Sedimentary characteristics and source of loess in Baranja (eastern Croatia)

pointdepressedΜηχανική

22 Φεβ 2014 (πριν από 3 χρόνια και 7 μήνες)

108 εμφανίσεις

1


Sedimentary characteristics and source of loess in Baranja (eastern
Croatia)

Adriano
Banak
a
*
, Davor
Pavelić
b
,
Marijan
Kovačić
c
,
Oleg Mandic
d

a
Croatian

Geological Survey, Department fo
r Geology, Sachsova 2, HR
-
10000 Zagreb,
Croatia

b
Faculty of Mining
,
Geology

and Petroleum Engineering, University of Zagreb, Pierottijeva 6, HR
-
10000
Zagreb, Croatia


c
Faculty of Science, Department of Mineralogy and Petrology, University of Zagreb, Horvatovac 95, HR
-
10000
Zagreb, Croatia

d
Natural H
istory

Museum

Vienna
,
Department

of
Geolog
y and P
aleontolog
y
, Burgring 7, A
-
1010, Wien, Austria

*Corresponding author. Tel.: +385

1 6160 708; fax:
+385 1 6160 799.

E
-
mail add
ress
:
abanak@hgi
-
cgs.hr

(A.Banak).

Abstract

Loess is
terrestrial

clastic

sediment, composed dominantly of silt
-
sized particles
formed by the accumulation of wind
-
blown dust. It is usually inter
-
bedded with soil horizons
forming loess
-
pal
a
eosol successions (LPS)
. Thickest LPS in Croatia are found in Baranja,
region bounded with two big rivers,
the
Danube and the Drava.
The results of grain
-
size and
modal analysis provide information about source material and wind direction in different time
periods during Pleistocene. Grain
-
size distribution is in good accorda
nce with other loess
localities in the Pannonian Basin. Garnet, epidote and amphibole mineral group are most
abundant heavy minerals in samples of Danube River sediment. Comparing heavy mineral
assemblage

(HMF)

from
southern and northern

LPS with that data, it is obvious that main
source area for loess in Baranja is from Danube flood plain sediments. Main transport
direction was from North or North
-
West. Nevertheless the higher concentration of amphiboles
in
sou
thern and northern LPS (mea
n 26.
3
% in HMF) then in the Danube plain suggests
additional source area. Western
Carpathians with Neogene calc
-
alkaline volcanic rocks is
major source for amphiboles. Alternatively those minerals could be from locally exposed
volcanic and metamorphic rock
s of the southward Slavonian Mts. Mt. Krndija and Mt. Papuk,
which are closest to Baranja of all Slavonian Mts., consist of amphibolites.
In that case,
s
mall
amount of silt

material for Baranja loess would be transported by WSW winds.
Results
obtai
ned from

sedimentological and SEM

analyses show fairly good congruence with results
f
rom other LPS in the Pannonian B
asin, with some differences in mineral composition which
imply diversity
and shifting
of source area for Baranja loess

during Late Pleistocene
.

Ke
y words
:

loess
,
Baran
j
a
,
aeolian,
heavy mineral fraction
,

quartz,
SEM images

1.Introduction



Quaternary sediments are widespread in Croatia.
They cover about 35.
7%
of the
Croatian teritory (
Bognar
, 1976
). Baranja, region in Eastern part of Croatia

(Fig
.

1)

is almost
completely covered w
ith Quaternary sediments (Pikija & Šikić
, 1991).
Special place within
,

2


regarding the origin, holds

loess and it's derivat
iv
es.
Loess is

terrestrial clastic sediment,
composed predominantly of silt
-
size particles, formed by

the accu
mulation of wind
-
blown
dust (Pye
, 1995).


Fig
.

1
.

Map showing
the position of Croatia, Baranja and two

investigated
loess
profiles at
slopes
Bansko brdo
hill.

One of the most specific characteri
stcs of loess is

grain size distribution
of

partic
les. Most
authors (Bognar, 1976;

Nemecz

et al
., 2000; Pecsi, 1990;

Smalley
, 1966b;

Smalley

et al
.,

3


2005;

Wright,

1995, 2001)
agree that typical loess has grain size distribution in
range
20
-
60

μm, which corelates with silt
-
size grains.
Origin of those silt

siz
e particles and mechanism of
it's

transport

is the main question in over one century of loess exploration. Two main thesis
which are opposite
-
pedogenetic and aeolian, describe formation of thick loess deposits. First
one emphasizes diagentic processes
in si
lty

material as beeing crucial in loess formation,
while second one favors aeolian transport of silt
-
size particles.
The aim of this study is

to
determine source area for loess in Baranja by identifying mineral composition of heavy
mineral fraction an
d morphology of quartz grains.
Geochemical and mineralogical bulk loess
sediment

analyses along parts of the Danube in NE and E Romania have

pointed out that the
main
sediment
for loess deposits
was transported by winds

from a NNW/NNE direction,
indicatin
g

forcing by northerly

winds from the Fennoscandian ice sheet

(Buggle et al., 2008)
.
Bokhorst et al. (2011)

suggest a domination of western winds

during the Early and Middle
Pleniglacial in central and eastern

Europe, while the Late Pleniglacial

was domina
ted by NW
or northern winds
.
Additionally this paper will propose a complex, five
-
phase p
rocess
necessary for thick

loess formation.


2. Geological setting

Geology of Baranja

(Fig
.

2)

is
simple at the surface, consisting of
dominantly
Pleistocene and Hol
ocene sediments, with some outcro
ps of Miocene
igneous rocks,

limestones and marls (CGS
, 2009)
. Holocene sediments are aluvial and marsh gravel, sand
and silt. Pleistocene sediments are domina
ntly loess, loess
-
like sediment

a
nd flood plain
sediments.
Only
high ground in Baranja region is Bansko brdo hill (244

m asl), an tectonic
complex horst, situated in northern part of region, stretching 20

km in NE
-
SW direction and
reaching banks of Danube river. Combination o
f active neotectonics (Hećimović
, 1991) and
Danube river erosion exposed b
ig outrocps of Middle and Upper
Pleistocene sediments

(loess
pal
a
eosol sequence
s
-
LPS
)

at surface

(Fig
.

2)
.

The oldest exposed rocks at Bansko Brdo

hill

belong to the Miocene volcano
-
sedimentary complex, and include basalt
-
andesite and pyroclastic rocks comprising volcanic
and tuffaceous breccia
s

and conglomerates

(Fig
.
2)
. K
-
Ar radiometric, whole
-
rock
measurements indicate an early Middle Miocene age
(13
.
8±0,

4 and 14.
5±0,

4 Ma, Pamić &
Pécskay
,
1996).


The basalt
-
andesite intercalates Middle Miocene (Badenian) marine
calcareous sand and marl, indicating the synsedimentary character

of the volcanic activity
(Pamić & Pikija, 1987; Lugović

et al., 1990).

Th
e oldest loess
in Bansko brdo hill is detected
at the Zmajevac locality an
d

infrared optically stimulated luminescence (IRSL)
method
produced
age of 217±22 ka, while the youngest loess has an
IRSL age of 16.7±1.
8 ka
(Galović

et al., 2009).

4



Fig
.

2
.

Geological map of Baranja region

(CGS,
2009
)
with

two loess profiles

at Bansko brdo hill, encircled with
red colour
.


3. Material

and methods

3.1. Field
methods


Field investigation and sampling wher
e carried out

during winter months, because lush
vegetation in spring and summer
,

disables finding and aproach to
loess
profiles
s. Aim was to
include maximu
m

thicknes of sediments. B
ulk samples (8
-
10

kg) were collected from
Zmajevac

(Zma
)
, Kotlina

(Kot
)
, Podolje

(Pod
)

and Branjina

(Br
)

loess outcrops

for
sedimentological analysis.

These four outcrops make two loess profiles: northern and
southern, regarding their position on the slopes of Bansko brdo hill.
Field investigation
included taking samples
,

m
easuring thicknes
s of sediment,

defining litology, GPS positioning
and
photographing.
Diferent
horizons in loess

where recognized and described. Depending on
that
in
-
situ horizon
diversity, sampling was
carried out. From
typical
loess, sampling
frequency was 1.
5

m on southern slopes of Bansko brdo hill. On northern slopes sampling
frequency
in loess
was
arround
0.
5

m.
Total of 3
0

loess
samples where taken.

3.2. Lab methods

Grain
-
size analyses combined wet sieving and the pipette method. Classification of the
grai
n siz
e distribution follows Wentworth

(1922). Mineral abundances (modal analyses) used
the 0.063
-
0.125 mm calcite
-
free fraction. Heavy and light mineral fraction (HMF and LMF)
were separated in bromoform liquid (CHBr
3
,
δ
=2.86 gcm
-
3
) by gravity. Qualitative

and
quantitative analyses of the fractions were based on 300
-
350 grains per sample and were
5


conducted using a po
larizing light microscope (Menge & Maurer
, 1992). The carbonate
(CaCO
3
) content was calculated from the weight difference before and after cold

hydrochloric
acid (5%) treatment
.

Photo
graphs

of quartz

grains

were made

using a

scanning electron

microscope

(SEM)

in the Laboratory

of Geochemistry,

INA
oil

Co
.

Out of 30 loess samples two with previously
extracted light mineral fraction were chosen
for scanning electron microscopy
.

One sample

from the southern

loess

profiles

(Z
ma

1/2)

and the other

from the northern

loess

profile

(
Br

1/12)
.
M
ore than

40 different

grains

were photographed
, and
best
photographs

that show

complete

quartz
grain
s

were sel
ected
.

Photographs were taken on

quartz grains in order to see
the
shape of

grains
and
detailed morphology

of grain surface
.
Microscopy was

made

with
in

magnification

range of

X350

-

X
3500
.

Selected

samples

were

glued

to a carrier

double
-
sided
tape
,
and
then

steamed with

gold

which
thickness is

25

nm.

Thus

prepared samples

were
placed

in a container

and analyzed

using

a scanning electron

microscope

JEOL
, model
JSM
-
6510
LV.

Micro
photo
graphs

were

recorded
.

The
microscope

scans the surface

of the sample

by
very precisely

focused beam of

electrons.

Source of electrons

is a

tungsten

filament

that

generates

the

high voltage

electrons.

The

beam

of electrones

excites

electrons

within the

atoms

of the sample.

Voltage is

15

kV

during
process of scanning
.

Energy

of
electrons from

the beam

is in direct

proportion to

interactively

excited

electrons

from the sample.

Energy

of
electrons

derived

from

samples

is
collected

and
measured

with special

detectors and

with the help of

a microprocessor

creates a

pseudo

image


with

wavelength

of electrons

unique
to an element

that is found

in the sample.

Detection

of
secondary electrons

enabled

images with
topographi
c
contrast
.

In this study, the quartz grain

surface microtexture classification method and implemented

terminology
are

based on studies
o
f Mahaney (1995b) and Strand et al. (2003).


4. Results


Four loess outcrops

were
investigated. Two are located o
n southern slopes of Bansko
b
rdo hill, a and other two are on

northern
, more steep

slopes of Bansko brdo hill.
The
southern

loess profile

make
s

up total of 18

m of loess

(Fig
.

3
)
, wh
ile northern loess profile

make
s

up total of 8.
5

m of loess

(Fig
.

3
)
.

Southern loess profile has four palaeosols
intercalated within five loess horizons

and is defined as typical loess
. Northern
loess profile
has two palaeosols intercalated within three loess horizins

and is defined as 'slope loess' or
'loess derivated coluvium', a terms previously decribed by several authors (Bognar, 1979; Pye
1995)
.

6



Fig
.

3
.

Southern and northern loess profiles

from Bansko brdo hill.

Detailed lithology of each loess horizon of sou
thern profile as

well with chronological
frame from MIS 6


MIS 2,

has been previously described (Banak et al., 2012; Galović et al.,
2009).
Grain
-
si
ze
analysis

indicate

silt as the dominant grain
-
size fraction in all 13 studied
loess samples

from southern profile

(Fig 4)
.

A
verage share

of silt
-
size particles is 88.
11%.
Coarse
-
grained silt is dominant silt fractio
n with average percentage of 41.
38%.
The
laminated unit is
composed of 81% sand, 11% silt and 8% clay
, with a median grain
-
size of
0.
22 mm.

Skewness is fairly constant in a
ll 13 samples, averaging of 0.
79
. Sorting is
dominantly po
or, with an average value 1.
64
.
The average CaCO
3

content
in loess samples
from
the
southern loess profile is 9.32%, with a maximum value of 23.
3%
that
was recorded
in the sample Kot 1/4, and the minimum
value of 2.
9% in a sample Zma 1/1.

7



Fig
.

4
.

Grain size distri
bution
and coefficients
of
southern loess profile from

Bansko brdo hill.

Grain
-
size analysis

indicate
s

silt as the dominant grain
-
size fraction in all
17

studied
loess samples

from northern profile

(Fig
.

5)
.
A
verage share

of silt
-
size particles is 90.
44%.
Coarse
-
grained silt is dominant silt fraction with average percentage of 42%. Skewnes is
constant with one significant peak
towards 0.
5
and average value is 0.
81.
Sorting is
dominantly poor, with an average value
of 1.
49

w
hich are

slightly positive compared
to
southern profil
e
.
The average CaCO
3

content
in loess samples from the northern profile
is
9.
92%,

with a maximum value of 15.
8%
recorded in the samples Pod 1/1 and Br

1/12

and the
minimum value of 5%
recorded
in the sample Br

1
/11
.


8



Fig
.

5
.

Grain size distribution
and coefficients
of
northern loess profile from

Bansko brdo hill.

Modal analysis of the light

(LM
F) and heavy mineral fraction (H
MF) was made on all
samples of loess and sho
ws the following results (Tab
le

1 and

2
).
In the southern loess profile

light mineral fraction is dominated by quartz with an average share of 58.38%, while
in
the
northern
loess profile average percentage of quartz is 60.29%. H
eavy mineral fraction is
dominated b
y amphibole, garnet and epidote wh
ile

chlorite is present in

small percentages
.
Each of the three major groups

(highlited with colour in Tab
le

1 and 2
)

of transparent heavy
minerals in the southern and northern loess profile is represented approximately in the range
of 25 to 30%, and th
e
relations between them

vary
.


It is worth noticing that incerased percentage of garnet in some samples is related with
decreasd percentage of amphibole and vice versa. This ratio is recorded in samples from both
southern (Kot 1/2, Kot 1/9, Zma 1/4) and nor
thern (Br 1/6, Br 1/8
, Br 1/12) loess profiles
(Tab
le

1 and 2
, numbers in red colour).

In all other samples percentage ratio between this two
groups has more narow range.


9



Tab
le

1

Mineral assemblage
s from southern loess profile
.


Note
:
op
=opaque minerals,
do
=dolomite,
bi
=biotite,
ch
=chlorite,
thm
=transparent heavy minerals,
tu
=tourmaline,
zr
=zirqonium,
ru
=rutile,

am
=amphibole,
py
=pyroksene,
ep
=epidote,
ga
=garnet,
cy
=cyanite,
st
=staurolite,
tit
=titanite,
czt
=clinocoisite,
cto
=chloritoide,
csp
=cromespinel,
si
=silimanite,
x
=undetermined,
q
=quartz,
f
=feldspar,
rf
=rock fragments,
mu
=muscovite



Tab
le

2

Mineral assemblages from northern loess profile
.

Note
:
op
=opaque minerals,
do
=dolomite,
bi
=biotite,
ch
=chlorite,
thm
=transparent heavy minerals,
tu
=tourmaline,
zr
=zirqonium,
ru
=rutile,

am
=amphibole,
py
=pyroksene,
ep
=epidote,
ga
=garnet,
cy
=cyanite,
st
=staurolite,
tit
=titanite,
czt
=clinocoisite,
cto
=chloritoide,
csp
=cromespinel,
si
=silimanite,
x
=undetermined,
q
=quartz,
f
=feldspar,
rf
=rock fragments,
mu
=muscovite

4.2. Scanning electron microscope (SEM) images

Electron
scanning microscope

analyzed two samples from

Bansko brdo hill loess. All
scanned

and photographed
gr
ains in both samples are sand
-
sized particles (>

63 um
)
. In both
samples angular grains dominate
. Their share is over 80% of the total number of grains

in the
samples. Smaller percentage of

grains are
very angular and partly rounded. R
ounded and
well
-
rounded grains

are not detected
. Over 70% of the grains in b
oth samples are
of
low

sphericity. On most grains schist
-
like fractures and ''V'' impresses are visible. Schist
-
like
fractures were detected in over 40% of the grains in the samples. The smaller numb
er of
grains have schist
-
like

fracture
s that are nearly t
he lenght of longer axis of grain
, while
the
majority of grains have

schist
-
like
fracture
s

with
size 1/3 or

1/4 of a grain. ''V'' impresses

on
10


the

surfaces are visible in 15% of total number of grains
. The

size of these textures is in range
from 3 μm to 8
μm. They are usually clustered

on smooth, flat surfaces of grain
s
, and
in

small
er

number of grains are prese
nt in form of individual'' V'' impress. Small percentage of
grains displayed sets of parallel striations. P
ercentage of grains with
striations

mark
s

is about
5% of total number of grains. Length of striation

marks on average

is

between
15
-
20
μ
m and
sets
are consisting of ten straitions in average
.

Sample Zma

1/1:



Fig
.

6
.

SEM images of quartz grains from southern loess profile.

Note
:
A: Detail of'' V'' impress
(
1
)

and lot of
small dents and cracks in the surface of grain. B: Whole grain, schist
-
like fracture
(
2
)

and'' V'' impress
(
1
)

in the
upper right part of the grain. C: The same grain (0104 pic.), with focus on the'' V'' impress
(
1
)
. D: Whole grain,
schist
-
like fractures
(
2
)
. E: Almost completely rounded grains with a schist
-
like fractures
(
2
)
. E: Almost
completely rounded grains with a schist
-
like fractures
(
2
)
. F: Grain surface displays the'' V'' impress
(
1
)
. Parallel
striations
(
3
)

are also visible.


Sample Br 1/12:

11



Fig
.

7
.

SEM images of quartz grains from northern loess profile.
Note
:

G: Subangular grain with partially visible
schist
-
like fracture
(
2
)
.

H: Whole, semispherical grains with a schist
-
like fracture
(
2
)

and severe'' V'' impress
(
1
)
.
I: The same grain (0112 pic.), zoomed in on'' V'' impress
(
1
)
.

J: The lower part of the gr
ain, partially rounded,
with ''
V'' impress
(
1
)
. K: Subangular grain with a big schist
-
like fracture
(
2
)
. L: Subangular grain with lots
of
striation marks and schist
-
like fracture
(
2
)
.


5. Discusion

Almost 5
0% of loess profiles

inves
tigated by Nemecz

et al
.

(2000) are constituated of
coarse silt (20
-
45 μm)

grain size.

Loess

profile
s from

Bansko

brdo hill

in most
sedimentological
characteristics
can be corelated with other loess profiles in Pannonian Basin.
Grain size distribution displays

a dominant share of coarse silt, as in other loess pr
ofiles from
the eastern Croatia, like

Šarengrad (Galović

et al., 2011) and
Vukovar (Wacha &

Frechen,
2011)

wh
ich are located in range of 50 km from

Bansko

brdo hill
. The
percentage of sand is
in a similar range as in other loess profiles in Pannonian Basin and

varies from 5
%

to 20
%.
N
orthern
loess
pro
file from Bansko brdo hill has

slightly highe
r

percentage of sand
-
size
particles,
espec
ially in the base of LPS.

When

reconstruct
ing the

loess deposition in a
specific
locality
geologist
must take into
account v
arious factors and concepts. Pye

(1995) described the necessary conditions an
d
processes t
hat lead to formation of loess deposits. According to him

there are two basic
requirements f
or the formation of thick

loess

deposits
: 1
.) Source

area sufficiently ri
ch with
silty material 2.) Presence of

a adequate 'entanglement' in

areas in which
silt is
accumulating.
If both conditions are met,

accumulation of loess

is possible and it

takes place in four phases
(Pye, 1995). The source
area for the primary material are mountain chains, which in this
case
of l
oess from Bansko brdo hill could be the Alps, th
e

Carpathians
and nearby Slavonian Mts.
surrounding Pannonian Basin (Banak et al
.
, 2012)
. Kuenen (1960, 1969), Wright (2000, 2007)
and Smalley et al. (2005) describe
d

physical and chemical processes of weathering of rocks,
result
ing in production of mater
ial

for the later

formation of loess. Weathering of parent rocks

is a combination of several processes. Glacial erosion, fluv
ial erosion, freez
-
thaw of water/ice
12


in rock crevasses, aeolian abrasion of rocks by

sand particles
and
tectonic movements are the
most important processe
s of weathering. This is the first phase

in the creation of loess

deposits. Subsequently, streams, floods and rivers

transport the
material
in

flood plains, which
constitutes the second phase. In the th
ird phase in the summer period,

dried river sediment
which is dominated

by silt
-
sized particles together with

small
amounts of fine
-
grained sand, is

exposed at surface and can be subjected to deflation. In the fourth phase northern

winds

(Hobbs, 1942,

1943)

blow away the

sediment
and tr
ansport it in large accomodation areas,
like

the steppes
of Pannonian Basin
.

It should be noted that deflation of silt
-
size and sand
-
size
particles was
possible

only if vegetation, which

act
s

as a
stabilizing factor of sediment, did not
grow
. At the same t
ime, the presence of veg
etation and microorganisms within, wich excrete
p
olysaccharides

and create so
-
called B
iological
Loess C
rust
(
BLC),
is
an essential factor in
the stabilization
and erosion prevention of wind deposited particles

(Smalley

et al., 2011). It is

unlikely that the deflation was effectiv
e in the winter, because snow or/and ice

covered the
sediment. It i
s known that in Europe

thickest loess deposits
are regularly very
close to major
rivers such as the Danube,
Rhine, Tisza, Dniep
ar and Dniesta
r (Smalley et al.
, 2009). It shows
how much the alluvial
, flood plain sediment

sediments
is

important in the formati
on of loess
deposits. Thickness
of loess in the vicinity of large rivers (
in average
50 km

from river beds)
suggests that most

of the silt and sand

particles
,

before final sedimentation
,

were transported

at
rel
atively short distances thus,

aeolian transport
has
proximal character. The predominan
t
mode of aeolian transport is saltation

and only the smallest particles
, such

as fine
-
grained silt,
are transported in air suspension

(Goossens, 1988; Smalley et al.
, 2009).

There is also a

division of

the processes that

result in the
formation

of loess

in three

phases.

So Smalley

et

al
.

(2009)

propose

the following three

stages:

1.)

w
eathering

2.)

fluvial

transport
/deposition

of

silt

and sand

3.)

a
eolian

deflation

of
sediment

from

floodplains.

They believe that

the process
can be

distinguished

in four

phases
, as indicated
in the previous

concept

(
Pye,
1995).

In doing
so,

the

second

pha
se

should be
divided

in
two
separate

processes: transport and sedimentation

(
Smalley

et

al
., 2009).
Both processes

are important

because they

enhance

the

sorting

of
particles
, particularly
because
they

removes

the
smallest

particles of

fine
-
grained
silt

and

clay

and thus

form a non
-
cohesive
sediments

with

high
ratio

of

silt
/
clay
.

That non
-
cohesive
sediment can be easely

subject
ed

to

deflation
.

This research

proposes the introduction of

five

phases

in the process

(Fig 8
)
,

describing the

formation of loess
-
palaeosol sequences (LPS)

in Baranja.

The same

principle could

be

applied
to describe formation of thick LPS in

the

entire

Pannonian
Basin
.


13



Fig
.

8
.

Schemtic presentation of processes
divided in five phases
required for loess forming.

Complex
process of

transporting

silt
and sand particles can be best explained with

SEM image
s which display

shape of quartz grains and
surface
textures
(Trewin,
1988). The
study of

surface
s

under high magnification provides i
nsight into the mechanical fractures of
quartz gr
ains. These fractures are 'fingerprints' of

different types of transportation. For this
purpose, numerou
s photographs of variety of quartz grains from

diferent
sediment
s were
scanned and atlases of

surface texture
s
, shape
s, and

fractures
were made
(Mahaney
, 2002). If
a quartz grains originate from crystalline rocks they

have an extremely low sphericity and are
full of cracks (Trewin, 1988). Precisely
this grain shape and forms are observed on quartz
grains from the loess profiles of Bansko brdo hill. Schist
-
like fractures (a), that are

visible in
most grains in both samples
, are effect of glacial

and aeolian transport of silt and sand (Strand
& Immonen, 2010). A large number of grains o
n the surface have so called '' V
''
imppresses
(b)

that o
ccur as a result

of the grinding, bouncing and collision between grains in the process
of saltation.

They can also be the effect of simmilar process but in the rivers and streams with
high
water
energy
(Mahaney, 2002; Strand & Immonen, 2010). A small number of grains
ha
ve

visible sets of parallel striations (c) which indicates

the process of g
lacial abrasion of
grains

under the pressure of ice (Strand & Immonen, 2010). These three most common
texture
s (a,b and c) on quartz grains in samples from

Bansko brdo hill
loess conf
irm the
compl
ex process of transporting

silt and sand from the pl
ace of their origin, to sedimentation
area
. Experi
mental work in the laboratory wind tunnels measured
wind energy,

and the speed
at which
certain mechanical grain damage
can be produced
(Bauer et al., 2004). It turned out
that the aeolian transport over short d
istances in short time interval

is
sufficient to cause
significant mechanical damage to the quartz grains (Costa et al., 2012). It was
found that
saltation

of

sa
nd
-
size particles st
arts at wind speed of 8 m/s

and when the wind speed is
increased up to 13 m/s sand
-
size particles were lifted and transported

in
air suspension

(Costa
et al., 2012). The experim
ent was conducted in environment with
a relatively high percentage
of moisture
in the air (about 80%), so we ca
n assume that in more arid

conditions

saltation
and airlift in suspension occur at lower wind speed. There is

no data
for silt
-
sized particles,
but
since they are smaller and lighter

then sand
-
size particles
, the minimum
wi
nd
speed
required to
start the

s
altation is much lower than 8 m/
s
.

By comparing the shapes and textu
res
recorded in quartz grains of

Bansko
brdo loess
with experimental data, it can be assumed that
14


the quartz grains
are of glacial origin and were

transport
ed

in Baranja in several phases. The
final phase
-
a
eolian transport, h
as proximal
character.

Comparing heavy mineral assemb
lage from

LPS

at Bansko brdo hill
, with

investigation of Thamó
-
Bozsó & Kovács (2007)
, it is obvious, that the main source area for
loess in Baranja is from the Danube flood plain sediments. The main transport direction was
from the North or North
-
West. Nevertheless, the higher concentration of amphiboles in
southern and northern loess pro
files
(if compared with
those from the Danube plain in c
entral
Hungary), suggests an additional source area. The Western Carpathians with Neogene calc
-
alkaline volcanic rocks is the major source for amphiboles (
Thamó
-
Bozsó & Kovács
, 2007).
The percentage of amphiboles in the HMF
in Baranja

is

fairly constant, averaging 27,2
%

in

southern loess profile and 25.
4% in northern loess profile
. Alternatively those minerals could
also be denudation products from locally exposed volcanic and
metamorphic rocks of the
southward nei
ghboring Slavonian Mts. (Jamičić

et al.
, 1987). Mt. Krndija and Mt. Papuk,
which are

the closest to Baranja of all
the Slavonian Mts., consist of amphibolites.
Furthermore, Pliocene sands from the northern slopes of Mt
. Krndija and Mt. Papuk are of
local origin and contain abundant am
phiboles (Jamičić

et al., 1987).

It is useful to determine

did the wind direction changed during Late Pleistocene.

Following the
work of Martinson et
al. (1987)
and adjustements made by

Wooilard & Mook (1982) and Vandenberghe (1985) it is
fair to say that MIS stages 4,

3, and 2 greatly conform with Early, Middle and Late
Pleniglacial in Europe loess region.
The
investigation of grain
-
size record and
mass
accumulation rates
of loess in cen
tral and eastern Europe
suggest a domination of western
winds

during the Early and Middle Pleniglacial in central and eastern

Europe, while the Late
Pleniglacial

was dominated by northwestern

or northern wind
s (Bokhorst et al., 2011). In this
investigation

it is obvious that certain
differences occured. Amphibole percentage in 8 upper
samples from southern loess profile display considerably higher values then samples from
Danube flood plain sediments. This two upper horizons of loess
are

deposited during MI
S 4
and MIS 2 (Banak et al., 2012). For MIS 3 stage there are no data, because palaeosol was not
sampled. Corealting ages of these two loess horizons with percentage of amphiboles in 8
samples it is possible to reconstruct the wind direction during MIS 4 a
nd MIS 2 for Baranja
region (Fig
.

9). Diference regarding investigation of Bokhorst et al. (2011) is present in loess
horizon deposited during MIS 4, while upper loess horizon

deposited during MIS 2

displays
same wind direction. During MIS 4 period wind di
rection that transported silty material to
Baranja was from nort
-
nortwest and not from west (Fig
.

9).


15



Fig
.

9
.

Wind direction
in central and eastern Europe
during Pleniglacial.

Corelation with MIS stages and Baranja
loess is set in Galović et al. (2009) and Banak et al. (2012).
Difference beetwen Baranja and other loess
profiles
in Pannonian Basin is present during Early Pleniglacial or MIS 4 stage.



6. Conclusion

Loess is not j
ust accumulation
of airborne dust (Pecsi
, 1990). T
his

is certa
i
nly true
statement
supported with

data

from

that

investigation. Never the l
ess, aeolian transport

is
16


critical factor in formation of thick loess deposits in Baranja, aswell in other

regions of

Pannonian
Basin
.

This final phase in complex
process of transporting silt and sand from
source area to acommodation space can be regarded as
conditio sine q
ua

non

in terms of

necessity for

generating
thick loess
-
palaesol sequences.

SEM images confirm com
plex multi
-
phase transport wich is divided in five phases.

Loess from Baranja is comparable with loess deposits from other Pannonian regions.
Grain
-
size distribution displays dominance od silt
-
size particles. CaCO
3

content is in
accordance with other loes
s profiles. Modal analysis indicate differences which are highlited
in percentage of amphibole group of minerals. Wind direction that transported silty material
for Baranja LPS during Early Pleniglacial was from north and
nort
h
wes
t

as opposed to other
LPS where western winds dominated.


Acknowledements

This paper was suported by Project
s

no.
181
-
181
-
1096
-
1093
,
195
-
1951293
-
2703

and
195
-
1951293
-
0237

of the Croatian Ministry of Science, Education and Sports.
H
ereby we
express our tha
n
k
s to Tamara Troskot
-
Čorbić and
Renata Slavković for providing technical
support
and knowledge
necessary for making SEM images of quartz grains.


References

Baćani
, A.,
Šparica, M., Velić
, J.
, 199
9. Quaternary deposits as the
hydrogeological

system of


Eas
tern Slavonia.

Geol. Croat.

52 (2), 141
-
152
.


Banak, A., Mandic, O., Kovačić
, M.,

Pavelić, D. 2012.

Late Pleistocene

climate history of the


Baranja loess plateau
-
ev
idence from the Zmajevac loess
-
pal
a
eosol
section



(northeastern Croatia). Geol. Croat.

65/3, 411
-
422
.

Banak, A.,

2012.

Reconstruc
tion of Late Pleistocene
climate change based on loess



sedimentology, malacofaunal palaeontology
and isotope analysis (Baranja, e
astern


Croatia). Ph
D Thesis, University of Zagreb.

Bauer
,
B.O., Houser, C.A.,
Nickling, W.G. 2004.

Analysis of velocity profile

measurements


from wind
-
tunnel experiments with saltation. Geomorphology
.

59, 81
-
98.

Bognar, A. 1976.

Les i lesu slični sedimenti i njihovo geografsko značenje [Loess and loess
like deposits and their geog
raphical meaning
-

in Croatian].
-

Geogr. Horizont.

1
-
2, 15
-
31.

Bokhorst, M.P., Vandenberghe, J.,

Sümegi, P., Łanczont, M.,
Gerasimenko, N.P., Matviishina,


Z.N., Marko
vić, S.B. & Frechen, M., 2011.

Atmospheric circulation patterns in central
17




and eastern E
urope during the

Weichselian Pleniglacial inferred

from loess grain
-
size


records. Quat
ern
. Int. 234, 62
-
74.

Buggle, B., Glaser, B., Zöller, L., Hambach, U., Markovi
ć
, S., Glaser, I., Gerasimenko, N.,

2008.

Geochemical characterization and origin of southeastern and eastern

European loesses (Serbia, Romania, Ukraine).

Quat
ern
. Sci. Rev.

27

(9
-
10), 1058
-

1075.


CGS
-
Department for Geology, 2009
.

Geologic
al Map of Republic of Croatia,

M 1:300.000.

Croatian Geo
logical Survey, Department for Geology, Zagreb.

Costa, P.J.M., Andrade, C., Mahaney, W.C., Marques Da Silva, F., F
reire,
P., Freitas, M.C.,


Janardo, C., Oliveira, M.A., Silva, T., Lopes
, V., 2012.

Aeolian microtextures in silica


spheres induce
d in a wind tunnel experiment:
Comparison with aeolian quartz.



Geomorpholgy, 4118.

Goossens
, D., 1988.

Scale model simulations of the deposi
tion of loess in hilly terrain. Earth


Surf. Proces. Land.

13, 533
-
544.

Haase, D., Fink
, J., Ruske, G., Pecsi, M.,

Richter, H., Altermann, M. & Jaeger
, K.
-
D., 2007.


Loess in Europe
-
its spatial di
stribution based on a European
Loess Map, scale



1:2,500,000.

Quatern. Sci. Rev.

26, 1301
-
1312.

Hećimović
, I.
,
1991
.

Tuma
č

OGK 1:100.000 list Moha
č
. Geomorfolo
ška analiza.

Fond

stru
č
. dok. IGI, 5 pp, Zagreb.

Hobbs
, W.H., 1942.

Wind
-
the dominant transportation agent within extramarginal
zones


to continental glaciers. J. Geol.

50, 556

559.

Hobbs
, W.H., 1943.
The glacial anticyclones and the European con
tinental glacier. Am.
J.


Sci.

241, 333

336.

Jamičić, D., Brkić, M., Crnko, J., Vragović
, M., 1987.

Osnovna geološka karta SFRJ
1:100000. Tumač za list Orahovica L33
-
96 [Basic Geological Map of SFRY 1:100000,
Geology of the O
rahovica sheet


in Croatian].

Inst. Za geol. Istraž., Zagreb, Sav.

geol.

zavod, Beograd, 1
-
72.


Kuenen, P
.H., 1960.

Experime
ntal abrasion 4: eolian action.

J.

Geol.

68,

427

449.


Kuenen
, P.H., 1969.

Origin of quartz silt. J. Sed. Petrol.

39, 1631

1633.


Lugović, B., Majer, V. &
Strumpfl
, W.E., 1990.

Geochemical Characteristics of Basaltic
Andesites from
Baranja (Croatia, Yugoslavia).

Geol. vjesnik, 43, 135
-
142.

Mahaney
, W.C., 1995b.

Glacial crushing, weathering and diagenetic histories of

quartz grains inferred from scanning elec
tron microscopy. In: Menzies, J. (Ed.),


18


Modern Glacial Environments
-

Processes, Dynamics and Sediments. Butterworth
-

Heinemann Ltd, Oxford, pp. 487
-
506.

Mahaney
, W.C., 2002.

Atlas of sand grain surface textures and applications. Ox
ford



University Press
, 237 pp.

Marković, S.B., Bokhorst, M.P., Vandenberghe, J., McCoy, W.D., Oches, E.A., Hambach, U.,
Gaudenyi, T., Jovanović, M., Zöller, L., Stevens, Machalett, B. 2008.

Late Pleistocene
loess
-
paleosol sequences in the
Vojvodina region, north Serbia. J.

Quat
ern. Sci.

23, 73
-
84.

Martinson, D., Pisias, M.,G., Hays, J.D.
, Imbrie, J., Moore, T., C.,
Shackleton
, M.J., 1987.


Age dating and th
e orbital theory of ice ages:
development of a high
-
resolution 0 to


300 000
-
year ch
ronostratigraphy. Quat. Res.
27,
1
-
30.

Menge, M.A.,

Maurer
, H.F.W.
, 1992.

Heavy minerals in Colo
r.
Chapman and Hall, London,
151 p.

Moss
, A.J.
,
1966
.

Origin, shaping and significance of quartz

sand grains. J. Geol. Soc. Aus.


13, 97
-
136.


Mutić, R.
1975a: Pijesak rijeke Drave u naslagama
bušotine B
-
12 nedaleko podravske



Slatine.

Geol. vjesnik, 28, 243
-
268, Zagreb.

Nemecz, E., Pecsi, M., Hartyani, Z. & Horvath
, T., 2000.

The origin of the silt

size quartz


grains and minerals in loess
.

Quatern. Int.

68
-
71, 199
-
208.


Pamić, J.,

P
é
cskay
,
Z., 1996.

Geological and K
-
Ar ages of Tertiary volcanic formations from
the southern part of the Pannonian Basin in Croatia
-

based o
n surface and subsurface
data.

Nafta, 6, 195
-
202.

Pamić, J.,

Pikija
, M., 1987.

Badenian andesites and pyroclastics from Baranja in n
orthern
Croatia (Yugoslavia). Acta Geol. Hung.

30/1
-
2, 111
-
132.

Pecsi
, M., 1990.

Loess is not just

accumulation of airborne dust.

Quatern. Int.
7/8, 1
-
21.

Pecsi
, M., 1995.

The role of principles and met
hods in loess
-
paleosol investigations.



GeoJournal 36, 117
-
131.


Pikija, M.,

Šikić
, K.
, 1991.

Osnovna geolo
š
ka karta 1:100.000 list Moha
č
.

Fond str.

dok. IGI, Zagreb.


Pye
, K., 1987.

Aeolian Dust and Dust Deposits.

Academic Press, London,
pp,
334.


Smalley
, I.J.
,
1966b
.

The properties of glacial loess and the

formation of loess deposits.


J. Sed. Petrol.

36, 669
-
676.


19


Smalley, I.J., Kumar
, R.,
O'hara
-
Dhand, K., Jefferson, I.,F. & Evans
, R.D., 2005.

The



formation of silt material for terrestrial sediments:

Particularly loess and


dust
.

Sed.


Geol
, 179, 321
-
328.


Smalley, I.J., O'hara
-
Dhand, K., Wint, J., Machalett, B. & Jefferson
, I., 2009.

Rivers and


loess: The significance o
f long river transportation

in
the
complex event
-
sequence


appro
ach to loess deposit formation.

Quatern. Int
, 198, 7
-
18.

Smalley
,

I.J., Marković, S.B.,

Svirčev
, Z., 2011.

Loess is (almost totally formed

by) the



accumulation of dust. Quatern. Int
.

240, 4
-
11.

Strand
, K., Passchier
,
S. &

Näsi
, J., 2003.

Implications of quartz grain microtextures for

onset


of Eocene/Oligocene glaciation in Prydz Bay, ODP Site 1166, Antarctica.

Palaeo.


Palaeo. Palaeo.

198, 101
-
111.


Strand, K. & Immonen
, N., 2010.

Dynamics of the Bare
nts
-
Kara ice shee
t as revealed
by


quartz sand grain microtextures of the Late Plei
stocene Arctic Ocean sediments.



Quat. Sci. Rev.

29, 3583
-
3589.

Thamo
-
Bozso, E. & Kovács
, L., O., 2007.

Evolution of Quaternary modern fluvial



network in the Mid
-
Hungarian plain, indicate
d by heavy mineral distributions and


statistical
analysis of heavy mineral data.

Develop.

in

Sed
. 58,
491
-
514.

Trewin
, N., 1988.

U
se of scanning electron microscope in sedimentology.
-

In: Techiques


in Sedimentology (Ed. M. E. TUCKER). Blacwell Scientif
ic

Publications, Oxford,

pp,


229
-
273.

Vandenberghe, J., 1985.

Paleoenvironment and stratigraphy during the last glacial in


the Belgian
-
Dutch border region.

Quat. Res.

24 (1), 23
-
38.

Van Loon
, A.J., 2006. Lost loesses.

Earth Sci. Rev
.

74, 309
-
316.

Wentworth
,

C., K., 1922.

A scale
of grade and cl
ass terms

for clastic sediment.

J.Geology V.


30, 377
-
392.

Whalley,
W.B., Marshall, J.R. & Smith
, B.J., 1982.

Origin of

desert loess from

some



experimental observation.

Nature
,
300, 433
-
435.


Woillard, G., Mook,
W., 1982.

Carbon
-
14 dates at Grande Pile: correlation of land


and sea
-
chronologies. Science 215, 159
-
161.


Wright
, J.S.
, 1995.

Glacial comminution of quartz sand grains and

the production of



lo
essic silt: a simulation study.

Quatern.

Sci. Rev
.

14, 669
-
6
80.


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, 231
-
256.


20