uncorrected proof

Catena xxx (2018) xxx-xxx

Contents lists available at ScienceDirect

Catena

PR OO F

journal homepage: www.elsevier.com

Loess in Italy: Genesis, characteristics and occurrence

Edoardo A.C. Costantinia⁠ ,⁠ ⁎⁠ , Stefano Carnicellib⁠ , Daniela Sauerc⁠ , Simone Prioria⁠ , Anna Andreettab⁠ , Annette Kadereitd⁠ , Romina Lorenzettia⁠ ,⁠ e⁠ a

Council for Agricultural Research and Economics CREA, Firenze, Italy Dipartimento Scienze della Terra DST, Università di Firenze, Italy Institute of Geography, University of Göttingen, Germany d Heidelberg Luminescence Laboratory, Institute of Geography, University of Heidelberg, Germany e Università del Molise, Department of Agricultural, Environmental and Food Sciences, Italy b c

ABSTRACT

Keywords: Loess Soils Database Meta-analysis Pleistocene Holocene Mediterranean Italy

There is currently a renewed interest in loess genesis and occurrence in Italy. Well-known loess profiles in northern Italy have been re-examined, and previously unknown loess deposits in central and southern Italy have been reported. This work combines a meta-analysis of published data with new data, presented here for the first time, to provide the state of the art on the spatial distribution, characteristics, genesis, and deposition ages of loess in Italy. A database of 98 soil horizons from 91 soil profiles was created and made available. It stores information on soils formed from loess or containing layers that show admixture of loess. Soil data include the source of information, the topographic, geomorphological and geological setting, kind of parent material and land use, main soil horizon properties as described in the field, soil classification, particle size distribution and chemical data. Loess is reported from almost all regions of Italy. It is generally pedogenized throughout the entire loess body and forms the parent material of deep and complex soil profiles. The thickness of the loess deposits varies from a few decimeters to a few meters; pedogenesis and local reworking partially change the original loess characteristics. In this study, we propose a set of parameters that help recognizing loess in soils: particle size distribution and sorting, geomorphological setting, pedostratigraphic position, shape of the grains under optical and scanning electron microscope (SEM) and soil-micromorphological characteristics. The ages of the loess deposits range between 70 ka BP and 18 ka BP. There are also a number of samples that have been dated to the Holocene. The meta-analysis suggests that climate is not the only driver of loess deposition in Italy; geomorphological instability and human disturbance, as well as the influence of Saharan dust, most likely played major roles, too. This study demonstrates that loess is much more widespread in Italy than previously estimated. Yet, further research on the spatial and temporal distribution of loess deposition across the Mediterranean region is needed to better understand its genesis, sources and trajectories, periods of enhanced loess formation, and the role of loess deposits in ecosystem functioning and resilience.

UN CO RR EC TE D

ARTICLE INFO

1. Introduction

Loess deposits are important palaeoenvironmental archives, covering about 10% of Earth's terrestrial surface (Yang et al., 2009). As Quaternary records, loess-palaeosol successions provide relevant information on climate changes and ecosystems' resilience (Muhs, 2006). Soils formed from loess are widely used for agricultural production, e.g. in

⁎

Corresponding author. Email address: [email protected] (E.A.C. Costantini)

https://doi.org/10.1016/j.catena.2018.02.002 Received 31 July 2017; Received in revised form 29 January 2018; Accepted 2 February 2018 Available online xxx 0341-8162/ © 2017.

the northern European and Asian loess belts (Pye, 1984), as loess is the parent material of some of the world's most fertile soils (Brahy et al., 2000). Loess and dust deposits that have been completely incorporated into the soil cover play an important role, too (Lequy et al., 2012), as they provide additional fine particles, calcium carbonate and base cations (Avila et al., 1998). The aeolian input may contribute a considerable proportion of the total amount of nutrients and can enhance soil fertility and carbon sequestration capacity (Lequy et al., 2012). Aeolian

E.A.C. Costantini et al.


gin of the loess was attributed to local sources. The times of its deposition included the late Upper Pleistocene as well as the middle Holocene. A set of recently published papers documented previously unknown occurrences of loess in different parts of Italy, especially in its northern regions, namely in Trentino (Borsato, 2009), Veneto (Peresani and Nicosia, 2015), Liguria (Rellini et al., 2009) and Piedmont (Frigerio et al., 2017), including both typical and reworked loess, deposited on various substrates in different geomorphological settings. Known loess exposures along the Po Plain were reinvestigated by applying modern techniques (Amit and Zerboni, 2013). Mileti et al. (2013) discussed the presence of aeolian sediments in soils of the northern, central and southern Apennines and compared them to volcanic ejecta. Sandy and sandy-silty aeolian deposits located near the Ionian coast were reported by several authors (Wagner et al., 2007; Sauer et al., 2010; Andreucci et al., 2012). The growing interest in loess in the Mediterranean region led to the establishment of the INQUA Project “AEOMED” in 2012–2015 (Sauer et al., 2015), and its follow-up INQUA Focus Group GEODUST (2016–2019). The aims of the present study are:

PR OO F

deposits on slopes, however, are highly prone to water erosion, and also lead to an enhanced risk of landslides (Mileti et al., 2013). Loess is defined as sediment that has been mobilized, transported, and re-deposited by wind. It is dominated by silt-sized particles (2–63 μm in diameter; Wright, 2001) and characterized by stabilization due to slight synsedimentary dissolution and re-precipitation processes of the contained carbonates (Pécsi and Richter, 1996). Thus, loess is not simply dust that has settled (Sprafke and Obreht, 2016), but its typical characteristics are also a result of various initial syn- or early post-sedimentary alteration processes, which can be summarized as “loessification”. Loess has been alternatively regarded as a product of sedimentation or pedogenesis (e.g., Smalley et al., 2011 versus Pécsi, 1991). It has been differentiated into primary (wind-deposited) loess and secondary (reworked) loess, which is the product of re-deposition and mixing of primary loess with other materials (Pye, 1984). Identifying and characterizing loess contribution to soils may be a difficult task, since pedogenesis progressively alters the original properties of the loess (Cremaschi, 1987). Such challenge is common in the Mediterranean region, and contributed to the underestimation of loess occurrence in the map of Haase et al. (2007), which did not depict loess deposits in Mediterranean Italy. In contrast, the map of loess deposits by Bertran et al. (2016), derived from the grain-size distributions of European soils, considers loess occurrence in Mediterranean Italy. However, it lacks ground verification and laboratory evidence. Geological maps of Italy report loess deposits only for a few exposures, mainly located in northern Italy, close to the Alpine margin (http://www.isprambiente.gov.it/Media/carg/). The recognition of the spatial distribution of Holocene and Pleistocene loess deposits is of paramount interest, since it may allow for identifying sources, modes of transport and phases of increased deposition, which can be correlated with climatic changes, tectonic instability, and anthropogenic landscape modifications (Pye, 1984). In addition, recognition that soils with loess contribution are much more widespread in the Mediterranean countries than usually acknowledged can provide new insights into ecosystem functioning and resilience. Cremaschi (1987) published the first comprehensive review of loess deposits in Italy. His documentation focused on loess exposures along the Po Plain and the southern Alpine foreland. Some occurrences were also reported from the Adriatic basin, namely the islands of Croatia and the Conero promontory in the Marche region. It was also Cremaschi (1987) who reported for the first time that loess deposits in Italy are mostly pedogenized, and that sequences including unweathered loess are limited to a few sites. This new notion laid the foundation for successive studies, integrating sedimentological and pedological approaches (Cremaschi, 2008; Cremaschi and Van Vliet-Lanöe, 1990; Cremaschi et al., 1990; Busacca and Cremaschi, 1998). The sources of the loess were attributed to periglacial environments and braided riverbeds. In terms of timing, loess accumulation was related to the Pleistocene glaciations. The maps published first by Cremaschi (1987) and successively refined thereafter (Cremaschi, 2004), provided the first representation of the geographical distribution of loess deposits in Italy, and the first acknowledgement of their, so far overlooked, importance. Later, Giraudi and co-authors carried out several studies on the high plateaus and karst depressions of the central and southern Apennines (Frezzotti and Giraudi, 1990; Giraudi et al., 2013; Giraudi, 2015). They reported the occurrence of thick and extensive aeolian deposits, which they interpreted as Saharan dust, though they also considered possible contributions from local sources. The deposition of these sediments was attributed to the Late Glacial and Holocene periods. Priori et al. (2009) and Costantini et al. (2009) also reported the presence of loess in soils on hills of central Italy. In these cases, the ori

UN CO RR EC TE D

(1) to establish an open database for gathering and disseminating information on soils developed from loess and soils with loess additions in Italy, in a format that can be used also in other countries, (2) to carry out a meta-analysis of the spatial distribution of loess and loess additions to soils in Italy, based on published and new soil data, (3) to extract from this meta-analysis a preliminary overview of the spatial distribution and characteristics of loess in Italy, related to glacial and non-glacial environments in different regions and geomorphological settings, and (4) to develop some preliminary hypotheses on the timing and probable processes involved in the genesis of loess in the various regions of Italy. 2. Materials and methods 2.1. Study area and general approach The study area comprises the whole of Italy, stretching between 36° and 47° north, from the central Mediterranean basin to the temperate zone of the northern hemisphere (Fig. 1). Because of its geographic position, Italy represents a bridge between the loess belt of central Europe and the desert loess originating from the Sahara. The interplay of the relief of the country with the wind regimes determines the sources, trajectories and areas of deposition of aeolian sediments. A major part of Italy is hilly or mountainous; the Alps represent the highest elevations of geographic Europe (Monte Bianco 4810 m) and act as a barrier with respect to the cold winds from the Arctic regions. The other important mountain chain of Italy, i.e. the Apennines, runs northwest to southeast, thus representing an obstacle for west winds coming from the Atlantic Ocean, but also for winds from northeastern directions. Occasionally, strong warm and humid winds blow from the south (the so-called “scirocco”), carrying dust from the Sahara Desert northwards to the Alps and beyond (Costantini et al., 2013). We compiled all published work on loess in Italy that we became aware of during an extensive literature research and subjected the obtained data, together with new data of several own studies, to a meta-analysis. This meta-analysis concentrated specifically on soils formed from loess, while soils formed from other aeolian sediments, e.g., volcanic ejecta or coastal dunes, were excluded. Thus, the focus of this study was on loess deposits whose aeolian nature was clearly recognized and documented, although local slope processes and anthro

2


UN CO RR EC TE D

PR OO F


Fig. 1. Distribution of the soil profiles characterized by aeolian deposits (■ sandy loess; ● glacial loess; ▲ non-glacial loess). Red = maximum glacial expansion during the last glacial period in the Alps and Apennines, modified from Giraudi (2015) and Florineth and Schlüchter (1998). Locations of dated loess: 1 Pecceto Valenza; 2 Bagaggiaro; 3 Monte Netto; 4 Val Sorda; 5 Gajum, Fumane Cave; 6 Elsa Valley; 7 Emilia (Parma Reggio Emilia area); 8 Central Apennines; 9 Gulf of Taranto; 10 Sardinia, 11 West Sicily; 12 Buca dei Corvi. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

pogenic impact might have reworked some of them, and possibly mixed them to some extent, e.g. with colluvial materials or autochthonous soil material.

3



The loess was then classified as follows:

Table 1 Structure of the database. Class sheet Data entry

Sheet

Class variable

Fields

dataentry site

Identity

Profile; site_name

PR OO F

1) typical loess (TL), i.e. loess showing characteristics typical for loess according to Pécsi and Richter (1996), mainly consisting of silt with minor other contributions, not reworked after its aeolian deposition, 2) reworked loess (RL), i.e. loess that has been mixed with other sediments and/or has been transported by agents other than wind after its original aeolian deposition, 3) sandy loess (SL), i.e. loess exhibiting the morphology and color of typical loess, but being characterized by significant sand contents, 4) typical and reworked loess trapped in karst depressions (KL) overlying and possibly mixed with variable proportions of residual material from limestone weathering, 5) typical and reworked loess released in the context of glacial deposits (GL).

Locationa⁠

Environmental data

dataentry horizon

In the last case, the loess deposits were differentiated with respect to their genesis, either related or non-related to glacial environments. This differentiation was based on the literature and on a new map (Fig. 1) showing the expansion of glaciers during the Last Glacial Maximum (LGM). We created this map by digitizing and combining information published by Florineth and Schlüchter (1998), and Giraudi (2015). Finally, the soil horizons formed in the loess were grouped into A or E, Bw, and Bt horizons (FAO, 2006), in order to assess the influence of the formation of these soil horizons on the properties of the loess.

Pedological data Source Identity

Location Loess classesb⁠

UN CO RR EC TE D

Physical data

Physical datac⁠ (texture)

2.2. Database

A database of the state-of-the-art knowledge on loess of Italy was created, including both published and new data. The structure of the database and its main information are shown in Table 1. Loess was assumed to be present at a site, if (i) its presence was explicitly reported in the respective published study or (ii) the similarity of the site in terms of setting, physical and chemical soil characteristics with published loess deposits pointed to the presence of loess. In the latter case, the following set of key indicators of the presence or absence of loess in soils was used:

Domain

1) topographic setting: stable geomorphological position, allowing for the preservation of aeolian deposits: terrace surfaces, plateaus, karst depressions or aggradation slopes; 2) stratigraphy: soil profiles showing lithological discontinuities between the loess or loess-containing sediment (indicated e.g., by enhanced silt contents) and underlying layers; 3) soil classification and lithology of the parent material: Andosols and soils from fluvial, coastal, lacustrine, and marine parent materials were excluded; 4) soil characteristics: texture dominated by silt, with the exception of sandy loess; rock fragments absent or very few, with the exception of reworked loess; carbonates absent or very low content; 5) (sub)microscopic soil characteristics: evidence of aeolian transport on particle surfaces, as observed under scanning electron microscope (SEM) or in soil thin sections.

Chemical data

country regions physiography morpho_element substratum hue structure

Decoding data Decoding data Decoding data Decoding data Decoding data Decoding data Decoding data

consistence

Decoding data

Country; region; lat; lon Elevation (m); slope (%); physiography; morpho_element; substratum Soil class WRB Reference Profile; name_hor; top (cm); bottom (cm) Country; region Loess type; karst; glacial Hue; value; chroma; skeleton (%); structure_type; structure_size; structure_grade; consistence Clay (%); silt (%); sand (%); fine silt (%); coarse silt (%); VFS (%); FS (%); MS (%); CS (%); VCS (%) pH; Org.C (dag/kg); Inorg.C (dag/kg) Code; description Code; description Code; description Code; description Code; description Hue Domain, code, description Code; description

a lat: latitude in decimal degrees WGS84 system; long: longitude in decimal degrees WGS84. b Loess type: typical reworked or sandy loess. c VFS (%): very fine sand (0.100–0.050 mm); FS (%): fine sand (0.250–0.100 mm); MS (%): mean sand (0.5–0.250 mm); CS (%): coarse sand (1–0.5 mm); VCS (%): very coarse sand (2–1 mm); fine silt (%): 0.020–0.002 mm; coarse silt (%): 0.050–0.020 mm.

The entries for several parameters are selected from predefined sets of values, stored topic-wise in eight separate lists (e.g. for parent material), following the codes of the national soil database (Costantini, 2007). In the version of the Excel spreadsheet prepared for the Supplementary material of this paper, the fields of the two sheets filled with the information on the profile sites and soil horizons are marked in green, and the fields of the lists containing the retrievable parameter values are marked in red. 2.3. Soil description and laboratory analysis

The data were organized in two spreadsheets. The first one provides relevant information on a site where a soil profile with at least one horizon having a loess component had been described. The documented site-specific parameters include the region, location, elevation, geomorphogenetic position (e.g. valley floor, uvala), relief form (e.g. plain, slope), substratum, soil classification according to WRB (IUSS Working Group WRB, 2006) and the reference to the source of the data, including the new data presented in this study for the first time. The second sheet contains the descriptions of the loess-containing soil horizons and their analytical data.

The new soil datasets that were acquired in the course of the present study were obtained by use of field and laboratory procedures in accordance with Schoeneberger et al. (1998) and Costantini (2007). Routine analysis of the air-dried fraction 10 μm) and fine materials (c/f ratio) supported the particle size analysis. This ratio differentiated the sand and coarse-medium silt grains from the groundmass (clay and fine silt, which were recognizable under the polarizing microscope not as grains but only as micromass). The c/f ratio was 1:1 or 1:2 in the loess-derived soils in Tuscany and very variable, ranging from 1.5:1 to 10:1, in the sandy loess of southern Sicily. The degree of pedality was higher in the buried soil horizons, whereas the loess-derived horizons often showed only weak ped sepa

8


UN CO RR EC TE D

PR OO F


Fig. 4. Representative soil developed in sandy loess at the Ionian coast of southern Italy.

in the groundmass. In one case (TUS1), the 2Btb horizon showed large, dense, incomplete infillings of silty material from the overlying loess cover, probably due to bioturbation (Fig. 8c, d). The soil mass was dominated by silt and characterized by numerous channels and non-interconnected, isolated voids (Fig. 8a, b). In all soils studied in Tuscany and Sicily, the coarse fraction consisted mainly of quartz, quartzite, quartz-arenite and siltstone, with rare occurrence of K-feldspar (microcline), plagioclase (andesine) and muscovite. No evidence of volcanic ash was found. The weathering degree of the grains was generally weak to moderate in the horizons developed in loess, and moderate to strong in the underlying horizons. The surface characteristics of quartz grains in the 100–500 μm fraction, observed under the SEM, showed two different associations of traces of mechanical impacts. The first one included angular and subangular grains. Their surfaces were often very smooth and characterized by conchoidal fractures, straight or curved grooves and V-shaped holes (Fig. 9), with only a few dissolution pits and little silica precipitation. This combination of submicroscopic grain-surface characteristics

Table 5 Thickness (cm) of different loess types. Loess type

Typical loess Reworked loess Sandy loess

Mean

Min

Max

95.7 77.0 135.3

3 10 7

320 296 355

ration (Fig. 8a, b). More advanced pedogenesis of the buried horizons was also demonstrated by greater abundance and higher complexity of clay coatings. In contrast, the soil horizons developed in the loess cover showed either scarce or no clay coatings (Fig. 8). The loess cover often included pedorelics (Fig. 8c, d), showing the same fabric and characteristics as observed in the associated buried soils. These pedorelics indicated an incorporation of material of the buried soils through colluvial processes, either contemporaneous or subsequent to the aeolian deposition. The pedorelics were often rounded and were randomly dispersed

9



3.4. Available numerical ages constraining the timing of loess deposition in Italy

PR OO F

Information on the timing of loess accumulation in Italy is compiled in Figs. 12 and 13, and Tables 2 and 3 (also including the references from which the ages were obtained). Altogether 40 numerical ages are available: 14 ages are from the region north of the Po River (N Italy; Table 2), and 26 ages are from sites south of the Po River (central Italy; Table 3). Many ages south of the Po River originate from two areas: (1) The Val d'Elsa in Tuscany (Costantini et al., 2009) and (2) the system of fluvial terraces linking the Po basin to the northern slope of the Apennines, between Reggio Emilia and Parma (Fig. 1, site 7). The latter area is underlain by active thrust fronts of the Apennine system and is thus strongly influenced by tectonic dynamics (Michetti et al., 2012). The 1⁠ 4C dates recently published by Frigerio et al. (2017) have been obtained from similar terraces, at the NW extremity of the system (Fig. 1, site 1). This compilation shows that the available ages are not evenly distributed, neither over time nor vertically through the published sections; this is clearly due to sampling decisions by the researchers carrying out the different studies (Tables 2 and 3) and/or to some loess beds that were considered unsuitable for dating. To reduce this arbitrary component, we reviewed the papers for the stratigraphy of each exposure reported therein. We recognized the individual loess units and tried to identify major phases of their deposition. In this way, we tried to avoid repeated counting of a single loess unit that was reported in several papers.

Fig. 5. Particle size distribution of typical (solid line), reworked (dotted line) and sandy loess (dashed line) in Italy. Wilks' Lambda = 0.0987, F = 17.46 p value 2 mm) dense incomplete infilling of soil material derived from loess (in the right part of the photo) in a red Luvisol developed on karstified limestone (on the left, TUS1 2Bt1/Bw2). White arrows point to rounded red pedorelics that are embedded within the infilling. These pedorelics indicate transport and mixing of soil material. Under NX, the voids did not show a clear black color and sometimes they show little greyish fragments. These are artifacts due to the soil thin section preparation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

the properties of grains subjected to long-distance transport from the Sahara, as described by Middleton et al. (2001). They reported that Sahara dust storm events produce quartz grains with scarce surface modifications, suggesting a mechanism of long-distance particle transport that does not markedly affect the original shape.

this way, fresh material is regularly replenished, and Mediterranean riverbeds represent efficient dust sources, especially in lowlands (Fig. 14). 4.2.3.2. Bare slopes In many hilly areas of the Mediterranean region additional suitable local dust sources occur. Especially where erosion into silty unconsolidated material produces bare surfaces, these surfaces may act as local sources for dust release or remobilization (Dugmore and Buckland, 1991; Costantini et al., 2009) (Fig. 15). This mechanism can also be active in mountains and may contribute to silty loess also at relatively high elevations in the Apennines (Mileti et al., 2013). Thin sections of soil horizons developed in loess deposits of central Tuscany showed high percentages of silt and fine sand, but also large proportions of fine groundmass (Table 7). All studied soil horizons derived from loess in this area had clay and fine silt contents summing up to about 30–40% (see Excel spreadsheet in Supplementary material). These percentages were higher than usual for loess in Italy (Ferraro, 2009). They were probably due to reworking processes, such as colluvial deposition and mixing by soil fauna and human action. Bioturbation was also indicated by the pedorelics and infillings in the buried horizons (Mirabella et al., 1992). The quartz sand grains of the loess of Tuscany, observed under SEM, included only few rounded grains that showed surface characteristics indicative of saltation. Their proportion was however still higher than in the buried non-aeolian horizons (Fig. 10a). Thus, a relatively increased proportion of rounded quartz sand grains (compared to an underlying soil horizon) may be indicative for aeolian contribution to soils, even if the absolute abundance of such grains is low.

4.2.3. Silty loess not related to glacial environments Typical loess related to glacial environments and sandy loess both showed clear spatial distribution patterns, which allowed for linking these deposits to potential source areas and interpreting the processes by which they formed. In contrast, silty loess not related to glacial environments showed no clear distribution patterns; its occurrence has been reported from all over Italy (Fig. 1). The identification of sources and reconstruction of formation processes are more challenging in this case, as this type of loess often deposited only in thin layers that subsequently underwent complete alteration by reworking, pedogenesis and/or mixing with other materials, such as residual clay where loess accumulated in karst landscapes (Fig. 3). Moreover, it must be taken into account that also silty non-aeolian sediments are widespread in Italy (http://www.isprambiente.gov. it/Media/carg/). Therefore, it is impossible to differentiate loess only based on particle size distribution. Potential dust sources for this type of loess include (1) dry riverbeds, (2) bare slopes and (3) Sahara. 4.2.3.1. Dry riverbeds Dry riverbeds represent potential areas of deflation, as the discharge of Mediterranean rivers is highly variable (Struglia et al., 2004). In summer, many riverbeds fall partially or completely dry. As river discharge wanes, transport capacity goes down and finer material settles within the riverbeds (Struglia et al., 2004). In

13


UN CO RR EC TE D

PR OO F


Fig. 9. Surface morphology of quartz sand grains from illuvial horizons of soils in Sicily, developed in sediments that are assumed to be of fluvial and marine origin. a) Subangular grain with smooth edges and evident V-shaped cavities. b) Close-up of V-shaped cavities (1b) of the previous photo. c) Linear grooves (1c). d) Conchoidal fractures (1d).

4.2.3.3. Saharan dust Saharan dust is well known to contribute to Mediterranean soils (Yaalon and Ganor, 1973; Yaalon, 1987) and is occasionally carried even to central Europe (Varga, 2016). Compared to other types of dust originating from regional or local sources, the long-distance transport of Saharan dust takes place at higher altitude in the atmosphere. It is thus independent from local topography. Therefore, the contribution of Saharan dust to the total dust deposition may show a relative increase on mountains, as reported by Giraudi (2015).

al. (2013) and Giraudi (2015) was considered as a single unit, in accordance with the interpretation by the authors. The set of available ages (Figs. 12, 13) shows that all dated loess deposits in Italy are of Late Pleistocene to Holocene age, suggesting that noticeable loess accumulation started in MIS 4. Among the available data, MIS 3 ages outnumber MIS 2 ages. This is in contrast with findings from central European sites, e.g. Nussloch (Antoine et al., 2009) and Datthausen (Sauer et al., 2016), where loess deposition during MIS 2 usually exceeded that of MIS 4 and MIS 3. This difference may partially be explained by a lack in numeric dating at some sites; e.g., at Monte Netto (Zerboni et al., 2015) a thick deposit of older loess, most likely from MIS 4, could not be dated. Furthermore, at several sites in northern (Accorsi et al., 1990; Ferraro, 2009) and central-southern Italy (Giraudi, 2015; Boretto et al., 2017) glacial, glaciofluvial or fluvial processes make the recognition of loess deposition during either MIS 4 or MIS 2 difficult. In addition to the reported loess deposition from MIS 4 to MIS 2, some of the ages point to the Holocene as a likely period of loess deposition, at least in the periods ca. 7.4–5.3 ka BP in the area north of the Po River (Table 2) and ca. 5.7–3.3 ka BP south of the Po River (Table 3). This is another marked contrast with loess in central Europe, supporting earlier observation of Holocene loess deposition by Costantini et al. (2009). However, the limited number of available ages of Holocene loess, and the risk of post-depositional disturbance affecting luminescence ages, especially under human impact (Zerboni et al., 2015), stress the need for additional attempts to date Holocene loess. Thus, the dataset compiled in this study suggests that loess accumulation in Italy was not confined to full glacial conditions, but took place, either continuously or repeatedly, since MIS 4. Evidence of loess accumulation in Italy that matches neither MIS 4 nor MIS 2 implies that factors other than glacial-periglacial environmental conditions were responsible for the genesis of some of the loess units.

4.3. Timing of loess deposition

4.3.1. General distribution of available ages of loess in Italy and differences compared to central Europe Some of the chronological data compiled in this study refer to loess units that were studied by Cremaschi (1987), Giraudi et al. (2013), and Giraudi (2015). The stratigraphic position of the main loess body in the Apennine piedmont was established by Cremaschi (1987), who termed it Ghiardo loess. The new datings obtained from the soil profiles P2, P6, P8 and P13 (Table 3) confirmed that in the Parma – Reggio Emilia area (Fig. 1, site 7) the Ghiardo loess is homogeneous in thickness and in age. The base of the unit was dated several times (Martini et al., 2001; Cremaschi et al., 2015) and is represented in this study by the samples P8-2, P2-2 and P6-1 (Table 3). As all ages obtained for the base of the Ghiardo loess are rather consistent, the base of the Ghiardo loess is presented in Figs. 12 and 13 only as one single data point. The newly obtained ages (P13-1, P2-1 and P8-1 in Table 3) indicate that the accretion of the Ghiardo loess took place from ca. 72 to 38 ka BP. The characteristics of the soil profiles, in agreement with the soil data in Cremaschi (1987), suggest one continuous cycle of pedogenesis, taking place contemporaneously to a more or less constant loess deposition at low rates. Consequently, the Ghiardo loess can be regarded as a single stratigraphic unit. Similarly, the loess body traced by Giraudi et

14


UN CO RR EC TE D

PR OO F


Fig. 10. Abundance (%) of the most common quartz-grain microfeatures in soil horizons developed in typical (TL) and reworked loess (RL) in Tuscany (Tus); buried horizons developed in colluvial deposits in Tuscany (COL_Tus); soil horizons developed in sandy loess (SL) and marine deposits (MAR) in Sicily (Sic).

mentary processes (Zerboni et al., 2014). The numeric ages that are available so far suggest that loess deposition within MIS 3 may have increased after ca. 45 ka BP, that is, after the onset of the series of interstadial Heinrich events (H5 to H3; Hemming, 2004). A possible explanation for such trend could be that rather stable climatic conditions, prevailing before 45 ka BP, allowed for the development of a dense vegetation cover, whereas the rapidly changing climatic conditions of the later part of MIS 3 led to instability in the vegetation cover and thus to an increase in potential dust sources. However, the suggested trend of increased loess accumulation during MIS 3 still needs to be confirmed by additional numeric dating.

4.3.2. Loess deposition during MIS 4 and MIS 3 Across the Po Plain, the loess covers glacial, glaciofluvial and fluvial sediment facies that are typical of broad alluvial fans and braided river plains, representing records of large sediment loads during phases of intense geomorphic activity. Alluvial fans and braided river plains acted as effective local dust sources, as confirmed by a strong local component evidenced by loess provenance studies (Cremaschi, 1987; Zerboni et al., 2014). On the southern side of the Po Plain, genesis of loess during MIS 4 and MIS 3 may have been driven by the tectonic activity of the Apennine orogeny (Boccaletti et al., 2004). Notably, loess deposition at the north-western edge of this loess region has been reported only for the time since MIS 3, after a major series of seismic events (Frigerio et al., 2017). North of the Po River, geomorphic processes were less dynamic during MIS 3, compared to MIS 4 and 2. Thus, loess deposits were not obscured by other geomorphic and sedi

4.3.3. Loess deposition during MIS 2 Published MIS 2 loess deposition ages are rather rare for northern Italy (Table 2). This may be due to predominating glacial and glacioflu 15


UN CO RR EC TE D

PR OO F


Fig. 11. Surface morphology of quartz sand grains from soil horizons developed in sediments in Sicily that are assumed to be of aeolian origin. a) Well-rounded grain with dish-shaped concavity (1a). b) Pitted microrelief (1b) and upturned silica plates (2b). c) and d) Other rounded grains with dish-shaped concavities (1c).

Fig. 12. OSL and TL ages of sediments, 1⁠ 4C ages of carbonate concretions in buried soils, and 1⁠ 4C ages of organic material from soil samples considered in the present study. Numeric ages with error bars for sites north of the Po River (a) and south of the Po River (b).

vial processes, leading to erosion or incorporation of the loess into their respective deposits. In addition, the upper part of the loess profile is often influenced by anthropogenic disturbance. Evidence of agricultural impact since about 8.5 ka BP has been documented (Corti et al., 2013). At that time humans in the central Mediterranean region became able to cultivate the land using fire and primitive cutting tools. The resulting rapid demographic growth induced an intense impact on the environment in terms of deforestation and soil erosion (Corti et al., 2013). Land abandonment during the Bronze Age, probably due to the impairment of soil fertility, has been documented in areas where loess

deposition took place (Cremaschi et al., 2006). Cremaschi et al. (2015) assumed human disturbance as the most probable reason for the absence of MIS 2 loess in the Ghiardo area. On the other hand, if anthropogenic disturbance led to removal of MIS 2 loess, this should also hold true for Holocene loess. However, at several sites (Costantini et al., 2009, unpublished site P10), Holocene loess directly overlies older deposits (MIS 4 loess or other deposits). At these sites, Holocene loess was preserved, even under agricultural land-use and in absence of MIS 2 deposits. Thus, at these locations either no MIS 2 loess had been pre

16



Italy, which might not have received much loess during MIS 2, because of their greater distance from the source areas and because of orographic barriers that limited dust influx from southern directions.

PR OO F

4.3.4. Holocene loess deposition Three potential sources of loess not related to glacial environments, namely, dry riverbeds, bare slopes, and Saharan dust (cf. Section 4.3.2), are not restricted to glacial periods. Dust release from these sources has been most likely strongly variable over time, including the Holocene period, under the influence of both climatic fluctuations and increasing human impact. Dry periods, involving temporary forest decline, like the one that occurred in central and southern Italy around 4.2 ka BP and lasted for several centuries (Magri, 1999; Magri and Sadori, 1999; Caroli and Caldara, 2007), may have triggered increased water and wind erosion, resulting in dust release from local sources. Pollen of African plant species that were deposited in central Italy during the arid events of 8.2 ka BP and 4.2 ka BP (Magri, 2009) suggest that air masses from Africa often reached central Italy. Thus, dust influx from dry riverbeds, locally outcropping silty sediments, and the Sahara Desert, may have been enhanced during Holocene dry periods. Similarly, enhanced dust release can be assumed for times of increased human impact, leading to reduced vegetation cover, e.g., during the early Bronze age (Costantini et al., 2009). Loess not related to glacial environments has lower contents of very fine sand and silt, and larger proportions of clay (Fig. 7), compared to glacial loess. These differences may be explained by various causes, such as differences in the distances from the sources and in the wind regimes, as well as Holocene weathering of the material prior to deflation. Especially chemical weathering was more intensive under Holocene environmental conditions than during the last glacial (Vance et al., 2009). Enhanced chemical weathering during the Holocene should result in finer texture of the material that is then blown out in a period of aridification or enhanced human impact. Nahon and Trompette (1982) and Pye (1984) examined potential sources of silt. They pointed out that weathering should lead to much more effective silt production than glacial grinding. Furthermore, they suggested that the main role of glacial conditions in generating dust is the related geomorphological instability, which facilitates silt mobilization from various land surfaces and materials. Rivers then spread silt-size material across floodplains, where it can be easily taken up by wind. Smalley and Leach (1978), Zárate (2003) and Smalley et al. (2009) proposed a

Fig. 13. Loess units at active deposition sites.

UN CO RR EC TE D

viously deposited, or the MIS 2 loess was completely consumed as a local component in the formation of Holocene loess. Giraudi (2015) and Giraudi et al. (2013) obtained MIS 2 ages for loess of the central and southern Apennine region. They referred these MIS 2 ages to one single loess unit (Giraudi et al., 2013), which was considered to be mainly derived from Saharan dust (Giraudi, 2015). Inspection of the topography at the site where Boretto et al. (2017) dated a MIS 2 loess unit showed that only dust influx from southern to south-western directions could have led to loess accretion at that site. Therefore, it is most likely that MIS 2 loess in the central and southern Apennines includes major contributions of dust coming from the Sahara Desert and/or the exposed continental shelf. The loess deposition would have then been primarily controlled by arid climatic conditions in the Sahara, by the frequency of incoming air masses from southern directions and by topography. This hypothesis of enhanced dust influx from southern directions during MIS 2 would also match very well with the contemporary accumulation of sandy loess in southern Italy, released from the southward adjacent shelf that fell dry during the LGM (Sauer et al., 2010). In this case, the shelf as a major source of sandy loess is evidenced by the presence of marine micro-fossils (Brückner, 1980; Sauer et al., 2010). A prominent role of southern sources for loess deposition in Italy during MIS 2 might also explain the rare occurrence of MIS 2 loess at the sites investigated in northern

Fig. 14. A “Fiumara” riverbed in Calabria, southern Italy. Large riverbeds acting as sources of loess are common in Italian lowlands and were probably much more widespread during phases of climatic and tectonic instability. (Photo by E.A.C. Costantini.) 17


PR OO F


UN CO RR EC TE D

Fig. 15. Bare slopes on silty marine sediments in Basilicata, southern Italy. Such slopes, similar to dry riverbeds (Fig. 14), may act as sources of loess, especially in hilly lands. Human impact during early Bronze Age, coupled with climatic and tectonic instability, may be responsible for a sudden increase of such surfaces. (Photo by D. Sauer.)

major role of floodplains as dust sources in areas not related to major ice sheets, while Badura et al. (2013) and Stauch (2015) assumed a similar role of floodplains even for classical loess regions such as Poland and Tibet. The data compiled in our study suggest that geomorphic instability triggered by causes other than glacial conditions, such as tectonics, human impact, and climatic dry spells, can be effective in inducing dust release and loess deposition. Consideration of these factors and processes may contribute to a better definition of the concept of “warm loess”, first expounded by Obruchev (1945).

i) ii) iii) iv) v)

glaciation of the Alps and related glaciofluvial dynamics, sea-level low-stands of the last glacial period (especially LGM), geomorphic instability (e.g., induced by tectonics in northern Italy), climatic variations (Heinrich events, drying out of the Sahara), and soil erosion caused by human activities (such as deforestation, slash-and-burn cultivation).

We conclude that, depending on the location of a loess deposit and the time of its deposition, various loess sources need to be considered, namely, (1) glaciofluvial deposits, (2) fluvial deposits (dry riverbeds), and (3) exposed shelfs, these three sources being particularly diffused in lowlands and along the coasts, (4) local silty sediments from bare slopes in hilly lands, where soils have been strongly eroded, (5) Saharan dust, which is relatively more important for the loess found in the mountains of central and southern Apennines, than in other regions of Italy. Loess deposition has occurred over the last ca. 70 ka, probably at low rates, and has intensified in certain periods. However, more research, including numeric dating, is needed to allow for a reliable discrimination of periods of enhanced loess accumulation. Our study suggests that loess accumulation might have not been confined to the Pleistocene, but may also have taken place during the Holocene, during periods of natural or anthropogenic disturbance. This issue needs to be investigated in more detail in further studies. In particular, additional numeric dating is needed, whereby utmost care is required when sampling loess deposits of possible Holocene age for dating, because of the risk of disturbance caused by pedogenesis and human activities. However, if confirmed, the existence of Holocene loess would imply important consequences for modelling the rates of pedogenic processes, the impact of soil erosion, and the risk of desertification in the Mediterranean (Lavee et al., 1998). In addition, further research on the relevance and spatial distribution of Saharan dust in Italy, which so far seems to be limited to the central-southern regions, is needed. Also its age range needs to be further constrained. At present, there is no dated loess older than MIS 2 in southern Italy.

5. Conclusions

Loess deposits in Italy are usually much thinner than those in the loess regions of central Europe and Asia (see e.g. Bronger and Heinkele, 1989). However, the present study shows that loess is present in most regions of Italy, from the northern temperate regions to Mediterranean Italy, and across various altitudes. This finding reveals a so far largely underestimated relevance of loess as a parent material of soils of Italy. This outcome of our study is important not only for basic research on loess and palaeoenvironmental reconstruction, but it is also relevant for environmental and ecological studies, as well as for soil and environmental protection. Since loess deposits and loess admixtures increase soil fertility, loess-borne soils should receive a particularly high level of research and protection. The identification of loess in Italy proved rather difficult, as the loess is usually pedogenized and often reworked and mixed with other materials, like colluvium and residuum of rock weathering, and because of the wide distribution of silty fluvial, lacustrine and marine sediments. Karst depressions are favourable for detecting aeolian sediments, although the loess trapped there tends to be mixed with limestone weathering residuum, resulting in loess soils that are more reddish and clayey than loess soils formed in other geomorphological positions. Glacial loess usually has a somewhat coarser texture, thus pointing mainly to a closer proximity of the source area. Based on the identified characteristics and distribution of loess in Italy, we suggest that the main driving forces triggering the release and aeolian dispersal of silt and fine sand in Italy are:

18



In conclusion, this study shows an unexpected ubiquity of loess all over Italy and stresses the necessity for further investigation of loess in Italy, which has implications far beyond basic research. Yet it also points to the need to extend the assessment of the spatial and temporal distribution of loess also beyond Italy, in order to fully understand the processes behind the genesis of loess and its role in Mediterranean soils.

PR OO F

Borsato, A., 2009. Depositi loessici in Trentino: caratteristiche morfologiche, tessiturali, mineralogiche e pedologiche. Studi Trentini Scienze Naturali 85, 51–59. Brahy, V., Deckers, J., Delvaux, B., 2000. Estimation of soil weathering stage and acid neutralizing capacity in a toposequence Luvisol–Cambisol on loess under deciduous forest in Belgium. Eur. J. Soil Sci. 51 (1), 1–13. Bronger, A., Heinkele, T., 1989. Micromorphology and genesis of paleosols in the Luochuan loess section, China: pedostratigraphic and environmental implications. Geoderma 45 (2), 123–143. Brückner, H., 1980. Marine Terrassen in Süditalien. Eine quartärmorphologische Studie über das Küstentiefland von Metapont. Düsseldorf Geographische Schriften 14, 1–235. Busacca, A., Cremaschi, M., 1998. The role of time versus climate in the formation of deep soils of the Apennine fringe of the Po Valley, Italy. Quat. Int. 51-52, 95–108. Carnicelli, S., Costantini, E.A.C., 2013. Time as a soil forming factor and the age of Italian soils. In: Costantini, E.A.C., Dazzi, C. (Eds.), Soils of Italy. Springer, Dordrecht Heidelberg New York London, pp. 93–104. Carnicelli, S., Benvenuti, M., Andreucci, S., Ciampalini, R., 2015. Late Pleistocene relic Ultisols and Alfisols in an alluvial fan complex in coastal Tuscany. Quat. Int. 376, 163–172. Caroli, I., Caldara, M., 2007. Vegetation history of Lago Battaglia (eastern Gargano coast, Apulia, Italy) during the middle-late Holocene. Veg. Hist. Archaeobotany 16, 317–327. Corti, G., Cocco, S., Brecciaroli, G., Agnelli, A., Seddaiu, G., 2013. Italian soil management from antiquity to nowadays. In: The Soils of Italy. Springer Netherlands, pp. 247–293. Costantini, E.A.C. (Ed.), 2007. Linee guida dei metodi di rilevamento e informatizzazione dei dati pedologici. CRA-ABP, Firenze, Italia, p. XV, (280). Costantini, E.A.C., Priori, S., Urban, B., Hilgers, A., Sauer, D., Protano, G., Trombino, L., Hülle, D., Nannoni, F., 2009. Multidisciplinary characterization of the middle Holocene eolian deposits of the Elsa River basin (central Italy). Quat. Int. 209, 107–130. Costantini, E.A.C., Fantappié, M., L'Abate, G., 2013. Climate and pedoclimate of Italy. In: Costantini, E.A.C., Dazzi, C. (Eds.), The Soils of Italy. World Soils Book Series Springer, pp. 19–37. Coudé-Gaussen, G., 1990. The loess and loess-like deposits along the sides of the western Mediterranean Sea: genetic and palaeoclimatic significance. Quat. Int. 5, 1–8. Coudé-Gaussen, G., Rognon, P., 1988. The Upper Pleistocene loess of southern Tunisia: a statement. Earth Surf. Process. Landf. 13, 137–151. Cremaschi, M., 1987. Palaeosols and Vetusols in the Central Po Plain (Northern Italy): A Study in Quaternary Geology and Soil Development. UNICOPLI, Milano, Italy, (306 pp). Cremaschi, M., 1990. Stratigraphy and palaeoenvironmental significance of the loess deposits on Susak Island (Dalmatian archipelago). Quat. Int. 5, 97–106. Cremaschi, M., 2004. Late Pleistocene loess. In: Antonioli, F., Vai, G.B., Aceti, A. (Eds.), Lithopalaeoenvironmental Maps of Italy During the Last Two Climatic Extremes. Climex Maps Italy Museo geologico Giovanni Cappellini, Bologna, Italy, pp. 34–37. Cremaschi, M., Van Vliet-Lanöe, B., 1990. Traces of frost activity and ice segregation in Pleistocene loess deposits of northern Italy. Deep seasonal freezing or permafrost?. Quat. Int. 5, 39–48. Cremaschi, M., Fedoroff, N., Guerreschi, A., Huxtable, J., Colombi, N., Castelletti, L., Maspero, A., 1990. Sedimentary and pedological processes in the Upper Pleistocene loess of northern Italy. The Bagaggera sequence. Quat. Int. 5, 23–38. Cremaschi, M., Pizzi, C., Valsecchi, V., 2006. Water management and land use in the terramare and a possible climatic co-factor in their abandonment: the case study of the terramara of poviglio Santa Rosa (northern Italy). Quat. Int. 151, 87–98. Cremaschi, M., Zerboni, A., Nicosia, C., Negrino, F., Rodnight, H., Spötl, C., 2015. Age, soil-forming processes, and archaeology of the loess deposits at the Apennine margin of the Po plain (northern Italy): new insights from the Ghiardo area. Quat. Int. 376, 173–188. Crouvi, O., Amit, R., Enzel, Y., Gillespie, A.R., 2010. Active sand seas and the formation of desert loess. Quat. Sci. Rev. 29 (17), 2087–2098. Culver, S.J., Bull, P.A., Campbell, S., Shakesby, R.A., Whalley, W.B., 1983. Environmental discrimination based on quartz grain surface textures: a statistical investigation. Sedimentology 30, 129–136. Dubar, M., 1979. Les caractères sédimentologiques des terrasses fluviatiles et leur couverture limoneuse en Moyenne Durance. Bulletin Association Francaise pour l'étude du Quaternaire 60 (3), 109–120. Dugmore, A., Buckland, P., 1991. Tephrochronology and late Holocene soil erosion in south Iceland. In: Maizels, J.K., Caseldine, C. (Eds.), Environmental Change in Iceland: Past and Present. Springer Netherlands, pp. 147–159. Dumas, B., 1977. Le Levant Espagnol; la genèse du relief. Université Paris-Sorbonne, (Thesis, 520 pp). FAO, 2006. Guidelines for Soil Description, 4th edition FAO, Rome, Italy. Ferraro, F., 2009. Age, sedimentation, and soil formation in the Val Sorda loess sequence, Northern Italy. Quat. Int. 204, 54–64. Florineth, D., Schlüchter, C., 1998. Reconstructing the Last Glacial Maximum (LGM) ice surface geometry and flowlines in the Central Swiss Alps. Eclogae Geol. Helv. 91, 391–407. Forno, M.G., 1990. Aeolian and reworked loess in the Turin hills (northwestern Italy). Quat. Int. 5, 81–87. Frezzotti, M., Giraudi, C., 1990. Late Glacial and Holocene aeolian deposits and geomorphological features near Roccaraso (Abruzzo – central Italy). Quat. Int. 5, 89–95. Frigerio, C., Bonadeo, L., Zerboni, A., Livio, F., Ferrario, M.F., Fioraso, G., Irace, A., Brunamonte, F., Michetti, A.M., 2017. First evidence for Late Pleistocene to Holocene earthquake surface faulting in the Eastern Monferrato Arc (Northern Italy): geology,

Uncited references Carnicelli and Costantini, 2013 Carnicelli et al., 2015 Zander et al., 2006 Acknowledgements

UN CO RR EC TE D

The authors thank Prof. Fabio Terribile, University of Naples, for the collaboration in sharing soil samples. We also thank P.S. Toms, who carried out all of the previously unpublished luminescence dating in 2004 in the Geochronology Laboratories of University of Gloucestershire and R. Barbetti, who contributed to the preparation of the loess database, attached as Excel spreadsheet in the Supplementary materials. The collaborative work was supported by the INQUA project “AEOMED: Loess and aeolian additions to current surface soils and palaeosols in Mediterranean climate”. We also thank two anonymous reviewers for their very thorough reviews that helped us to considerably improve a previous version of the manuscript. Appendix A. Supplementary data

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.catena.2018.02.002. References

Accorsi, C.A., Aitken, M., Cremaschi, M., Ferraris, M., McElroy, C., Questiaux, D., Van Vliet-Lanöe, B., 1990. The loess deposits of the Rivoli moraine system. In: Cremaschi, M. (Ed.), The Loess in Northern and Central Italy: A Loess Basin Between the Alps and the Mediterranean Region. Quaderni di GeodinamicaAlpina e Quaternaria 1, pp. 21–39. Alekseeva, V.A., 2005. Micromorphology of quartz grain surface as indicator of glacial sedimentation conditions: evidence from the Protva river basin. Lithol. Miner. Resour. 40 (5), 420–428. Amit, R., Zerboni, A., 2013. Report on the INQUA-AEOMED field-trip workshop ‘Reconsidering Loess in Northern Italy’ (Po Plain, 1–3 July 2013). Alp. Mediterr. Quat. 26 (2), XI–XV. Andrenelli, M.C., Fiori, V., Pellegrini, S., 2013. Soil particle-size analysis up to 250 μm by X-ray granulometer: device set-up and regressions for data conversion into pipette-equivalent values. Geoderma 192, 380–393. Andreucci, S., Bateman, M.D., Zucca, C., Kapur, S., Akşit, , Dunajko, A., Pascucci, V., 2012. Evidence of Saharan dust in upper Pleistocene reworked palaeosols of North-west Sardinia, Italy: palaeoenvironmental implications. Sedimentology 59 (3), 917–938. Antoine, P., Rousseau, D.D., Moine, O., Kunesch, S., Hatté, C., Lang, A., … Zöller, L., 2009. Rapid and cyclic aeolian deposition during the Last Glacial in European loess: a high-resolution record from Nussloch, Germany. Quat. Sci. Rev. 28 (25), 2955–2973. Avila, A., Alarcon, M., Queralt, I., 1998. The chemical composition of dust transported in red rains—its contribution to the biogeochemical cycle of a holm oak forest in Catalonia (Spain). Atmos. Environ. 32 (2), 179–191. Badura, J., Jary, Z., Smalley, I., 2013. Sources of loess deposits in Poland and parts of Central Europe: the lost Big River. Quat. Int. 296, 15–22. Ben-Israel, M., Enzel, Y., Amit, R., Erel, Y., 2015. Provenance of the various grain-size fractions in the Negev loess and potential changes in major dust sources to the Eastern Mediterranean. Quat. Res. 83 (1), 105–115. Bertran, P., Liard, M., Sitzia, L., Tissoux, H., 2016. A map of Pleistocene aeolian deposits in Western Europe, with special emphasis on France. J. Quat. Sci. 31 (8), 844–856. Boccaletti, M., Bonini, M., Corti, G., Gasperini, P., Martelli, L., Piccardi, L., Tanini, C., Vannucci, G., 2004. Seismotectonic Map of the Emilia-Romagna Region, 1:250,000. Regione Emilia-Romagna-CNR, SELCA, Firenze, Italy. Boretto, G., Zanchetta, G., Ciulli, L., Bini, M., Fallick, A.E., Lezzerini, M., Colonese, A.C., Zembo, I., Trombino, L., Regattieri, E., Sarti, G., 2017. The loess deposits of Buca dei Corvi section (Central Italy): revisited. Catena 151, 225–237.

19


Catena xxx (2018) xxx-xxx Peresani, M., Cremaschi, M., Ferraro, F., Falguères, C., Bahain, J.-J., Gruppioni, G., Sibilia, E., Quarta, G., Calcagnile, L., Dolo, J.-M., 2008. Age of the final Middle Palaeolithic and Uluzzian levels at Fumane Cave, Northern Italy, using 1⁠ 4C, ESR, 2⁠ 34U/2⁠ 30Th and thermoluminescence methods. J. Archaeol. Sci. 35, 2986–2996. Priori, S., Costantini, E.A., Capezzuoli, E., Protano, G., Hilgers, A., Sauer, D., Sandrelli, F., 2008. Pedostratigraphy of Terra Rossa and Quaternary geological evolution of a lacustrine limestone plateau in central Italy. J. Plant Nutr. Soil Sci. 171 (4), 509–523. Pye, K., 1984. Loess. Prog. Phys. Geogr. 8 (2), 176–217. Pye, K., 1992. Aeolian dust transport and deposition over Crete and adjacent parts of the Mediterranean Sea. Earth Surf. Process. Landf. 17 (3), 271–288. Rellini, I., Trombino, L., Firpo, M., Rossi, P.M., 2009. Extending westward the loess basin between the Alps and the Mediterranean region: micromorphological and mineralogical evidences from the northern scope of the Ligurian Alps, Northern Italy. Geogr. Fis. Din. Quat. 32, 103–116. Sauer, D., Wagner, S., Brückner, H., Scarciglia, F., Mastronuzzi, G., Stahr, K., 2010. Soil development on marine terraces near Metaponto (Gulf of Taranto, southern Italy). Quat. Int. 222 (1–2), 48–63. Sauer, D., Amit, R., Carnicelli, S., Costantini, E.A.C., 2015. Rates of soil forming processes and the role of aeolian influx. Quat. Int. 376, 1–4. Sauer, D., Kadereit, A., Kühn, P., Kösel, M., Miller, C.E., Shinonaga, T., … Stahr, K., 2016. The loess-palaeosol sequence of Datthausen, SW Germany: characteristics, chronology, and implications for the use of the Lohne Soil as a marker soil. Catena 146, 10–29. Schoeneberger, P.J., Wysocki, D.A., Benham, E.C., Broderson, W.D., 1998. Field Book for Describing and Sampling Soils. Natural Resources Conservation Service, USDA National Soil Survey Center, Lincoln, NE. Smalley, I.J., Leach, A.J., 1978. The origin and distribution of the loess in the Danube basin and associated regions of East-Central Europe - a review. Sediment. Geol. 21 (1), 1–26. Smalley, I., O'Hara-Dhand, K., Wint, J., Machalett, B., Jary, Z., Jefferson, I., 2009. Rivers and loess: the significance of long river transportation in the complex event-sequence approach to loess deposit formation. Quat. Int. 198, 7–18. Smalley, I.J., Markovic, S.B., Svircev, Z., 2011. Loess is [almost totally formed by] the accumulation of dust. Quat. Int. 240, 4–11. Sprafke, T., Obreht, I., 2016. Loess: rock, sediment or soil - what is missing for its definition?. Quat. Int. 399, 198–207. Stauch, G., 2015. Geomorphological and palaeoclimate dynamics recorded by the formation of aeolian archives on the Tibetan Plateau. Earth Sci. Rev. 150, 393–408. Stoops, G., 2003. Guidelines for Analysis and Description of Soil and Regolith Thin Sections. Soil Science Society of America Inc., U.S.A.. Struglia, M.V., Mariotti, A., Filograsso, A., 2004. River discharge into the Mediterranean Sea: climatology and aspects of the observed variability. J. Clim. 17 (24), 4740–4751. Torrent, J., Schwertmann, U., Schulze, D.G., 1980. Iron oxide mineralogy of some soils of two river terrace sequences in Spain. Geoderma 23 (3), 191–208. Vance, D., Teagle, D.A., Foster, G.L., 2009. Variable Quaternary chemical weathering fluxes and imbalances in marine geochemical budgets. Nature 458 (7237), 493–496. Varga, G., Cserháti, C., Kovács, J., Szalai, Z., 2016. Saharan dust deposition in the Carpathian Basin and its possible effects on interglacial soil formation. Aeolian Res. 22, 1–12. Vos, K., Vandenberghe, N., Elsen, J., 2014. Surface textural analysis of quartz grains by scanning electron microscopy (SEM): from sample preparation to environmental interpretation. Earth Sci. Rev. 128, 93–104. Wacha, L., Pavlakovic, S.M., Novothny, A., Crnjakovic, M., Frechen, M., 2011. Luminescence dating of Upper Pleistocene Loess from the Island of Susak in Croatia. Quat. Int. 234, 50–61. Wagner, S., Costantini, E.A.C., Sauer, D., Stahr, K., 2007. Soil genesis in a marine terrace sequence of Sicily, Italy. Revista Mexicana de Ciencias Geológicas 2, 247–260. Wright, J.S., 2001. “Desert” loess versus “glacial” loess: quartz silt formation, source areas and sediment pathways in the formation of loess deposits. Geomorphology 36 (3), 231–256. Yaalon, D.H., 1987. Saharan dust and desert loess: effect on surrounding soils. J. Afr. Earth Sci. (1983) 6 (4), 569–571. Yaalon, D.H., Ganor, E., 1973. The influence of dust on soils during the Quaternary. Soil Sci. 116 (3), 146–155. Yang, X., Liu, T., Yuan, B., 2009. The Loess Plateau of China: aeolian sedimentation and fluvial erosion, both with superlative rates. In: Geomorphological Landscapes of the World. Springer Netherlands, pp. 275–282. Zander, A.-M., Fülling, A., Brückner, H., Mastronuzzi, G., 2006. OSL dating of upper Pleistocene littoral sediments: a contribution to the chronostratigraphy of raised marine terraces bordering the Gulf of Taranto, South Italy. Geogr. Fis. Din. Quat. 29, 33–50. Zárate, M.A., 2003. Loess of southern South America. Quat. Sci. Rev. 22, 1987–2006. Zerboni, A., Trombino, L., Frigerio, C., Livio, F., Berlusconi, A., Michetti, A.M., Rodnight, H., Spötl, C., 2015. The loess-paleosol sequence at Monte Netto: a record of climate change in the Upper Pleistocene of the central Po Plain, northern Italy. J. Soils Sediments 15, 1329–1350.

UN CO RR EC TE D

PR OO F

pedostratigraphy and structural study of the Pecetto di Valenza site. Quat. Int. https://doi. org/10.1016/j.quaint.2016.12.022. Fuchs, F., Semmel, A., 1974. Pleistozäne kaltzeitliche Ablagerungen in der Sila und Basilicata (Süd-Italien). Catena 1, 387–400. Giraudi, C., 2015. The Upper Pleistocene deglaciation on the Apennines (peninsular Italy). Cuadernos de investigación geográfica 41 (2), 337–358. Giraudi, C., Zanchetta, G., Sulpizio, R., 2013. A Late-Pleistocene phase of Saharan dust deposition in the high Apennine mountains (Italy). Alp. Mediterr. Quat. 26 (2), 110–122. Haase, D., Fink, J., Haase, G., Ruske, R., Pécsi, 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. Quat. Sci. Rev. 26, 1301–1312. Helland, P.E., Holmes, M.A., 1997. Surface textural analysis of quartz sand grains from ODP Site 918 off the southeast coast of Greenland suggests glaciation of southern Greenland at 11 Ma. Palaeogeogr. Palaeoclimatol. Palaeoecol. 135 (1), 109–121. Hemming, S.R., 2004. Heinrich events: massive late Pleistocene detritus layers of the North Atlantic and their global climate imprint. Rev. Geophys. 42 (1). IUSS Working Group WRB, 2006. World Reference Base for Soil Resources. In: World Soil Resources Reports, vol. 103, (Rome, 132 pp.). IUSS Working Group WRB, 2014. World Reference Base for Soil Resources. In: World Soil Resources Reports, vol. 106, (Rome, 181 pp.). Krinsley, D.H., Doornkamp, J.C., 1973. Atlas of Quartz Sand Surface Textures. Cambridge University Press, Cambridge, UK, (91 pp). Krinsley, D.H., Friend, P.F., Klimentidis, R., 1976. Eolian transport textures on the surfaces of sand grains of Early Triassic age. Geol. Soc. Am. Bull. 87, 130–132. Lambeck, K., Antonioli, F., Anzidei, M., Ferranti, L., Leoni, G., Scicchitano, G., Silenzi, S., 2011. Sea level change along the Italian coast during the Holocene and projections for the future. Quat. Int. 232 (1), 250–257. Lavee, H., Imeson, A.C., Sarah, P., 1998. The impact of climate change on geomorphology and desertification along a mediterranean-arid transect. Land Degrad. Dev. 9 (5), 407–422. Lequy, , Conil, S., Turpault, M.P., 2012. Impacts of Aeolian dust deposition on European forest sustainability: a review. For. Ecol. Manag. 267, 240–252. Magri, D., 1999. Late Quaternary vegetation history at Lagaccione near Lago di Bolsena (central Italy). Rev. Palaeobot. Palynol. 106, 171–208. Magri, D., 2009. Advances in Italian palynological studies: late Pleistocene and Holocene records. GFF 129 (4), 337–344. Magri, D., Sadori, L., 1999. Late Pleistocene and Holocene pollen stratigraphy at Lago di Vico (central Italy). Veg. Hist. Achaeobotany 8, 247–260. Manickam, S., Barbaroux, L., 1987. Variations in the surface texture of suspended quartz grains in the Loire River: a SEM study. Sedimentology 34, 495–510. Martini, M., Sibilia, E., Croci, S., Cremaschi, M., 2001. Thermoluminescence (TL) dating of burnt flints: problems, perspectives and some examples of application. J. Cult. Herit. 2, 179–190. Michetti, A.M., Giardina, F., Livio, F., Mueller, K., Serva, L., Sileo, G., Vittori, E., Devoti, R., Riguzzi, F., Carcano, C., Rogledi, S., Bonadeo, L., Brunamonte, F., Fioraso, G., 2012. Active compressional tectonics, Quaternary capable faults, and the seismic landscape of the Po Plain (northern Italy). Ann. Geophys. 55 (5), 969–1001. Middleton, N.J., Betzer, P.R., Bull, P.A., 2001. Long-range transport of ‘giant’ aeolian quartz grains: linkage with discrete sedimentary sources and implications for protective particle transfer. Mar. Geol. 177 (3), 411–417. 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 207, 291–309. MiPAAF, M. D. P. A. A. e Forestali, 1997. Metodi Ufficiali di Analisi Fisica del Suolo. Determinazione della massa volumica apparente. Met. II 1 (1), 4–6. Mirabella, A., Costantini, E.A.C., Carnicelli, S., 1992. Genesis of a polycyclic Terra Rossa (chromic cambisol on rhodic nitisol) at the Poggio del Comune in Central Italy. J. Plant Nutr. Soil Sci. 155 (5), 407–413. Muhs, D.R., 2006. Loess deposits, origins and properties. In: Elias, S.A. (Ed.), Encyclopedia of Quaternary Science. pp. 1405–1418. Murphy, C.P., 1986. Thin Section Preparation of Soils and Sediments. AB Academic Publishers, Berkhamsted, England. Murray, A.S., Wintle, A.G., 2000. Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiat. Meas. 32 (1), 57–73. Murray, A.S., Wintle, A.G., 2003. The single aliquot regenerative dose protocol: potential for improvements in reliability. Radiat. Meas. 37 (4), 377–381. Nahon, D., Trompette, R., 1982. Origin of siltstones: glacial grinding versus weathering. Sedimentology 29, 25–35. Nemecz, E., Pécsi, M., Hartyáni, Z., Horváth, T., 2000. The origin of the silt size quartz grains and minerals in loess. Quat. Int. 68, 199–208. Obruchev, V.A., 1945. Loess types and their origin. Am. J. Sci. 243 (5), 256–262. Pécsi, M., 1991. Loess is not just the accumulation of dust. Quat. Int. 7 (8), 1–21. Pécsi, M., Richter, G., 1996. Löss. Herkunft - Gliederung - Landschaften. Z. Geomorphol. Suppl. 98, (391 pp). Peresani, M., Nicosia, C., 2015. Comparative study of two late Pleistocene sequences with paleosols and aeolian deposits at the southern alpine foreland. Geogr. Fis. Din. Quat. 38 (1), 41–53.

20