathering and pedogenesis was closely related to the alteration of their host minerals such as clastic detritus eroded from the basin and pyroclastic material, respectively, the correlation of major elements to trace metals may explain the mobility from both parent materials (Zhang et al., 2008). To evaluate this issue, correlations between trace element and either Fe2O3 or S2O5 concentrations in RS, paleosols and carbonate sediments are given (Fig. 7). Trace elements Zn, V, Cr and As are significantly correlated to Fe2O3,

paleosols, Zn and Cr contents are <100 mg/kg and well associated with V and Pb (Fig. 7c and d). In RS samples (the horizons A, Bw1 and Bw2) the Zn, Cr and V contents decrease with depth and this could indicate that they probably derived from weathering processes of the primary minerals which behaves conservatively along the RS. Many trace elements (As, Cd and Pb) are also associated with anthropic activities. For example Pb increases in recent soil (RS) samples, ranging from 23 to 27 mg/kg, and decreases in the deeper paleosols (Fig. 8d). In general terms, Solum IV, lower paleosol unit, and Solum I-2 (clayly paleosol) represent a quite different soil environment compared with RS on the basis of the trace elements content and their correlations with Fe and S (Fig. 7c and d). The increasing amount of trace elements (V, Cr, As) in S1-2 and SIV-4 (Fig. 7) could be associated with the alteration of fine silt content that could be explained through a high explosive volcanic event like the Campanian Ignimbrite (>39 ka; Rolandi et al., 2003; Santacroce et al., 2008) or an older Plinian destructive eruption. Nevertheless, the high content of trace elements observed in most palesosol is coherent with the hypothesis about the impact of volcanic activity on their genesis.

Table 3 Descriptive statistics (mean in black and standard deviation in grey) of the total content of trace elements of the buried paleosols in the Bojano alluvial sequence. Sample

Depth (cm)

Cu

Ni

Zn

As

Be

B

Cd

Cr

Pb

Sr

V

16 0.1 21 0.2 23 0.4 11 0.2 16 0.3 27 0.4 31 0.7 28 0.4 28 0.3 30 0.6 30 0.6 29 0.7 29 0.7 31 0.6 29 0.6 31 0.6

63 0.6 72 0.5 71 0.5 30 0.3 18 0.1 60 0.7 87 0.9 61 0.7 67 0.6 58 0.4 79 0.8 65 0.6 58 0.5 60 0.6 57 0.4 74 0.5

14 0.3 15 0.2 14 0.1 6 0.01 4 0.01 10 0.2 16 0.4 9 0.2 10 0.1 10 0.2 11 0.2 9 0.1 9 0.1 9 0.2 10 0.1 12 0.1

4.8 0.04 5.7 0.03 6.1 0.05 1.3 0.01 0.8 0.01 2.2 0.03 5.3 0.03 1.6 0.03 2.4 0.02 2.1 0.04 2.2 0.04 1.8 0.02 1.8 0.03 1.8 0.03 2.2 0.05 3.2 0.04

8 0.05 8 0.03 6 0.02 6 0.02 5 0.01 9 0.05 11 0.07 11 0.05 10 0.06 10 0.04 12 0.07 11 0.07 12 0.07 12 0.06 11 0.06 12 0.05

1.6 0.03 1.7 0.03 2.1 0.05 0.8 0.02 0.7 0.01 1.0 0.01 2.2 0.05 0.9 0.01 1.0 0.01 1.1 0.02 1.1 0.05 0.9 0.03 0.9 0.05 0.9 0.04 0.9 0.04 1.2 0.02

29 0.6 35 0.8 41 0.8 21 0.3 12 0.1 34 0.7 54 0.9 35 0.7 40 0.9 37 0.6 53 0.9 43 0.5 42 0.6 44 0.7 39 0.5 44 0.6

23 0.2 26 0.3 27 0.5 7 0.1 4 0.03 13 0.1 25 0.3 11 0.01 14 0.07 14 0.04 15 0.1 13 0.03 12 0.04 12 0.02 13 0.01 18 0.05

61 0.5 55 0.4 66 0.6 161 1.4 154 1.8 88 0.9 31 0.2 96 0.9 86 0.9 74 1.1 58 0.5 83 0.6 78 0.9 72 0.5 75 0.4 48 1.1

48 0.7 54 0.7 48 0.5 19 0.2 13 0.1 43 0.8 72 0.9 41 0.8 45 0.7 53 0.8 54 0.5 45 0.6 50 0.5 48 0.3 52 0.4 68 0.8

mg kg1 RS A

0e20

RS Bw1

20e40

RS Bw2

40e80

RS C1

80e100

RS C2

100e150

Solum I-1

280e300

Solum I-2

320e340

Solum II-1

460e480

Solum II-2

460e520

Solum III-1

840e860

Solum III-2

900e920

Solum III-3

920e940

Solum IV-1

1000e1020

Solum IV-2

1020e1040

Solum IV-3

1040e1060

Solum IV-4

1060e1080

19 0.2 21 0.3 18 0.2 17 0.2 7 0.2 24 0.3 36 0.5 22 0.3 23 0.4 22 0.1 32 0.3 28 0.3 22 0.2 26 0.2 21 0.1 30 0.4

with (r > 0.84; p < 0.01), suggesting the behavior of these trace elements was closely related to the pedogenetic formation of Fe minerals resulting from the weathering of the volcanic parent material (Fig. 8a). The inverse correlation observed between Fe and S (0.96, p < 0.001) may depict soil leaching in the paleosol sequence. We do not observe the same behaviors between P and Fe. Probably the P minerals originated form sedimentary rocks and was poorly correlated with Fe. Cu and Ni contents of the recent soil (RS) which developed on volcanic ash were lower than the mean values in the paleosols, indicating a different relationship with parent material and local sedimentation. In most

Magnetic susceptibility c data are discussed in order to understand the main differences between paleosols associated with the Fe mineralogy. RS show higher c values ranging between 4 and 6 106 m3kg1 typical of Andosols, generally one order of magnitude higher than soil that were formed on metamorphic and sedimentary rocks (Lu, 2000). Magnetic susceptibility c show an interesting decreasing trend in the paleosol sequence, reducing in the SI paleosols at 2e4 106 m3kg1 to 1 106 m3kg1 in the SII, SIII and SIV. Only in selected samples (SIV paleosol) do c values showed an increasing trend into the sub sample (1e2 106 m3kg1). This behavior could be explained because many of the paleosols are

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

C. Colombo et al. / Quaternary International xxx (2015) 1e13

10

a)

8

RS

6 4

SI,II,III,IV 2 0 0

0.2

0.3

0.4

χfd (10-6m3kg-1)

c)

0.4

0.5

2

b)

1.6

RS 1.2 0.8 0.4 0 0

8

RS Two trends

0.1

0.2

0.3

0.4

χfd (10-6m3kg-1)

0.5

d)

6

ARM

χfd (10-6 m3 kg-1)

0.5

0.1

and wet periods facilitate the formation of ultra-fine grained pedogenic iron oxides (majorly maghemite and hematite) and enhance the magnetic properties of soils (Zhou et al., 1990; Liu et al., 2005, 2012). For the classic loess profiles from the Chinese loess Plateau, Spain, and Argentina, cfd% is usually larger than 10e13% (Liu et al., 2010a, b; 2012), whereas the cfd% values for the studied paleosols are much lower (4e8%). Moreover, two linear trends between cfd and c are observed for the upper most and the other samples, respectively (Fig. 9). This indicates two major populations of grain size distributions, which, in turn, indicates that there might be two possible origins of the finegrained ferrimagnetics in the samples. In addition, an evident negative correlation is observed between the goethite content and cfd, while the hematite content is not correlated with the cfd. The relatively high c and cfd% values for the RS could indicate partial addition of pedogenic ferrimagnetics typical of Andosols, which were formed during the Holocene. Moreover, the negative correlation between the goethite content (sensitive to paleoclimatic changes) and the cfd (indicator for SP þ SD particles) indicates that at least some of the SP þ SD ferromagnetic particles are associated with paleoclimatic changes. To summarize, we suggest that the fine-grained SP þ SD ferromagnetic particles in the studied soil and paleosol sequence are a mixture of two populations decreasing with depth originated both from an aeolian volcanic ash deposit and local residual sediment sediments, respectively. The paleosol sequence of the Bojano basin generally has low magnetic mineral concentrations and low

DRS band intensity IHm (10-4)

DRS band intensity IGt (10 -4)

abruptly truncated with mixed properties of B and C horizons. c, cfd, and ARM, are all generally highest in the A horizon, and gradually decrease to the background value down to the lower part of the B horizon or the C horizon. To the first order, magnetic susceptibility c represents the concentration variations in the magnetic minerals if the grain size distribution of magnetic minerals remains relatively uniform (Liu et al., 2012). Moreover, the frequency-dependent c (cfd) is more sensitive to a narrow grain size crossing the superparamagnetic (SP) and single-domain (SD) boundary (~20e25 nm for magnetite and maghemite). cfd% usually indicates the grain size distribution of SP þ SD particles and higher cfd% values correspond to narrower grain size distribution, and vice versa (Worm, 1998). Fig. 8 exhibits an overall upwardly increasing trend for c and cfd, and ARM. The similar trends among these paleosols indicate that the magnetic enhancement is dominantly controlled by increased amounts of ultrafine-grained ferrimagnetics (magnetite or maghemite) and antiferromagnetic phases (goethite). This indicates that the grain size distribution of magnetic minerals in the paleosol sequence remains relatively stable. The enhancement of magnetic properties is mainly due to the increase in concentration of SP þ SD particles. The cfd% is 7% larger for the RS soil and decreases by 1e2% for all paleosols (Fig. 9). Therefore, the magnetic particles in the RS soil have a narrower grain size distribution than in the paleosol sequence. The magnetic mineral assemblage and the associated magnetic properties of soils depend highly on paleoclimatic (or paleoenvironmental) changes (Liu et al., 2012). Usually, warm

11

0.3 0.2

SI,II,III,IV

SI,II,III,IV

RS

4

2

0.1 0

0 0

2

4

6

χ (10-6m3kg-1)

8

0

0.1

0.2

0.3

0.4

χfd (10-6m3kg-1)

0.5

Fig. 9. Correlation between the diffuse reflectance (DR) goethite band intensity (IGt) and frequency dependent magnetic susceptibility (cfd) (a), hematite band intensity (IHm) and frequency dependent magnetic susceptibility (cfd) (b), frequency dependent magnetic susceptibility (cfd) and magnetic susceptibility (c) (c), and anhysteretic remanent magnetization (ARM) and frequency dependent magnetic susceptibility (cfd) (d). Solid lines in (a), (c), and (d) indicates the linear trends.

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

12

C. Colombo et al. / Quaternary International xxx (2015) 1e13

values of frequency dependent susceptibility (Fig. 9). This result is coherent with the type of Fe minerals that were formed in an intense regime of soil leaching processes. The negative correlation between the goethtite content and the cfd indicates that the lower content of ferromagnetic particles are associated with higher goethite content in the most developed paleosols (Solum I-1, I-2). 5. Conclusion The results of the pedological and geomorphological observation together with the GSD, DRS, trace element content and the associated magnetic properties is consistent with the occurrence of several paleoclimate trends in the Bojano paleosquences. In general terms, GSD describes a more complex assemblage of parent material ranging from very fine and fine silt particles, connected with the volcanic ash produced by the numerous Plinian eruptions in the Campanian district, and medium sand mainly attributed to alluvial processes. The recent soil (Andosols) showed a more important coarse silt component with respect to the Paleosol sequence. The increasing amount of trace elements in the two part of the paleosol sequence (SI and SIV) could be associated with the alteration of fine silt. This result could be explained by the presence of volcanic ash of a strong explosive volcanic event, such as the Campanian Ignimbrite or older Plinian eruption. Only the well-developed Andosols on the top of the sequence show evidence of slightly magnetic content in their silt fraction. The four different Solum (SI, SII, SIII and SIV) show negative correlation between the goethtite content and the frequency-dependent magnetic susceptibility (cfd) indicates that the SP þ SD ferromagnetic particles are associated with different regimes of soil leaching processes. The SI and part of SIV pedoenvironment seems to indicate a more humid climate, with seasonal regime, that does not provide good conditions for the hematite accumulation and magnetic particles. In the other paleosols (SII and SII), the weathering of parent rock and Fe bearing minerals was not intense enough to produce well crystalline iron oxides. Thus, the paleosol sequence in the Bojano basin possibly records a phase of climatic change from the Last Interglacial epoch to the end of the Glacial epoch. By integrating the geochemical and rock magnetic data, two scenarios can explain the low-magnetic mineral contents, the recent volcanic soil in the upper part was formed after the yellow tuff (12e15 ka BP) at the bottom the Solum I, II and III (as indicated by low c, cARM) could be related to Wurm glaciation (approximately 110,000 years). The changes in the pedogenesis of parent material in the SIV unit with increasing magnetic susceptibility c and frequence-dependent c (cfd) could be associated with very fine volcanic ash produced in the numerous Plinian eruptions in the Campanian district. References Amato, V., Aucelli, P.P.C., Cesarano, M., Pappone, G., Petrosino, P., Rosskopf, C., Russo Ermolli, E., 2010. New chrono-stratigraphic data on the Boiano basin infilling (Molise, Italy). Rendiconti Online Della Societ
a Geologica Italiana 11, 614e615. Amato, V., Aucelli, P.P.C., Cesarano, M., Pappone, G., Rosskopf, C.M., Russo Ermolli, E., 2011. The Sessano intra-montane basin: new multi-proxy data for the Quaternary evolution of the Molise sector of the Central-Southern Apennines (Italy). Geomorphology 128, 15e31. Amato, V., Aucelli, P.P.C., Russo Ermolli, E., Rosskopf, C., Cesarano, M., Pappone, G., 2012. Quaternary morpho-evolution, tectonic and environmental changes in the Boiano intermontane basin (Central-Southern Italy). Rendiconti Online
Geologica Italiana 21, 1225e1227. Della Societa Aucelli, P.P.C., Amato, V., Baranello, S., Cesarano, M., Mastronardi, R., Monaco, R., Rosskopf, M.C., 2011a. Nuovi dati sull’evoluzione pleistocenica dei bacini intramontani di Boiano e di Sessano (Molise, Italia meridionale). Rendiconti
Geologica Italiana 12, 3e7, 614. Online Della Societa

Aucelli, P.P.C., Amato, V., Cesarano, M., Pappone, G., Rosskopf, C., Russo Ermolli, E., Scarciglia, F., 2011b. New morphostratigraphic and chronological constraints for the Quaternary palaeosurfaces of the Molise Apennines (Southern Italy). Geologica Carpathica 62 (1), 17e26. Aucelli, P.P.C., Cesarano, M., Di Paola, G., Filocamo, F., Rosskopf, C.M., 2013. Geomorphological map of the central sector of the Matese Mountains (Southern Italy): an example of complex landscape evolution in a Mediterranean mountain environment. Journal of Maps 9 (4), 604e616. Blumetti, A.M., Caciagli, M., Di Bucci, D., Guerrieri, L., Michetti, A.M., Naso, G., 2000. Evidenze di fagliazione superficiale olocenica nel Bacino di Boiano (Molise). Atti del XIX Convegno Nazionale del Gruppo Nazionale Geofisica della Terra Solida. C.N.R, Roma, pp. 12e15. Brancaccio, L., Cinque, A., Sgrosso, I., 1979 b. Forma e genesi di alcuni versanti di faglia in rocce carbonatiche: il riscontro naturale di un modello teorico. Rendiconti della Accademia Scienze Fisiche e Matematiche di Napoli, serie IV 46, 1e21. Brancaccio, L., Sgrosso, I., Cinque, A., Orsi, G., Pece, R., Rolandi, G., 1979a. Lembi residui di sedimenti lacustri pleistocenici sul versante settentrionale del Matese presso S. Massimo. Bollettino della Societa dei Naturalisti in Napoli 88, 275e286. Colombo, C., Sellitto, M.V., Palumbo, G., Terribile, F., Stoops, G., 2007. Characteristics and genesis of volcanic soils from South Central Italy: Mt Gauro (Phlegraean  Bartoli, F., Buurman, P., Fields, Campania) and Vico lake (Latium). In: Arnalds, O.,  Oskarsson, H., Stoops, G., García-Rodeja, E. (Eds.), Soils of Volcanic Regions in Europe. Springer, Berlin, pp. 197e229. Colombo, C., Sellitto, V.M., Palumbo, G., Di Iorio, E., Terribile, F., Schulze, D.G., 2014. Clay formation and pedogenetic processes in tephra-derived soils and buried soils from Central-Southern Apennines (Italy). Geoderma 213, 346e356. Coltorti, M., Dramis, F., Pambianchi, G., 1983. Stratified slope-waste deposits in the Esino River Basin, Umbria-Marche Apennines, Central Italy. Polarforschung 53, 59e66. Corrado, S., Di Bucci, D., Naso, G., Butler, R.W.H., 1997. Thrusting and strike-slip tectonics in the Alto Molise region (Italy): implications for the NeogeneQuaternary evolution of the Central Apennine orogenic system. Journal, Geological Society of London 154, 679e688. Corrado, S., Di Bucci, D., Naso, G., Valensise, G., 2000. The role of pre-existing structures in Quaternary extensional tectonics of the Southern Apennines, Italy: the Boiano basin case history. Journal of the Czech Geological Society 45 (3/4), 217. De Vivo, B., Rolandi, G., Gans, P.B., Calvert, A., Bohrson, W.A., Spera, F.J., Belkin, H.E., 2002. New constraints on the pyroclastic eruptive history of the Campanian volcanic Plain (Italy). Mineralogica et Petrographica Acta 73, 47e65. Deino, A.L., Orsi, G., De Vita, S., Piochi, M., 2004. The age of the Neapolitan Yellow Tuff caldera forming eruption (Campi Flegrei caldera e Italy) assessed by 40Ar/39Ar dating method. Journal of Volcanology and Geothermal Research 133, 157e170. Di Bucci, D., Naso, G., Corrado, S., Villa, I.M., 2005. Growth, interaction and seismogenetic potential of coupled active normal faults (Isernia Basin, CentralSouthern Italy). Terra Nova 17, 44e55. Di Vito, M.A., Isaia, R., Orsi, G., Southon, J., de Vita, S., D'Antonio, M., Pappalardo, L., Piochi, M., 1999. Volcanism and deformation in the past 12 ka at the Campi Flegrei caldera (Italy). Journal of Volcanology and Geothermal Research 91, 221e246. Dudas, M.J., Pawluk, S., 1969. Naturally occurring organo-clay complexes of orthic black chernozems. Geoderma 3, 5e17. Eyles, N., Kocsis, S., 1988. Sedimentology and clast fabric of subaereal debris flow facies in a glacially-influenced alluvial fan. Sedimentary Geology 59, 15e28. Frezzotti, M., Giraudi, C., 1992. Evoluzione geologica tardo-pleistocenica ed olocenica del conoide complesso di Valle Majelama (Massiccio del Velino, Abruzzo). Il Quaternario 5 (1), 33e50. Frezzotti, M., Narcisi, B., 1996. Late quaternary tephra-derived paleosols in central Italy's carbonate Apennines range: stratigraphical and paleoclimatological implications. Quaternary International 34e36, 147e153. Galli, P., Galadini, F., Capini, S., 2002. Analisi archeosismologiche nel Santuario di Ercole di Campochiaro (Matese). Evidenze di un terremoto distruttivo sconosciuto ed implicazioni sismo tettoniche. Il Quaternario 15 (2), 155e167. Guerrieri, L., Scarascia Mugnozza, G., Vittori, E., 1999. Analisi stratigrafica e geomorfologica della conoide tardo-quaternaria di Campochiaro ed implicazioni per la conca di Bojano in Molise. Il Quaternario 12 (2), 237e247. Hrouda, F., 2011. Model of frequency-dependent susceptibility of rock sand soils revisited and broadened. Geophysical Journal International 187, 1259e1269. Iamarino, M., Terribile, F., 2008. The importance of andic soils in mountain ecosystems: a pedological investigation in Italy. European Journal of Soil Science 59, 1284e1292. Kampf, N., Schwertmann, U., 1983. Goethite and hematite in a climosequence in Southern Brazil and their application in classification of kaolinitic soils. Geoderma 29, 27e39. Kimble, J.M., Ping, C.L., Sumner, M.E., Wilding, L.P., 2000. Andisols. In: Sumner, M.E. (Ed.), Handbook of Soil Science. CRC PRESS LLC, Boca Raton, FL, pp. 209e224. Liu, Q.S., Torrent, J., Maher, B.A., Yu, Y.J., Deng, C.L., Zhu, R.X., Zhao, X.X., 2005. Quantifying grain size distribution of pedogenic magnetic particles in Chinese loess and its significance for pedogenesis. Journal of Geophysical Research (B11), B11102. http://dx.doi.org/10.1029/2005JB003726.  n, V., Zhao, X.Y., Jiang, Z.X., Su, Y.L., 2010a. EnviLiu, Q.S., Hu, P.X., Torrent, J., Barro ronmental magnetic study of a Xeralf chronosequence in northwestern Spain: indications for pedogenesis. Palaogeography, Palaoclimatology, Palaeoecology 293, 144e156.

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004

C. Colombo et al. / Quaternary International xxx (2015) 1e13 s, H., Hong, A., Jiang, Z.X., Su, Y.L., 2010b. SuperLiu, Q.S., Torrent, J., Morra paramagnetism of two modern soils from the northeastern Pampean region, Argentina, and its paleoclimatic indications. Geophysical Journal International 183, 695e705. ~ a, J.C., Banerjee, S.K., Guyodo, Y., Tauxe, L., Liu, Q.S., Roberts, A.P., Larrasoan Oldfield, F., 2012. Environmental Magnetism: Principles and Applications. http://dx.doi.org/10.1029/2012RG000393. Reviews of Geophysics, RG4002. Lu, S.G., 2000. Lithological factors affecting magnetic susceptibility of subtropical soils, Zhejiang Province, China. Catena 40 (4), 359e373.  Bartoli, F., Buurman, P., Lulli, L., 2007. Italian volcanic soils. In: Arnalds, O.,  Oskarsson, H., Stoops, G., García-Rodeja, E. (Eds.), Soils of Volcanic Regions in Europe. Springer, Berlin, pp. 51e67. Lulli, L., Bidini, D., 1980. A climosequence of soils fromtuffs on slopes of an extinct volcano in Southern Italy. Geoderma 24, 129e142. Magliulo, P., Terribile, F., Colombo, C., Russo, F., 2006. A pedostratigraphic marker in the geomorphological evolution of the Campanian Apennines (Southern Italy): the Paleosol of Eboli. Quaternary International 156, 97e117. Malucelli, F., Terribile, F., Colombo, C., 1999. Mineralogy, micromorphology and chemical analysis of andosols on the Island of Sao Miguel (Azores). Geoderma 88, 73e98. Mileti, F.A., Langella, G., Prins, M.A., Vingiani, S., Terribile, F., 2013. The hidden nature of parent material in soils of Italian mountain ecosystems. Geoderma 207e208, 291e309. MiPAF (Ministero delle Politiche Agricole e Forestali), 2000. Metodi di analisi chimica del suolo. Collana di metodi analitici per l'agricoltura. Franco Angeli Editore, Milan, Italy. Narcisi, B., 1996. Tephrochronology of a late Quaternary lacustrine record from the Monticchio maar (Vulture Volcano, Southern Italy). Quaternary Science Reviews 15, 155e165. Quantin, P., Lorenzoni, P., 1992. Weathering of leucite to clay minerals to tephrite of the Vico volcano. Mineralogica et Petrographica Acta 35, 289e296. Quantin, P., Gautheyoru, J., Lorenzoni, P., 1988. Halloysite formation through in situ weathering of volcanic glass from trachytic pumices, Vico's volcano, Italy. Clay Minerals 23, 423e437. Rachocki, A.H., 1981. Alluvial Fans. Wiley, Chichester, p. 161. Ramrath, A., Zolitschka, B., Wulf, S., Negendank, J.F.W., 1999. Late Pleistocene climatic variations as recorded in two Italian maar lakes (Lago di Mezzano, Lago Grande di Monticchio). Quaternary Science Reviews 18, 977e992. Rolandi, G., Bellucci, F., Heizler, M.T., Belkin, H.E., De Vivo, B., 2003. Tectonic controls on the genesis of ignimbrites from the Campanian Volcanic Zone, Southern Italy. Mineralogica et Petrographica Acta 79, 3e31. Rosskopf, C.M., Scorpio, V., 2013. Geomorphologic map of the Biferno River valley floor system (Molise, Southern Italy). Journal of Maps 9 (1), 106e114. Russo, F., Terribile, F., 1995. Osservazioni geomorfologiche, stratigrafiche e pedologiche sul Quaternario del Bacino di Boiano (Campobasso). Il Quaternario 8 (1), 239e254. Russo Ermolli, E., Aucelli, P.P.C., Di Rollo, A., Mattei, M., Petrosino, P., Porreca, M., Rosskopf, C.M., 2010. An integrated stratigraphical approach to the late Middle Pleistocene succession of the Sessano lacustrine basin (Molise, Italy). Quaternary International 225, 114e127.

13

Santacroce, R., Sulpizio, R., Zanchetta, G., Cioni, R., Marianelli, P., Sbrana, A., Donahue, D.J., Joron, J.L., 2008. Age and whole rockseglass composition of proximal pyroclastics from the major explosive eruptions of Vesuvius: a review as a tool for distal tephrostratigraphy. Journal of Volcanology and Geothermal Research 177, 1e18. Scandone, R., Bellucci, F., Lirer, L., Rolandi, G., 1991. The structure of the Campanian Plain and the activity of the Neapolitan volcanoes (Italy). Journal of Volcanology and Geothermal Research 48, 1e31. Scarpati, C., Cole, P., Perrotta, A., 1993. The Neapolitan Yellow Tuffda large volume multiphase eruption from Campi Flegrei, Southern Italy. Bulletin of Volcanology 55, 343e356. Schwertmann, U., 1971. Transformation of hematite to goethite in soils. Nature 232, 624e625. Schwertmann, U., 1985. The effect of pedogenic environments on iron oxide minerals. Advances in Soil Science 1, 171e200. Schwertmann, U., 1987. Occurrence and formation of iron oxides in various pedoenvironments. In: Stucki, Joseph W., Goodman, Bernard A., Schwertmann, Udo (Eds.), Iron in Soils and Clay Minerals. D. Reidel, Dordrecht, pp. 267e308.  n, V., Colombo, C., 2009. Comparing two different Sellitto, V., Fernandes, R., Barro spectroscopic techniques for the characterization of soil iron oxides: diffuse versus bi-directional reflectance. Geoderma 149, 2e9. Stoner, E.,R., Baumgardner, M.,F., 1981. Characteristic variations in reflectance of surface soils. Soil Science Society of America Journal 45, 1161e1165. n, V., 1993. Laboratory measurement of soil color: theory and Torrent, J., Barro practice. In: Bigham, J.M., Ciolkosz, E.J. (Eds.), Soil Color. SSSA Special Publication, vol. 31. Soil Science Society of America, Madison, WI, pp. 21e33. Vacca, A., Adamo, P., Pigna, M., Violante, P., 2003. Genesis of tephra-derived soils from the Roccamonfina Volcano, South Central Italy. Soil Science Society of America Journal 67, 198e207. Vacca, S., Capra, G.F., Coppola, E., Rubino, M., Madrau, S., Colella, A., Langella, A., Buondonno, A., 2009. From andic non-allophanic to non-andic allophanic Inceptisols on alkaline basalt in Mediterranean climate A toposequence study in the Marghine district (Sardinia, Italy). Geoderma 151, 157e167. Wada, K., 1985. The distinctive properties of Andosols. In: Advances of Soil Science, vol. 2. Springer-Verlag, New York, pp. 173e228. Worm, H.U., 1998. On the superparamagnetic-stable single domain transition for magnetite, and frequency dependence of susceptibility. Geophysical Journal International 133, 201e206. Yang, T., Zhu, Z., Gao, Q., Rao, Z., Han, J., Wu, Y., 2010. Trace element geochemistry in topsoil from East China. Environmental Earth Sciences 60 (3), 623e631. Zhang, H., Selim, H.M., 2008. Reaction and transport of arsenic in soils: equilibrium and kinetic modeling. Advances in Soil Science 98, 45e115. Zhang, C.S., Fay, D., McGrath, D., Grennan, E., Carton, O.T., 2008. Statistical analyses of geochemical variables in soils of Ireland. Geoderma 146 (1e2), 378e390. Zhou, L.P., Oldfield, F., Wintle, A.G., Robinson, S.G., Wang, J.T., 1990. Partly pedogenic origin of magnetic variations in Chinese loess. Nature 346, 737e739.

Please cite this article in press as: Colombo, C., et al., Soil development in a Quaternary fluvio-lacustrine paleosol sequence in Southern Italy, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.004