A natron source at Pikrolimni Lake in Greece? Geochemical evidence
A natron source at Pikrolimni Lake in Greece? Geochemical evidence
A natron source at Pikrolimni Lake in Greece? Geochemical evidence
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
Contents lists available <strong>at</strong> ScienceDirect<br />
Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion<br />
journal homepage: www.elsevier.com/loc<strong>at</strong>e/jgeoexp<br />
A <strong>n<strong>at</strong>ron</strong> <strong>source</strong> <strong>at</strong> <strong>Pikrolimni</strong> <strong>Lake</strong> <strong>in</strong> <strong>Greece</strong> <strong>Geochemical</strong> <strong>evidence</strong><br />
E. Dotsika a, ⁎, D. Poutoukis b , I. Tzavidopoulos a , Y. Mani<strong>at</strong>is a , D. Ign<strong>at</strong>iadou c , B. Raco d<br />
a Lab. of Archaeometry, NCSR “Demokritos”, 153 10 Aghia Paraskevi, Attiki, <strong>Greece</strong><br />
b General Secretari<strong>at</strong> for Research and Technology, 10-14 Messogion, Athens, <strong>Greece</strong><br />
c Archaeological Museum of Thessaloniki, 6 M. Andronikou Street, Thessaloniki, <strong>Greece</strong><br />
d Institute of Geosciences and Earth Re<strong>source</strong>s, Via G. Moruzzi 1, 56124 Pisa, Italy<br />
article<br />
<strong>in</strong>fo<br />
abstract<br />
Article history:<br />
Received 2 February 2009<br />
Accepted 6 August 2009<br />
Available onl<strong>in</strong>e 21 August 2009<br />
Keywords:<br />
Non-mar<strong>in</strong>e salt<br />
N<strong>at</strong>ron<br />
Br<strong>in</strong>e<br />
Alkal<strong>in</strong>e w<strong>at</strong>er<br />
<strong>Lake</strong> <strong>Pikrolimni</strong><br />
The geochemical conditions th<strong>at</strong> are responsible for the form<strong>at</strong>ion of “Chalastraion nitron” <strong>in</strong> the bas<strong>in</strong> of<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> (Northern <strong>Greece</strong>) were <strong>in</strong>vestig<strong>at</strong>ed <strong>in</strong> this study. The goal of the study is to confirm Pl<strong>in</strong>y's<br />
description: “<strong>at</strong> Clitae <strong>in</strong> Macedonia it is found <strong>in</strong> abundance the best, called soda of Chalestra…” The ionspecific<br />
<strong>in</strong>teraction model based on Pitzer's equ<strong>at</strong>ions showed th<strong>at</strong> the lake br<strong>in</strong>e samples are s<strong>at</strong>ur<strong>at</strong>ed<br />
versus carbon<strong>at</strong>e–bicarbon<strong>at</strong>e sodium and sodium–chloride–sulf<strong>at</strong>e m<strong>in</strong>erals. The results of <strong>in</strong>dividual X-<br />
Ray Diffraction (XRD) analyses on the evapor<strong>at</strong>ive deposits showed th<strong>at</strong> the salts consist mostly of trona,<br />
burkeite, thenardite and halite confirm<strong>in</strong>g Pl<strong>in</strong>y's description.<br />
© 2009 Elsevier B.V. All rights reserved.<br />
1. Introduction<br />
Dur<strong>in</strong>g the last decades, the orig<strong>in</strong> of <strong>n<strong>at</strong>ron</strong> (Na 2 CO 3 ) has been a<br />
major issue among the researchers of ancient glassmak<strong>in</strong>g. N<strong>at</strong>ron<br />
was used dur<strong>in</strong>g the Greek–Roman period as a <strong>source</strong> of soda, which is<br />
one of the three basic components of glass. Its use was loc<strong>at</strong>ed <strong>in</strong><br />
Eastern Mediterranean, Egypt and Phoenicia, from the Bronze Age<br />
until the medieval time. In Mesopotamian glassmak<strong>in</strong>g, the ma<strong>in</strong><br />
<strong>source</strong> of soda was halophyte ash. The glasses made of plant ash have<br />
different composition <strong>in</strong> rel<strong>at</strong>ion to those made of <strong>n<strong>at</strong>ron</strong>, so it is<br />
practically possible to dist<strong>in</strong>guish the West Asian glass products from<br />
those from Mediterranean areas. Glasses from Archaic, Classical and<br />
Hellenistic times which have been found <strong>in</strong> <strong>Greece</strong> are all made of<br />
<strong>n<strong>at</strong>ron</strong> (Ign<strong>at</strong>iadou, 2002).<br />
N<strong>at</strong>ural <strong>n<strong>at</strong>ron</strong> deposits, cre<strong>at</strong>ed by evapor<strong>at</strong>ion and desicc<strong>at</strong>ion<br />
processes <strong>in</strong> isol<strong>at</strong>ed lakes, exist <strong>in</strong> Wadi N<strong>at</strong>run (De Cosson, 1936;<br />
Coulson and Leonard, 1979; Henderson, 1985), northwest of Cairo,<br />
Egypt. This <strong>source</strong> was used <strong>in</strong> antiquity but also until today and it is<br />
considered the ma<strong>in</strong> or even the only <strong>source</strong> of <strong>n<strong>at</strong>ron</strong> for glassmak<strong>in</strong>g.<br />
N<strong>at</strong>ron deposits also existed <strong>in</strong> <strong>Greece</strong>, <strong>in</strong> the region of Macedonia.<br />
N<strong>at</strong>ron is referred <strong>in</strong> liter<strong>at</strong>ure, with earliest the quot<strong>at</strong>ion of Pl<strong>at</strong>o,<br />
who refers to it as Chalastraion, used as detergent (Testimonium 1).<br />
Educ<strong>at</strong>ion, he says, must be like an <strong>in</strong>delible dye th<strong>at</strong> cannot be<br />
washed out by detergents. It must not be affected by pleasure, which<br />
is stronger than the most powerful chalastraion. One of his<br />
comment<strong>at</strong>ors added th<strong>at</strong> “Chalastra is a city <strong>in</strong> Macedonia and a lake<br />
⁎ Correspond<strong>in</strong>g author. Tel.: +30 2106503305; fax: +30 2106519430.<br />
E-mail address: edotsika@ims.demokritos.gr (E. Dotsika).<br />
where the Chalastraion nitron is formed or dissolved over a period of<br />
n<strong>in</strong>e days” (Testimonium 2). Others mention Chalastra as a city or lake<br />
<strong>in</strong> Macedonia where <strong>n<strong>at</strong>ron</strong> was be<strong>in</strong>g burned (Testimonium 3) or<br />
<strong>n<strong>at</strong>ron</strong> as Chalastraion nitron, named after Chalastra, the lake <strong>in</strong><br />
Macedonia (Testimonium 4).<br />
A rel<strong>at</strong>ively unknown quot<strong>at</strong>ion <strong>in</strong> <strong>in</strong>tern<strong>at</strong>ional studies about<br />
ancient glass is th<strong>at</strong> of Pl<strong>in</strong>y, the Elder, which <strong>in</strong>cludes the most<br />
thorough and enlighten<strong>in</strong>g reference to Chalastraion nitron, referr<strong>in</strong>g<br />
th<strong>at</strong> it abounds is the Macedonian city of Clitae (Testimonium 5). In<br />
this quot<strong>at</strong>ion, the follow<strong>in</strong>g are mentioned:<br />
“At Clitae <strong>in</strong> Macedonia it is found <strong>in</strong> abundance the best, called<br />
Chalastraion, white and pure like salt. There is an alkal<strong>in</strong>e lake<br />
there, with a little spr<strong>in</strong>g of fresh w<strong>at</strong>er ris<strong>in</strong>g up <strong>in</strong> the center.<br />
Nitron forms there when the Dog Star rises for n<strong>in</strong>e days, ceases<br />
for n<strong>in</strong>e days, comes to the surface aga<strong>in</strong> and then ceases”.<br />
Additional characteristics of the lake follow and the author cont<strong>in</strong>ues<br />
referr<strong>in</strong>g to the other famous nitron, the one from Egypt. “Here it is<br />
n<strong>at</strong>ural, but <strong>in</strong> Egypt it is made artificially and <strong>in</strong> much gre<strong>at</strong>er<br />
abundance but of <strong>in</strong>ferior quality, because it is darker and stony”.<br />
Based on the reference of Pl<strong>in</strong>y and other newer liter<strong>at</strong>ure <strong>source</strong>s,<br />
H<strong>at</strong>zopoulos and Loukopoulou (1989) suggested th<strong>at</strong> the lake<br />
Chalastra is the modern lake <strong>Pikrolimni</strong>.<br />
The whole issue is of gre<strong>at</strong> importance <strong>in</strong> studies of glassmak<strong>in</strong>g, as<br />
glass products were abundant <strong>in</strong> Macedonia dur<strong>in</strong>g the period of<br />
Pl<strong>in</strong>y. The possibility th<strong>at</strong> Chalastraion nitron was used for their<br />
manufacture can lead to the determ<strong>in</strong><strong>at</strong>ion of the orig<strong>in</strong> of the<br />
products, both Roman and pre-Roman.<br />
0375-6742/$ – see front m<strong>at</strong>ter © 2009 Elsevier B.V. All rights reserved.<br />
doi:10.1016/j.gexplo.2009.08.003
134 E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> is loc<strong>at</strong>ed 20 km to the NW of the town of<br />
Thessaloniki, <strong>in</strong> the region of Macedonia, <strong>in</strong> Northern <strong>Greece</strong>, where a<br />
spa and <strong>in</strong>stall<strong>at</strong>ions for mud b<strong>at</strong>hs exist.<br />
In order to confirm Pl<strong>in</strong>y's description, we <strong>in</strong>vestig<strong>at</strong>ed the<br />
geochemical d<strong>at</strong>a of the area and the evapor<strong>at</strong>ive conditions, which<br />
are responsible for the form<strong>at</strong>ion of “Chalastraion nitron”.<br />
2. N<strong>at</strong>ron use<br />
Evaporitic rocks have been used s<strong>in</strong>ce the dawn of human history.<br />
Salt and br<strong>in</strong>es have been used <strong>in</strong> all cultures, i.e. Knossos Palace is built<br />
on Crete Island with big blocks, some of which were pure massive<br />
gypsum. Also <strong>in</strong> more “modern” civilis<strong>at</strong>ions, the salt was used for<br />
payments. The Romans paid a “salarium” <strong>in</strong> NaCl and therefore the<br />
actual word “salary” came up for wages, orig<strong>in</strong>ally mean<strong>in</strong>g salt money.<br />
Ancient use of sodium carbon<strong>at</strong>e is known s<strong>in</strong>ce 5500–4000 BC.<br />
However, f<strong>in</strong>d<strong>in</strong>gs of the m<strong>at</strong>erial are rare. In general, the ma<strong>in</strong> use of<br />
<strong>n<strong>at</strong>ron</strong> was as a purifier <strong>in</strong> mummific<strong>at</strong>ion. Herodotus provides the<br />
best description of the use of <strong>n<strong>at</strong>ron</strong> <strong>in</strong> mummific<strong>at</strong>ion.<br />
N<strong>at</strong>ron seems to have been used <strong>in</strong> glassmak<strong>in</strong>g <strong>in</strong> Egypt s<strong>in</strong>ce the<br />
Badarian Period (early 4th millennium BC), <strong>in</strong>dic<strong>at</strong>ed by the very low<br />
potash contents of glasses (Shortland et al., 2006). The use of <strong>n<strong>at</strong>ron</strong><br />
becomes apparent s<strong>in</strong>ce the 1st millennium BC. Reported glass from the<br />
tomb of Nesikhons <strong>in</strong> Egypt with low content <strong>in</strong> potash, magnesia and<br />
lime <strong>in</strong>dic<strong>at</strong>es th<strong>at</strong> the soda was derived from <strong>n<strong>at</strong>ron</strong> (Schlick-Nolte and<br />
Werthmann, 2003). Other <strong>evidence</strong> of <strong>n<strong>at</strong>ron</strong> use <strong>in</strong> the same period is<br />
the blue glasses from 8–9th century BC Nimrud, Iraq (Reade et al., 2005).<br />
The Roman glassmak<strong>in</strong>g <strong>in</strong>dustry seems to have been based on <strong>n<strong>at</strong>ron</strong>,<br />
<strong>in</strong>dic<strong>at</strong>ed by the low-magnesia content. Until the 9th century AD, <strong>n<strong>at</strong>ron</strong><br />
cont<strong>in</strong>ued to be the basic m<strong>at</strong>erial <strong>in</strong> glassmak<strong>in</strong>g <strong>in</strong> Levant, the<br />
Mediterranean and Europe. The use of <strong>n<strong>at</strong>ron</strong> as the flux <strong>in</strong> glass<br />
production decl<strong>in</strong>es <strong>in</strong> Near East and <strong>in</strong> Europe <strong>in</strong> the 9th century AD,<br />
when soda-rich plant ash replaced it (Shortland et al., 2006). However,<br />
<strong>at</strong> al-Barnuj <strong>in</strong> the Western Delta, Egypt, <strong>n<strong>at</strong>ron</strong> was collected dur<strong>in</strong>g<br />
periods th<strong>at</strong> w<strong>at</strong>ers were dried up until the 18th century AD (Mart<strong>in</strong> and<br />
Sauneron, 1982). From the 1920s and onwards <strong>n<strong>at</strong>ron</strong> was collected <strong>at</strong><br />
the same area by the Egypt Salt and Soda Company (Evelyn-White and<br />
Hauser, 1926–1933).<br />
In ancient <strong>Greece</strong> <strong>n<strong>at</strong>ron</strong> was mentioned for the first time by Pl<strong>at</strong>o<br />
<strong>in</strong> the 5th century <strong>in</strong> Republic (Testimonium 1).<br />
Today, sodium carbon<strong>at</strong>e is formed <strong>in</strong> the subtropical region of<br />
<strong>Lake</strong> N<strong>at</strong>ron <strong>in</strong> Tanzania and <strong>Lake</strong> Magadi <strong>in</strong> Kenya (Magadi means<br />
bitter, exactly as the name of <strong>Lake</strong> <strong>Pikrolimni</strong>, which means bitterlake).<br />
In Table 2, the m<strong>in</strong>eral sequence th<strong>at</strong> precipit<strong>at</strong>es dur<strong>in</strong>g mar<strong>in</strong>e<br />
w<strong>at</strong>er evapor<strong>at</strong>ion and lacustr<strong>in</strong>e w<strong>at</strong>er evapor<strong>at</strong>ion appears (<strong>Lake</strong><br />
N<strong>at</strong>ron, <strong>Lake</strong> Magadi, Deep Spr<strong>in</strong>gs <strong>Lake</strong> and Wadi N<strong>at</strong>run).<br />
3. Meteorological and geological sett<strong>in</strong>g<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> is loc<strong>at</strong>ed <strong>in</strong> the bas<strong>in</strong> of Kilkis pla<strong>in</strong>, near Thessaloniki<br />
(23 km), <strong>in</strong> northern <strong>Greece</strong>. It is a small shallow lake which<br />
usually dries out dur<strong>in</strong>g summer. It has an average depth of about 0.5–<br />
0.7 m and covers an area about 4.5 km 2 when it is flooded. The mean<br />
annual precipit<strong>at</strong>ion <strong>in</strong> meteorological st<strong>at</strong>ion of Thessaloniki (dur<strong>in</strong>g<br />
1959–1997, Hellenic N<strong>at</strong>ional Meteorological service) is 448 mm, while<br />
the temper<strong>at</strong>ure ranges from 1.3 to 32 °C for the same period. Mean<br />
temper<strong>at</strong>ure dur<strong>in</strong>g July and August is 26.3 °C and mean precipit<strong>at</strong>ion is<br />
22 mm (1959–1997, Hellenic N<strong>at</strong>ional Meteorological service). The<br />
annual real evapotranspir<strong>at</strong>ion for Thessaloniki area (Thermi) <strong>in</strong> 2005<br />
was estim<strong>at</strong>ed 482.01 mm, with precipit<strong>at</strong>ion 481.8 mm and mean<br />
temper<strong>at</strong>ure 15.4 °C (An<strong>at</strong>oliki AE, 2006). Dur<strong>in</strong>g summer months, July<br />
and August (2005), mean temper<strong>at</strong>ure was 22.3 °C and 25.5 °C<br />
respectively, mean precipit<strong>at</strong>ion was 95 and 48 mm respectively while<br />
real evapotranspir<strong>at</strong>ion was 95 and 48 mm for the same months<br />
(An<strong>at</strong>oliki AE, 2006). Also Dalezios et al. (2002) report th<strong>at</strong> the maximum<br />
values of reference evapotranspir<strong>at</strong>ion r<strong>at</strong>es are observed ma<strong>in</strong>ly over<br />
the northern region and particularly over the pla<strong>in</strong>s surround<strong>in</strong>g the<br />
st<strong>at</strong>ions of Thessaloniki and Kilkis <strong>in</strong> the majority of months. This could<br />
be merely due to the fact th<strong>at</strong> ma<strong>in</strong>ly <strong>in</strong> summer months strong w<strong>in</strong>ds<br />
prevail over <strong>Greece</strong>, which along with the lack of precipit<strong>at</strong>ion cause a<br />
gre<strong>at</strong> deal of aridity variability among different sites. In particular, the<br />
observed maximum values of reference evapotranspir<strong>at</strong>ion r<strong>at</strong>es <strong>in</strong> the<br />
summer months over the Thessaloniki pla<strong>in</strong> could be due to a strong and<br />
dry N–NW w<strong>in</strong>d p<strong>at</strong>tern named Vartharis.<br />
Sediments of Neogene and Qu<strong>at</strong>ernary age are prevalent <strong>in</strong> the<br />
region of <strong>Lake</strong> <strong>Pikrolimni</strong> (Fig. 1). The Qu<strong>at</strong>ernary deposits are<br />
Holocene river and lake sediments, alluvial we<strong>at</strong>her<strong>in</strong>g m<strong>at</strong>erial<br />
composed ma<strong>in</strong>ly of schists and carbon<strong>at</strong>e sandy clays, sands and<br />
gravels of lower terrace system (carbon<strong>at</strong>es). Pleistocene deposits<br />
conta<strong>in</strong> gravels, red clays and sands with calcareous concretions and a<br />
big amount of mica. The prevalent Neogene form<strong>at</strong>ion (Mountrakis,<br />
1985) is the sandstone–marl series, <strong>in</strong>clud<strong>in</strong>g Hipparion mediterraneum,<br />
Mastodon sp., Gazella cf. deperdita.<br />
The bedrock <strong>in</strong> the region consists of Mesozoic rocks, ma<strong>in</strong>ly by<br />
conglomer<strong>at</strong>es from Upper Jurassic, <strong>in</strong>clud<strong>in</strong>g Ner<strong>in</strong>eae, Cladocoropsis sp.<br />
and Pseudocyclamm<strong>in</strong>a sp. (Mercier, 1966), with quartzite, greywacke<br />
(quartz and feldspars), quartz and limestone pebbles, basic rocks<br />
(gabbro) from Upper Jurassic with pyroxene, altered <strong>in</strong>to biotite and<br />
act<strong>in</strong>olite, hornblende, plagioclase feldspar and epidote and limestone<br />
from Middle–Upper Triassic. In a distance less than 10 km from <strong>Lake</strong><br />
<strong>Pikrolimni</strong>, the Paleozoic basement is uncovered consist<strong>in</strong>g of maficrocks<br />
(amphibolite with amphibole as hornblende) and plagioclase feldspars.<br />
4. Sampl<strong>in</strong>g and analysis<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> is not fed by any major river and the hydrography<br />
of the area shows th<strong>at</strong> no perennial streams enter the lake directly. In<br />
the past, the major <strong>source</strong> of w<strong>at</strong>er to the lake was ground w<strong>at</strong>er<br />
(Pl<strong>in</strong>y; Many public testimonies vouch th<strong>at</strong> there were a lot of spr<strong>in</strong>gs<br />
<strong>in</strong> the middle of the lake, but after the big earthquake of 1981 these<br />
spr<strong>in</strong>gs have disappeared). Specifically, there were spr<strong>in</strong>gs very close<br />
to the lake and with<strong>in</strong> the lake itself. These percol<strong>at</strong><strong>in</strong>g groundw<strong>at</strong>ers<br />
probably brought the necessary ionic charge <strong>in</strong>to the lake. So dur<strong>in</strong>g<br />
the summer, accord<strong>in</strong>g to Pl<strong>in</strong>y, the precipit<strong>at</strong>ion of the salt was<br />
tak<strong>in</strong>g place. Actually, the ma<strong>in</strong> <strong>in</strong>put of fresh w<strong>at</strong>er to <strong>Pikrolimni</strong> is<br />
the ra<strong>in</strong>. Also, a number of wells and two spr<strong>in</strong>gs are loc<strong>at</strong>ed <strong>in</strong> the<br />
lake area (one spr<strong>in</strong>g is loc<strong>at</strong>ed <strong>in</strong> the south perimeter of the lake,<br />
nr. 21 <strong>in</strong> Table 1 and one spr<strong>in</strong>g emerges <strong>in</strong> the <strong>Pikrolimni</strong> Village).<br />
The spr<strong>in</strong>g w<strong>at</strong>er (nr. 21 <strong>in</strong> Table 1), ranges <strong>in</strong> temper<strong>at</strong>ure from 28 to<br />
30 °C and does not vary <strong>in</strong> discharge and concentr<strong>at</strong>ion. It emerges <strong>in</strong><br />
the ancient bas<strong>in</strong> of the lake and feeds the lagoon which is perennially<br />
wet marg<strong>in</strong>al lake areas. Dur<strong>in</strong>g the w<strong>in</strong>ter, when temper<strong>at</strong>ure is low,<br />
the <strong>in</strong>flow of w<strong>at</strong>er (ma<strong>in</strong>ly ra<strong>in</strong> w<strong>at</strong>er) <strong>in</strong>to <strong>Lake</strong> <strong>Pikrolimni</strong> exceeds<br />
the evapor<strong>at</strong>ion and the lake fills with w<strong>at</strong>er. Dur<strong>in</strong>g the summer,<br />
when the temper<strong>at</strong>ure and evapor<strong>at</strong>ion <strong>in</strong>crease, the level of the lake<br />
beg<strong>in</strong>s to fall, the concentr<strong>at</strong>ion of the ions <strong>in</strong>creases and evaporite<br />
m<strong>in</strong>erals beg<strong>in</strong> to precipit<strong>at</strong>e out of the solution. Dur<strong>in</strong>g some<br />
summers, when the lake w<strong>at</strong>er is exposed to very <strong>in</strong>tense evapor<strong>at</strong>ive<br />
concentr<strong>at</strong>ion, the br<strong>in</strong>es cover all the bottom of the lake. This br<strong>in</strong>e<br />
body vanishes dur<strong>in</strong>g the dry seasons expos<strong>in</strong>g salt deposit, as shown<br />
<strong>in</strong> Fig. 1. In the next year, dur<strong>in</strong>g the ra<strong>in</strong>y seasons, the lake refills with<br />
meteoric w<strong>at</strong>er and this w<strong>at</strong>er dissolves the salts and the shallow<br />
br<strong>in</strong>e body cover<strong>in</strong>g the ma<strong>in</strong> lake surface. This sequence of fill<strong>in</strong>g/<br />
depletion of the lake was observed dur<strong>in</strong>g the five years of this study.<br />
Therefore, evapor<strong>at</strong>ive concentr<strong>at</strong>ions dom<strong>in</strong><strong>at</strong>e the chemical composition<br />
of the lake w<strong>at</strong>er and it is the driv<strong>in</strong>g force for the evolution of<br />
<strong>Pikrolimni</strong> br<strong>in</strong>es, although only one segment is observed directly: the<br />
lake br<strong>in</strong>es. The spr<strong>in</strong>g is postul<strong>at</strong>ed as sub-surface flow.<br />
Samples were collected, <strong>in</strong> different times, from the borehole <strong>in</strong><br />
the thermal spa of <strong>Pikrolimni</strong> (which is loc<strong>at</strong>ed <strong>at</strong> the banks of <strong>Lake</strong><br />
<strong>Pikrolimni</strong>), from spr<strong>in</strong>g w<strong>at</strong>er and samples of br<strong>in</strong>e and salts from the<br />
lake itself. We also sampled fresh w<strong>at</strong>er of the region. The depth of the
E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
135<br />
Fig. 1. Geological map of the area of <strong>Lake</strong> <strong>Pikrolimni</strong> and sampl<strong>in</strong>g sites (•).<br />
perfor<strong>at</strong>ion of the borehole <strong>in</strong> the thermal spa is approxim<strong>at</strong>ely 250 m.<br />
This w<strong>at</strong>er is n<strong>at</strong>urally sparkl<strong>in</strong>g, with a metallic aftertaste and a slight<br />
organic smell. Str<strong>at</strong>igraphic d<strong>at</strong>a are absent for the area, however<br />
reports of the borehole <strong>at</strong> the thermal spa mention th<strong>at</strong> across the<br />
totality of the borehole's depth, only lake sediments were drilled.<br />
The samples were taken twice dur<strong>in</strong>g the year: <strong>in</strong> summer and <strong>in</strong><br />
w<strong>in</strong>ter (2002, 2003, 2004, 2005, 2006 and 2007). The samples were taken<br />
<strong>in</strong> different seasons. Samples with codes PL-8/2002 and PL-9/2006 regard<br />
residual br<strong>in</strong>es on the lake bottom. The lake was totally dried out once, on<br />
2002, and never aga<strong>in</strong> until 2006. The samples from 2003 to 2005 concern<br />
mix<strong>in</strong>g between the residual br<strong>in</strong>es and meteoric w<strong>at</strong>er. The analytical<br />
scheme <strong>in</strong>cludes field measurements of temper<strong>at</strong>ure, conductivity and pH.<br />
Samples were also subjected to chemical analysis. Filtered<br />
(0.45 µm), acidified (with HNO 3 1:1) w<strong>at</strong>er samples were collected<br />
for determ<strong>in</strong><strong>at</strong>ion of c<strong>at</strong>ions (Ca 2+ ,Mg 2+ ,K + and Na + ). Untre<strong>at</strong>ed<br />
samples were collected for analyses of anions (Cl − ,SO 4 2− ,CO 3 2− and<br />
HCO 3 − ). The major chemical constituents were analyzed with<br />
standard methods described <strong>in</strong> Apha (1989). The anions and c<strong>at</strong>ions<br />
of the w<strong>at</strong>er samples were analyzed with ion chrom<strong>at</strong>ography and<br />
<strong>at</strong>omic absorption.
Table 1<br />
Chemical compositions (mmol/L) of w<strong>at</strong>ers from boreholes, spr<strong>in</strong>gs, w<strong>at</strong>er tubes and lake.<br />
a/a Sample name Sample description D<strong>at</strong>e T (°C) pH Cl −<br />
(mmol/L)<br />
SO 4<br />
2−<br />
(mmol/L)<br />
1 S/1/2002 Spr<strong>in</strong>g–<strong>Pikrolimni</strong> Vil. 1/2002 18 9.8 3.81 1.23 3.20 5.61 0.08 0.12 14.78 0.08 0.15 Na–HCO 3<br />
2 b/2/2002 Priv<strong>at</strong>e borehole–<strong>Pikrolimni</strong> Vil. 1/2008 19 8.7 1.69 0.18 0.47 3.43 0.13 0.70 4.35 0.13 0.00 Na–HCO 3<br />
3 b/3/2002 Priv<strong>at</strong>e borehole–<strong>Pikrolimni</strong> Vil. 1/2002 19 8.6 0.59 0.38 0.87 10.61 0.48 1.03 9.78 0.13 0.00 Na–HCO 3<br />
4 b/4/2002 Priv<strong>at</strong>e borehole–<strong>Pikrolimni</strong> Vil. 1/2002 18 8.4 4.01 0.30 0.87 22.26 0.55 10.66 5.52 0.23 0.00 Mg–HCO 3<br />
5 b/A/2002 Borehole–<strong>Pikrolimni</strong> Vil. 2002 21 6.6 0.79 0.11 0.80 5.20 1.05 1.93 1.26 0.03 Mg–HCO 3<br />
6 wt/B/2002 W<strong>at</strong>er tube 2002 22.5 6.8 3.70 0.30 51.00 7.60 11.60 15.43 0.18 Mg–HCO 3<br />
7 b/B/7/2004 Borehole 2004 15.5 6.3 2.60 0.21 48.52 8.28 14.44 15.30 0.46 Mg–HCO 3<br />
8 b/B/8/2004 Borehole 8/2004 22 6.6 3.10 0.27 55.25 9.78 14.73 17.35 1.82 0.16 Mg–HCO 3<br />
9 b/G/8/2002 Borehole 2002 21 6.6 3.19 0.23 53.00 3.85 16.17 15.87 0.20 Mg–HCO 3<br />
10 b/G/9/2002 Borehole 2002 18 6.6 3.02 0.24 55.51 3.98 16.67 14.57 0.18 0.18 Mg–HCO 3<br />
11 b/G/2003 Borehole 2003 18 6.8 3.67 0.28 48.11 3.80 16.05 14.78 0.18 0.29 Mg–HCO 3<br />
12 b/G/2004 Borehole 2004 18 6.3 3.05 0.28 41.72 7.33 10.78 15.17 0.28 Mg–HCO 3<br />
13 b/G/7/2004 Borehole 7/2004 19 7.7 1.44 0.92 6.20 1.63 2.18 2.70 0.33 1.48 Mg–HCO 3<br />
14 b/P1/7/2004 Borehole 7/2004 18 6.9 3.07 0.26 46.56 11.03 15.19 14.39 0.26 0.16 Mg–HCO 3<br />
15 b/SPA Borehole/SPA 3/2007 18.2 6.1 6.77 0.25 39.49 7.10 12.92 13.83 0.15 Mg–HCO 3<br />
16 wt/SPA W<strong>at</strong>er tube/SPA 7/2004 26.9 6.1 7.42 0.32 38.00 7.23 10.91 15.96 0.20 Mg–HCO 3<br />
17 MW/SPA Municipality w<strong>at</strong>er 7/2004 22.1 7.7 8.86 1.30 5.89 3.55 2.18 7.78 0.10 Na–Cl<br />
18 PB Priv<strong>at</strong>e borehole/Philadelphia 7/2004 18.1 6.9 3.81 1.02 9.89 4.23 2.14 4.83 0.00 Ca–HCO 3<br />
19 MWP-1 Municipality w<strong>at</strong>er <strong>Pikrolimni</strong> 7/2004 23 7.2 5.11 1.49 6.00 2.90 2.30 6.39 0.08 Na–HCO 3<br />
20 MWPh-1 Municipality w<strong>at</strong>er Philadelphia 7/2004 27.5 6.9 6.46 1.29 6.49 3.35 1.93 5.17 0.10 Ca–HCO 3<br />
21 S/IGME Spr<strong>in</strong>g/near lake 7/2004 28–30 6.5 13.79 1.18 34.89 3.98 5.97 30.74 5.81 Na–HCO 3<br />
21a Ker<strong>at</strong>ea a Spr<strong>in</strong>g 1977 23 6.5 13.79 1.19 34.90 3.98 5.97 30.70 0.59 Na–HCO 3<br />
21b <strong>Pikrolimni</strong>-mud a Borehole mud spa 2000/2000 6.1 4.46 0.30 46.10 8.05 10.08 14.48 0.19 Mg–HCO 3<br />
22 PL-8/2002 <strong>Lake</strong> <strong>Pikrolimni</strong> 8/2002 27.4 8.5 4146.69 331.98 826.67 609.84 0.09 0.29 5143.43 10.26 Na–Cl<br />
23 PL-1/2003 <strong>Lake</strong> <strong>Pikrolimni</strong> 1/2003 22 9.6 170.13 18.20 28.00 14.00 8.83 0.74 246.87 0.51 Na–Cl<br />
24 PL-3/2003 <strong>Lake</strong> <strong>Pikrolimni</strong> 3/2003 23 9 72.50 8.13 6.67 20.16 3.88 0.29 104.57 0.26 Na–Cl<br />
25 PL-5/2003 <strong>Lake</strong> <strong>Pikrolimni</strong> 5/2003 25 9.2 120.37 14.01 12.00 19.10 6.83 0.45 174.13 0.41 Na–Cl<br />
26 PL-8/2003 <strong>Lake</strong> <strong>Pikrolimni</strong> 8/2003 30.6 8.7 247.95 25.52 28.83 26.07 10.73 0.78 340.00 0.59 Na–Cl<br />
27 PL-4/2004 <strong>Lake</strong> <strong>Pikrolimni</strong> 4/2004 25 8 241.47 25.63 26.17 43.44 1.35 2.26 412.61 7.83 Na–Cl<br />
28 PL-5/2004 <strong>Lake</strong> <strong>Pikrolimni</strong> 5/2004 25.4 9.1 261.78 27.98 27.33 45.57 2.08 2.92 460.87 10.03 Na–Cl<br />
29 PL-7/2004 <strong>Lake</strong> <strong>Pikrolimni</strong> 7/2004 28 9.5 499.29 54.58 32.07 44.00 22.53 7.41 723.04 14.14 Na–Cl<br />
30 PL-8/2004 <strong>Lake</strong> <strong>Pikrolimni</strong> 8/2004 27.1 8.9 1357.97 150.21 36.83 40.74 13.48 7.94 1745.65 18.72 Na–Cl<br />
31 PL-1/2005 <strong>Lake</strong> <strong>Pikrolimni</strong> 1/2005 27.5 9.1 556.56 62.08 30.50 50.90 17.10 6.13 881.30 30.43 Na–Cl<br />
32 PL-9/2006 <strong>Lake</strong> <strong>Pikrolimni</strong> 9/2006 27.5 9.1 4100.00 527.92 2795.83 0.03 0.12 6321.74 20.03 Na–Cl<br />
33 PL-3/2007 <strong>Lake</strong> <strong>Pikrolimni</strong> 3/2007 23.3 9.2 442.00 47.15 111.00 0.43 1.60 590.96 1.56 Na–Cl<br />
34 PL-6/2007 <strong>Lake</strong> <strong>Pikrolimni</strong> 6/2007 34.2 8.9 603.50 68.24 142.00 0.35 1.65 838.57 2.25 Na–Cl<br />
a<br />
Unpublished d<strong>at</strong>a, <strong>source</strong>s <strong>in</strong> text.<br />
CO 3<br />
2−<br />
(mmol/L)<br />
HCO 3<br />
−<br />
(mmol/L)<br />
Ca 2+<br />
(mmol/L)<br />
Mg 2+<br />
(mmol/L)<br />
Na +<br />
(mmol/L)<br />
K +<br />
(mmol/L)<br />
NO 3<br />
−<br />
(mmol/L)<br />
Type<br />
136 E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143
E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
137<br />
5. Chemical results<br />
5.1. Solute acquisition<br />
Four chemically different w<strong>at</strong>er types (Table 1) are recognized <strong>in</strong> the<br />
<strong>Pikrolimni</strong> area for diluted w<strong>at</strong>er and one type for the lake w<strong>at</strong>er. The<br />
four recognized groundw<strong>at</strong>er types, based on both chemical and TDS,<br />
are: Mg–HCO 3 ,Na–HCO 3 ,Ca–HCO 3 ,andNa–Cl type (only one sample).<br />
The two first types represent fresh w<strong>at</strong>er th<strong>at</strong> has flown from boreholes<br />
and spr<strong>in</strong>g <strong>in</strong> the old bottom of <strong>Lake</strong> <strong>Pikrolimni</strong>. The municipality w<strong>at</strong>er<br />
of <strong>Pikrolimni</strong> village is also Na–HCO 3 type although municipality w<strong>at</strong>er<br />
which w<strong>at</strong>ers the spa is Na–Cl type. The Ca–HCO 3 represents the aquifer<br />
of local groundw<strong>at</strong>er system <strong>in</strong> Philadelphia area. The lake w<strong>at</strong>ers are<br />
Na–Cl, however when the lake began to evapor<strong>at</strong>e until the evapor<strong>at</strong><strong>in</strong>g<br />
conditions were such th<strong>at</strong> lake w<strong>at</strong>er is nearly totally evapor<strong>at</strong>ed, then<br />
the residual w<strong>at</strong>er is Na–Cl(CO 3 –SO 4 ) type. Fig. 2 presents the<br />
<strong>Pikrolimni</strong> area w<strong>at</strong>er, groundw<strong>at</strong>er and <strong>Pikrolimni</strong> br<strong>in</strong>e, <strong>in</strong> terms of<br />
the major ions. The groundw<strong>at</strong>ers are predom<strong>in</strong>antly HCO 3 , whereas the<br />
c<strong>at</strong>ions are mixtures of Na, Ca and Mg. In contrast, the c<strong>at</strong>ions of the<br />
br<strong>in</strong>es are dom<strong>in</strong><strong>at</strong>ed by Na and the anions are a mixture. This d<strong>at</strong>a is <strong>in</strong><br />
agreement with the model proposed by Hardie and Eugster (1970).<br />
The <strong>Pikrolimni</strong> sal<strong>in</strong>e alkal<strong>in</strong>e lake is enriched <strong>in</strong> dissolved m<strong>in</strong>erals<br />
th<strong>at</strong> have accumul<strong>at</strong>ed <strong>in</strong> the br<strong>in</strong>es follow<strong>in</strong>g evapor<strong>at</strong>ion. Such br<strong>in</strong>es<br />
<strong>in</strong>dic<strong>at</strong>e a very considerable range <strong>in</strong> ionic composition and concentr<strong>at</strong>ion.<br />
The total sal<strong>in</strong>ity ranges from sal<strong>in</strong>e w<strong>at</strong>er to br<strong>in</strong>e: total dissolved<br />
solid (TDS) concentr<strong>at</strong>ions rang<strong>in</strong>g from 7.5 to 426 g/L. The pH ranges<br />
from 8.8 to 10.5 and <strong>in</strong>creases as sal<strong>in</strong>ity <strong>in</strong>creases. The surface w<strong>at</strong>er<br />
temper<strong>at</strong>ure is between 22 and 31 °C. All the sal<strong>in</strong>e lake w<strong>at</strong>ers are Na–<br />
Cl type, with Na + and Cl − represent<strong>in</strong>g about 95% and 70% respectively<br />
of the total c<strong>at</strong>ions and anions. Concentr<strong>at</strong>ions of Ca 2+ and Mg 2+ are<br />
very low <strong>in</strong> the br<strong>in</strong>es of 8/2002 and 9/2006.<br />
The two types of groundw<strong>at</strong>er of <strong>Pikrolimni</strong> area, Na–HCO 3 ,Mg–<br />
HCO 3 types, are tepid (temper<strong>at</strong>ure rang<strong>in</strong>g from 18 to 30 °C), nearneutral<br />
to alkal<strong>in</strong>e (pH rang<strong>in</strong>g from 6.3 to 9.8) and have total<br />
sal<strong>in</strong>ities th<strong>at</strong> range from fresh to brackish w<strong>at</strong>er: total dissolved solid<br />
(TDS) concentr<strong>at</strong>ions rang<strong>in</strong>g from 0.50 to 4.5 g/L. This variability of<br />
TDS is due to the HCO 3 − concentr<strong>at</strong>ions.<br />
The dom<strong>in</strong>ant ionic species <strong>in</strong> the Na–HCO 3 (soda) groundw<strong>at</strong>er<br />
are Na + and HCO 3 − –CO 3 2− with these two ions compris<strong>in</strong>g 70–80% of<br />
all ions <strong>in</strong> solution. The concentr<strong>at</strong>ion of other major ions is low, with<br />
Mg 2+ and Ca 2+ contents less than 2 mmol/L. The Cl − ,K + and SO 4<br />
2−<br />
concentr<strong>at</strong>ions are also low.<br />
In the Mg–HCO 3 type w<strong>at</strong>er the dom<strong>in</strong>ant ionic species are Mg 2+<br />
and HCO 3 − –CO 3 2− with these two ions compris<strong>in</strong>g 70–80% of all ions <strong>in</strong><br />
solution.<br />
The Na–HCO 3 w<strong>at</strong>ers were found <strong>in</strong> the <strong>Pikrolimni</strong>/Ker<strong>at</strong>ea area,<br />
approxim<strong>at</strong>ely 2 km away from <strong>Lake</strong> <strong>Pikrolimni</strong>, and these w<strong>at</strong>ers end up<br />
<strong>in</strong> the bas<strong>in</strong> of <strong>Lake</strong> <strong>Pikrolimni</strong>. Sample Ker<strong>at</strong>ea (IGME, unpublished d<strong>at</strong>a)<br />
belongs <strong>in</strong> this c<strong>at</strong>egory, which was measured with SiO 2 =31 mg/L and<br />
CO 2 =668 mg/L. The Mg–HCO 3 groundw<strong>at</strong>er was found with<strong>in</strong> the<br />
Holocene lacustr<strong>in</strong>e sediments of <strong>Pikrolimni</strong>. Sample <strong>Pikrolimni</strong>-mud of<br />
2000 (Manasis Mitrakas, unpublished d<strong>at</strong>a) belongs <strong>in</strong> this w<strong>at</strong>er type,<br />
which was measured SiO 2 =47 mg/L and CO 2 =1650 mg/L. These<br />
w<strong>at</strong>ers, of meteoric orig<strong>in</strong>, flow through the clay m<strong>in</strong>erals, rich <strong>in</strong> mica,<br />
altern<strong>at</strong><strong>in</strong>g with carbon<strong>at</strong>e rocks. The Mesozoic and Paleozoic basements<br />
<strong>in</strong> the area consist of basic–mafic rocks. When the aquifer m<strong>at</strong>rix<br />
is composed of alluvial fill, silic<strong>at</strong>e hydrolysis is the typically <strong>in</strong>voked<br />
explan<strong>at</strong>ion for the form<strong>at</strong>ion of Mg–HCO 3 groundw<strong>at</strong>er and Na–HCO 3<br />
groundw<strong>at</strong>er. So it is tempt<strong>in</strong>g to assume th<strong>at</strong> this process is controll<strong>in</strong>g<br />
the w<strong>at</strong>er chemistry. However, many other we<strong>at</strong>her<strong>in</strong>g reactions<br />
contribute to the solute load of <strong>in</strong>flow w<strong>at</strong>ers. Thus, reconsider<strong>in</strong>g the<br />
facts and d<strong>at</strong>a, it could be said th<strong>at</strong> generally ra<strong>in</strong> w<strong>at</strong>er and we<strong>at</strong>her<strong>in</strong>g<br />
reactions are the pr<strong>in</strong>cipal solute <strong>source</strong>s. Ra<strong>in</strong> w<strong>at</strong>er contributes to all<br />
pr<strong>in</strong>cipal solutes of dilute w<strong>at</strong>ers, except<strong>in</strong>g silica, but the most<br />
significant contribution will be to Na + , Cl − , SO 4 2− and HCO 3 − . The<br />
amounts of these solutes vary with the distance from seaw<strong>at</strong>er and<br />
pollution. Regard<strong>in</strong>g the we<strong>at</strong>her<strong>in</strong>g reactions, one of the most<br />
important is the congruent dissolution of soluble m<strong>in</strong>erals, e.g. gypsum<br />
or halite th<strong>at</strong> can load the solution with very high concentr<strong>at</strong>ion — this is<br />
a very important mechanism for recycl<strong>in</strong>g evaporites — and the silic<strong>at</strong>e<br />
hydrolysis. The alter<strong>at</strong>ion of feldspar to clay m<strong>in</strong>erals charges the w<strong>at</strong>ers<br />
with Na + ,HCO 3 − and silica. Other silic<strong>at</strong>es provide the additional c<strong>at</strong>ions<br />
Ca 2+ , Mg 2+ and K + but such w<strong>at</strong>ers are always dom<strong>in</strong><strong>at</strong>ed by<br />
bicarbon<strong>at</strong>e derived by <strong>at</strong>mosphere or from soil processes. The <strong>in</strong>crease<br />
<strong>in</strong> Ca 2+ ,Na + ,Mg 2+ versus HCO 3 − from the <strong>Pikrolimni</strong>/Ker<strong>at</strong>ea area<br />
(nr. 1,2,3,19) to boreholes (nr. 4 to 16) near <strong>Lake</strong> <strong>Pikrolimni</strong> seems to<br />
support silic<strong>at</strong>e hydrolysis as the we<strong>at</strong>her<strong>in</strong>g reactions th<strong>at</strong> contribute<br />
to the solute load of <strong>in</strong>flow w<strong>at</strong>er. However silic<strong>at</strong>e hydrolysis cannot be<br />
the only control on the w<strong>at</strong>er chemistry. Moreover, the critical control<br />
on the precipit<strong>at</strong>ion of trona, <strong>in</strong> <strong>Lake</strong> <strong>Pikrolimni</strong> w<strong>at</strong>ers, is the rel<strong>at</strong>ive<br />
amount of Ca 2+ and CO 3<br />
2−<br />
TOT ([CO 3 2− ]+[HCO 3 − ]+[H 2 CO 3 ], where<br />
bracketed symbols refer to concentr<strong>at</strong>ion) <strong>in</strong> the w<strong>at</strong>er to be<br />
evapor<strong>at</strong>ed. Accord<strong>in</strong>g to the conceptual model of Hardie–Eugster<br />
of evaporites form<strong>at</strong>ion (Hardie and Eugster, 1970), the first m<strong>in</strong>eral<br />
Fig. 2. Ternary diagrams of w<strong>at</strong>er samples from <strong>Pikrolimni</strong> area. White diamonds: Mg–HCO 3 groundw<strong>at</strong>ers; white circles: Na–HCO 3 groundw<strong>at</strong>ers; white squares: Ca–HCO 3<br />
groundw<strong>at</strong>ers; black diamonds: <strong>Pikrolimni</strong> <strong>Lake</strong> br<strong>in</strong>es.
138 E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
to form from evapor<strong>at</strong><strong>in</strong>g w<strong>at</strong>er is calcite, and calcite precipit<strong>at</strong>ion<br />
cont<strong>in</strong>ues until either Ca 2+ or CO 3 2− is exhausted. W<strong>at</strong>ers with a<br />
low proportion of CO 3<br />
2−<br />
TOT rel<strong>at</strong>ive to Ca 2+ will be carbon<strong>at</strong>e depleted<br />
after calcite form<strong>at</strong>ion, and will thus not yield Na-carbon<strong>at</strong>e<br />
m<strong>in</strong>erals upon further evapor<strong>at</strong>ion. So the elev<strong>at</strong>ed alkal<strong>in</strong>ity/2[Ca 2+ ]<br />
r<strong>at</strong>io is a ma<strong>in</strong> requirement, accord<strong>in</strong>g to Hardie and Eugster (1970),for<br />
the form<strong>at</strong>ion of significant quantities of Na-carbon<strong>at</strong>e m<strong>in</strong>erals.<br />
Also, accord<strong>in</strong>g to Hardie and Eugster (1970) the appearance of sepiolite<br />
and gypsum determ<strong>in</strong>es the evapor<strong>at</strong>ive sequence. So, after the sepiolite<br />
precipit<strong>at</strong>ion the w<strong>at</strong>er can become carbon<strong>at</strong>e-enriched and alkal<strong>in</strong>e<br />
earth-poor or vice versa. If alkal<strong>in</strong>ity is higher than 2[Ca 2+ +Mg 2+ ]<br />
then the sepiolite precipit<strong>at</strong>ion is not capable to modify the evolution of<br />
the solution along the p<strong>at</strong>h of alkal<strong>in</strong>e facies (Risacher, 1992). The<br />
significantly high proportion of CO 3<br />
2−<br />
TOT rel<strong>at</strong>ive to 2[Ca 2+ +Mg 2+ ]<br />
<strong>in</strong> <strong>Pikrolimni</strong> area w<strong>at</strong>ers causes them to move along the p<strong>at</strong>h,<br />
<strong>in</strong> the Hardie–Eugster model, of alkal<strong>in</strong>e faces, form<strong>in</strong>g Na–CO 3<br />
m<strong>in</strong>erals.<br />
However, because the r<strong>at</strong>io of alkal<strong>in</strong>ity to Ca 2+ and Mg 2+ controls<br />
the types of evaporite m<strong>in</strong>erals formed, and specially the deposition of<br />
trona m<strong>in</strong>eral, it is very important to determ<strong>in</strong>e the potential<br />
processes th<strong>at</strong> affected the HCO 3 − contents <strong>in</strong> these w<strong>at</strong>ers.<br />
5.2. Potential processes affected the concentr<strong>at</strong>ion of HCO 3<br />
−<br />
There are a number of processes th<strong>at</strong> can affect the HCO − 3 contents <strong>in</strong><br />
groundw<strong>at</strong>er [these w<strong>at</strong>ers are rich <strong>in</strong> HCO − 3 (about 1.50–3 g/L)].<br />
Eugster (1980) described <strong>in</strong> detail processes th<strong>at</strong> can affect the<br />
concentr<strong>at</strong>ion of alkal<strong>in</strong>ity <strong>in</strong> groundw<strong>at</strong>er: dissolution or precipit<strong>at</strong>ion<br />
of carbon<strong>at</strong>e m<strong>in</strong>erals, chemical we<strong>at</strong>her<strong>in</strong>g of silic<strong>at</strong>es, redox reactions,<br />
especially the reduction of NO − 3 to NH + 4 and SO 2− 4 to HS − , microbial<br />
respir<strong>at</strong>ion or anaerobic decay and the conversion of CO 2 of deep orig<strong>in</strong><br />
to HCO − 3 <strong>in</strong> the aquifer. The dissolution/precipit<strong>at</strong>ion of carbon<strong>at</strong>e<br />
m<strong>in</strong>eral is a common control of HCO − 3 contents <strong>in</strong> groundw<strong>at</strong>er. These<br />
w<strong>at</strong>ers show an important over-s<strong>at</strong>ur<strong>at</strong>ion with respect to calcite<br />
(Parkhurst and Appelo, 1999). In Fig. 3a the rel<strong>at</strong>ionship Ca 2+ versus<br />
alkal<strong>in</strong>ity (HCO − 3 +2CO 2− 3 ) is shown. For the Mg–HCO 3 -type w<strong>at</strong>ers it is<br />
observed th<strong>at</strong>, from <strong>Pikrolimni</strong> village (Na–HCO 3 ) to boreholes near<br />
<strong>Pikrolimni</strong> <strong>Lake</strong> (Mg–HCO 3 ), <strong>in</strong>crease of HCO 3 contents takes place: the<br />
Mg–HCO 3 groundw<strong>at</strong>er th<strong>at</strong> was sampled with<strong>in</strong> the Holocene<br />
lacustr<strong>in</strong>e sediments of <strong>Pikrolimni</strong> has HCO − 3 contents higher than<br />
38 mmol/L. It is also observed th<strong>at</strong> the HCO − 3 <strong>in</strong>creases with rel<strong>at</strong>ively<br />
slight change <strong>in</strong> Ca 2+ , suggest<strong>in</strong>g th<strong>at</strong> carbon<strong>at</strong>e m<strong>in</strong>eral dissolution/<br />
precipit<strong>at</strong>ion reactions are not an important control on HCO − 3 .Alsothe<br />
absence of chloride <strong>in</strong> the soda and Mg–HCO 3 type w<strong>at</strong>ers suggest th<strong>at</strong><br />
the dissolution of mar<strong>in</strong>e carbon<strong>at</strong>es is not the only <strong>source</strong> of the carbon.<br />
Furthermore, these w<strong>at</strong>ers are plotted to the right of the 2(Ca 2+ )=<br />
(HCO − 3 ) equilibrium l<strong>in</strong>e (Fig. 3a). If the dissolution of the calcite was the<br />
dom<strong>in</strong>ant process produc<strong>in</strong>g CO 2 <strong>in</strong> the system then these w<strong>at</strong>ers would<br />
plot along the 2(Ca 2+ )=(HCO − 3 ) equilibrium l<strong>in</strong>e.<br />
Chemical we<strong>at</strong>her<strong>in</strong>g of silic<strong>at</strong>es is another major <strong>in</strong>fluence on<br />
groundw<strong>at</strong>er. The Na/HCO 3 r<strong>at</strong>io of dilute Na–HCO 3 <strong>in</strong>flow is very close<br />
to unity. When accompanied by the absence of Ca–HCO 3 groundw<strong>at</strong>er,<br />
this shows th<strong>at</strong> these Na–HCO 3 w<strong>at</strong>ers do not evolve from Ca–HCO 3<br />
groundw<strong>at</strong>er th<strong>at</strong> undergoes normal ion exchange. Therefore the r<strong>at</strong>io<br />
Na/HCO 3 , very close to unity, suggests th<strong>at</strong> the chemistry of soda w<strong>at</strong>er<br />
is controlled by simple we<strong>at</strong>her<strong>in</strong>g reactions like Na-feldspars:<br />
Na–AlSi 3 O 8 þ H 2 CO 3 þ 4; 5H 2 O→Na þ þ HCO − 3 þ 2H 4 SiO 4<br />
þ 1=2Al 2 Si 2 O 5 ðOHÞ 4<br />
:<br />
The hydrolysis of silic<strong>at</strong>e m<strong>in</strong>erals takes place rapidly produc<strong>in</strong>g<br />
bicarbon<strong>at</strong>e rich w<strong>at</strong>er with high silica content (sample Ker<strong>at</strong>ea,<br />
nr. 21a). Probably this fresh w<strong>at</strong>er is from the deep cell of the local<br />
groundw<strong>at</strong>er system and probably circul<strong>at</strong>es <strong>in</strong> the bedrock. On the<br />
contrary, the Na/HCO 3 r<strong>at</strong>io of dilute Mg–Na–HCO 3 <strong>in</strong>flow (Fig. 3b)<br />
Fig. 3. Ca 2+ versus alkal<strong>in</strong>ity of all groundw<strong>at</strong>ers (a) and Na + versus HCO 3 − of Mg–HCO 3<br />
groundw<strong>at</strong>ers (b). Same symbols as Fig. 2.<br />
<strong>in</strong>dic<strong>at</strong>es th<strong>at</strong> not only silic<strong>at</strong>e alter<strong>at</strong>ion but also more processes are<br />
<strong>in</strong>volved <strong>in</strong> the chemistry of these w<strong>at</strong>ers. The correl<strong>at</strong>ion between<br />
HCO 3 − and Mg 2+ <strong>in</strong>dic<strong>at</strong>es th<strong>at</strong> basic rocks become a major target of<br />
we<strong>at</strong>her<strong>in</strong>g reactions <strong>in</strong> the Mg–HCO 3 − w<strong>at</strong>ers. In summary, as referred<br />
by Eugster (1980), the composition of <strong>in</strong>flow w<strong>at</strong>er depends largely on<br />
the m<strong>in</strong>erals present <strong>in</strong> the w<strong>at</strong>ershed: basic and ultrabasic rocks are<br />
probable to give Mg-HCO 3 w<strong>at</strong>ers.<br />
Redox reactions are an additional possible control on HCO 3 − <strong>in</strong><br />
particular the reduction of NO 3 − to NH 4 + and SO 4 2− to HS − ,bothofwhich<br />
produce CO 2 . So, concern<strong>in</strong>g the reduction reactions as a possible control<br />
of HCO 3 − , although Eh was not measured, the high SO 4 2− and NO 3 − contents<br />
<strong>in</strong> these w<strong>at</strong>ers suggest th<strong>at</strong> these reactions are not likely to be occurr<strong>in</strong>g.<br />
Also the microbial respir<strong>at</strong>ion or anaerobic decay process (potential<br />
<strong>source</strong>s of CO 2 th<strong>at</strong> could be converted to HCO 3 − ) appears unlikely to<br />
apply significant control on the ground w<strong>at</strong>er chemistry because the<br />
m<strong>at</strong>erial th<strong>at</strong> would <strong>in</strong>dic<strong>at</strong>e large accumul<strong>at</strong>ion of organic m<strong>at</strong>ter <strong>in</strong> the<br />
borehole logs is absent.<br />
The f<strong>in</strong>al major reaction th<strong>at</strong> could <strong>in</strong>fluence the HCO 3 − contents <strong>in</strong><br />
the deep aquifer is the addition <strong>in</strong> the system of CO 2 of deep orig<strong>in</strong>. The<br />
pCO 2 values of the most Mg–Na–HCO 3 type w<strong>at</strong>ers (10 − 0.5 –10 − 1 )<br />
are higher than the <strong>at</strong>mospheric value (10 − 3.5 ) and the value of the<br />
soil (10 − 2.5 –10 − 1.5 ), suggest<strong>in</strong>g th<strong>at</strong> <strong>in</strong>put of <strong>at</strong>mospheric/soil is not a<br />
major <strong>source</strong> and evidenc<strong>in</strong>g th<strong>at</strong> gas is be<strong>in</strong>g added to the aquifer<br />
from the deep. The highest values are observed for the Mg–HCO 3 type<br />
w<strong>at</strong>ers emerg<strong>in</strong>g <strong>in</strong> the Holocene lacustr<strong>in</strong>e sediments of <strong>Lake</strong><br />
<strong>Pikrolimni</strong>. These w<strong>at</strong>ers present also the highest Na + ,Mg 2+ and<br />
HCO 3 − contents, from <strong>Pikrolimni</strong> area (<strong>Pikrolimni</strong> Village) to <strong>Lake</strong><br />
<strong>Pikrolimni</strong> (borehole near <strong>Pikrolimni</strong>), suggest<strong>in</strong>g probably th<strong>at</strong> basic<br />
m<strong>in</strong>eral hydrolysis driven by <strong>in</strong>jection of deep (mantle or metamorphism<br />
<strong>source</strong>) CO 2 is act<strong>in</strong>g to control the concentr<strong>at</strong>ion of HCO 3 − <strong>in</strong><br />
the lacustr<strong>in</strong>e bas<strong>in</strong>.
E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
139<br />
5.3. Br<strong>in</strong>e evolution<br />
The plot of Na + ,SO 4 2− ,CO 3 2− ,HCO 3 − versus Cl − contents of the<br />
borehole sample <strong>in</strong> the thermal spa of <strong>Lake</strong> <strong>Pikrolimni</strong> and the doma<strong>in</strong> of<br />
br<strong>in</strong>e from <strong>Lake</strong> <strong>Pikrolimni</strong> are illustr<strong>at</strong>ed <strong>in</strong> Fig. 4. In these diagrams we<br />
also put published d<strong>at</strong>a for <strong>Lake</strong> N<strong>at</strong>ron <strong>in</strong> Tanzania (Gueddari, 1984)<br />
and <strong>Lake</strong> Magadi <strong>in</strong> Kenya (Gueddari, 1984), where the <strong>n<strong>at</strong>ron</strong> actually<br />
precipit<strong>at</strong>es and for Wadi N<strong>at</strong>run which was the <strong>n<strong>at</strong>ron</strong> <strong>source</strong> <strong>in</strong><br />
antiquity (Taher, 1999). Accord<strong>in</strong>g to all diagrams, it is concluded th<strong>at</strong><br />
the chemical composition of <strong>Lake</strong> <strong>Pikrolimni</strong>'s w<strong>at</strong>er is similar to th<strong>at</strong> of<br />
the lakes th<strong>at</strong> precipit<strong>at</strong>e <strong>n<strong>at</strong>ron</strong> today. Such solute concentr<strong>at</strong>ions<br />
support th<strong>at</strong> the evapor<strong>at</strong>ive concentr<strong>at</strong>ion is the dom<strong>in</strong>ant process <strong>in</strong><br />
the chemical evolution of these lake w<strong>at</strong>ers. The Na–Cl rel<strong>at</strong>ion (Fig. 4a)<br />
illustr<strong>at</strong>es <strong>in</strong> general the stability of Na/Cl r<strong>at</strong>io <strong>in</strong> the lake br<strong>in</strong>e. The Na/<br />
Cl r<strong>at</strong>io of <strong>Lake</strong> N<strong>at</strong>ron and <strong>Lake</strong> Magadi, as well as of Wadi N<strong>at</strong>run which<br />
was <strong>n<strong>at</strong>ron</strong> <strong>source</strong> <strong>in</strong> antiquity, rema<strong>in</strong>s stable between diluted w<strong>at</strong>ers<br />
and concentr<strong>at</strong>ed br<strong>in</strong>es. Compar<strong>at</strong>ively to the w<strong>at</strong>ers of <strong>Lake</strong> N<strong>at</strong>ron<br />
(Na/Cl r<strong>at</strong>io=3.3) and Magadi (Na/Cl r<strong>at</strong>io=2), the w<strong>at</strong>er of <strong>Lake</strong><br />
<strong>Pikrolimni</strong> presents a lower Na/Cl r<strong>at</strong>io (Na/Cl r<strong>at</strong>io from 1.2 to 1.7),<br />
almost two times. The Na/Cl r<strong>at</strong>ios of <strong>Lake</strong> N<strong>at</strong>ron, <strong>Lake</strong> Magadi and<br />
Wadi N<strong>at</strong>run tend to unity, close to <strong>Pikrolimni</strong> br<strong>in</strong>es, only <strong>in</strong> the most<br />
concentr<strong>at</strong>e br<strong>in</strong>es (Gueddari, 1984) show<strong>in</strong>g a very small devi<strong>at</strong>ion<br />
from the evapor<strong>at</strong>ive trend (from diluted to br<strong>in</strong>e w<strong>at</strong>ers), until with<strong>in</strong><br />
the concentr<strong>at</strong>ion range appropri<strong>at</strong>e for trona s<strong>at</strong>ur<strong>at</strong>ion. Once trona<br />
precipit<strong>at</strong>es, the Na + extraction from the solution enriches the br<strong>in</strong>e <strong>in</strong><br />
chloride and thus, Na/Cl r<strong>at</strong>io decreases. In fact <strong>in</strong> the diagram log Na/Cl<br />
versus Cl (Fig. 4b) the values of diluted w<strong>at</strong>ers and br<strong>in</strong>es from <strong>Lake</strong><br />
Magadi are given as well as the correspond<strong>in</strong>g d<strong>at</strong>a from <strong>Lake</strong> <strong>Pikrolimni</strong>.<br />
As precisely referred by Eugster and Jones (1979), “from groundw<strong>at</strong>ers<br />
to all but the most concentr<strong>at</strong>es br<strong>in</strong>es Na/Cl rema<strong>in</strong>s constant, because<br />
no fraction<strong>at</strong>ion occur. The most concentr<strong>at</strong>ed br<strong>in</strong>es are s<strong>at</strong>ur<strong>at</strong>ed<br />
with respect to trona and precipit<strong>at</strong>ion clearly leads to a decrease <strong>in</strong><br />
Na/Cl”. In <strong>Lake</strong> Pikrol<strong>in</strong>i the most concentr<strong>at</strong>ed br<strong>in</strong>e (8/2002)<br />
shows a decrease <strong>in</strong> Na/Cl <strong>in</strong> rel<strong>at</strong>ion to the most groundw<strong>at</strong>ers th<strong>at</strong><br />
are rel<strong>at</strong>ed to trona precipit<strong>at</strong>ion. Between groundw<strong>at</strong>er and lake<br />
br<strong>in</strong>es, a decrease of Ca/Cl, Mg/Cl and HCO 3 /Cl r<strong>at</strong>ios is observed<br />
(Table 1). Carbon<strong>at</strong>e species are subject to removal from solution by<br />
different mechanisms. In Fig. 4c, the gre<strong>at</strong>er than 1:1 slope of the<br />
regression l<strong>in</strong>e for Cl − versus CO 3 2− +HCO 3 − contents illustr<strong>at</strong>es the<br />
gradual loss of carbon<strong>at</strong>e species, accompany<strong>in</strong>g evapor<strong>at</strong>ive concentr<strong>at</strong>ion<br />
of <strong>Lake</strong> Magadi (Jones et al., 1977). The same is observed for the<br />
w<strong>at</strong>ers from <strong>Pikrolimni</strong> area. This loss is <strong>at</strong>tributed to equilibr<strong>at</strong>ion with<br />
the <strong>at</strong>mosphere [2HCO 3 − (aq) → CO 3 2− (aq)+CO 2 (gas)+H 2 O], calcite<br />
precipit<strong>at</strong>ion from diluted w<strong>at</strong>ers [Ca 2+ (aq)+HCO 3 − (aq) → CaCO 3<br />
(sol)+H + ] and trona precipit<strong>at</strong>ion from s<strong>at</strong>ur<strong>at</strong>ed br<strong>in</strong>e [3Na + (aq)+<br />
HCO 3 − (aq)+CO 3 2− (aq)+2H 2 O → Na 2 CO 3 ,NaHCO 3 ,2H 2 O (sol)]. In<br />
most diluted lake w<strong>at</strong>ers, the carbon<strong>at</strong>e loss is probably rel<strong>at</strong>ed to both<br />
degass<strong>in</strong>g and carbon<strong>at</strong>e precipit<strong>at</strong>ion. Loss to <strong>at</strong>mosphere is expected<br />
for pCO 2 gre<strong>at</strong>er than 10 −3.5 <strong>at</strong>m (Jones et al., 1977). In fact, <strong>in</strong> these<br />
diluted w<strong>at</strong>ers the pCO 2 is lower than 10 −3.2 . In the lake br<strong>in</strong>es, the<br />
Fig. 4. Na + versus Cl − (a), log (Na/Cl) versus log Cl (b), Cl − versus CO 3 2− +HCO 3 − (c), and SO 4 2− versus Cl − (d). Black diamonds: <strong>Pikrolimni</strong> br<strong>in</strong>es; white diamonds: groundw<strong>at</strong>ers;<br />
black circles: <strong>Lake</strong> N<strong>at</strong>ron spr<strong>in</strong>gs (Gueddari, 1984); white circles: <strong>Lake</strong> N<strong>at</strong>ron br<strong>in</strong>es (Gueddari, 1984); crosses: Wadi N<strong>at</strong>run (Shortland, 2004); white squares: <strong>Lake</strong> Magadi<br />
(Gueddari, 1984; Jones et al., 1977); white triangles: Wadi N<strong>at</strong>run (Taher, 1999).
140 E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
carbon<strong>at</strong>e loss is probably rel<strong>at</strong>ed to calcium and sodium carbon<strong>at</strong>e<br />
precipit<strong>at</strong>ion. An additional mechanism is also associ<strong>at</strong>ed with efflorescent<br />
crust (Eugster, 1966). These crusts are the products of complete<br />
desicc<strong>at</strong>ion of lake w<strong>at</strong>er and consist predom<strong>in</strong>antly of soluble salts.<br />
When ra<strong>in</strong> or dilute runoff comes <strong>in</strong> contact with these crusts only the<br />
most soluble salts, such chlorides, carbon<strong>at</strong>es and sulf<strong>at</strong>es are resolved<br />
<strong>in</strong>to solution and <strong>in</strong>creas<strong>in</strong>g the solute load, while the less soluble phases<br />
like the alkal<strong>in</strong>e earth carbon<strong>at</strong>es and silica rema<strong>in</strong> <strong>in</strong> the crust. The<br />
dissolution of these crusts leads to high alkal<strong>in</strong>e w<strong>at</strong>er with a specifically<br />
lower HCO 3 /CO 3 r<strong>at</strong>io. The high pH (9.6) of some w<strong>at</strong>ers of <strong>Lake</strong><br />
<strong>Pikrolimni</strong> (like PL-1/2003) can be <strong>at</strong>tributed to recycl<strong>in</strong>g of residual<br />
br<strong>in</strong>es (PL-8/2002) from the fresh w<strong>at</strong>er. The f<strong>in</strong>d<strong>in</strong>g of abundant calcite<br />
precipit<strong>at</strong>ion <strong>in</strong> lake beds th<strong>at</strong> are covered by crusts, consist<strong>in</strong>g of trona,<br />
burkeite, thenardite and halite m<strong>in</strong>erals, as shown by the XRD analysis<br />
(Table 2) of <strong>Pikrolimni</strong> crusts, shows th<strong>at</strong> the above play a competitive<br />
role <strong>in</strong> the geochemistry processes of <strong>Pikrolimni</strong> br<strong>in</strong>es. In fact, salt<br />
analyses from different parts of the lake showed th<strong>at</strong> they consist of<br />
carbon<strong>at</strong>es (calcite, dolomite), trona, burckeite and halite. Salts from<br />
Wadi N<strong>at</strong>run show similar m<strong>in</strong>eral composition (Shortland et al., 2006).<br />
In the graph of SO 4 2− versus Cl − (Fig. 4d), the molar SO 4 2− /Cl − r<strong>at</strong>io<br />
<strong>in</strong>dic<strong>at</strong>es generally neither loss nor ga<strong>in</strong> of SO 4 2− between diluted and<br />
br<strong>in</strong>e w<strong>at</strong>ers. The rel<strong>at</strong>ions SO 4 2− versus Cl − suggest a possible<br />
<strong>in</strong>cipient precipit<strong>at</strong>ion of a sulph<strong>at</strong>e m<strong>in</strong>eral only for PL-8/2002. All<br />
the other lake w<strong>at</strong>ers <strong>in</strong>dic<strong>at</strong>e an excellent conserv<strong>at</strong>ive behaviour of<br />
sulph<strong>at</strong>e rel<strong>at</strong>ive to chloride. This also suggests th<strong>at</strong> sulf<strong>at</strong>e reduction<br />
is not a major process <strong>at</strong> <strong>Lake</strong> <strong>Pikrolimni</strong>.<br />
6. <strong>Geochemical</strong> model<br />
6.1. Thermodynamic calcul<strong>at</strong>ion model for groundw<strong>at</strong>ers and lake w<strong>at</strong>er<br />
An ion-specific <strong>in</strong>teraction model based on Pitzer's equ<strong>at</strong>ions has been<br />
used (EQL/EVP by Risacher and Clement, 2001) to calcul<strong>at</strong>e the s<strong>at</strong>ur<strong>at</strong>ion<br />
<strong>in</strong>dex versus different solid phases, of the <strong>Pikrolimni</strong> w<strong>at</strong>ers and br<strong>in</strong>es.<br />
The lake br<strong>in</strong>e samples are put <strong>in</strong> ionic strength diagram versus the<br />
s<strong>at</strong>ur<strong>at</strong>ion <strong>in</strong>dex (Fig. 5a, b, c) and <strong>in</strong> Table 3 the m<strong>in</strong>erals th<strong>at</strong> are<br />
supers<strong>at</strong>ur<strong>at</strong>ed <strong>in</strong> the lakes <strong>Pikrolimni</strong>, Magadi, N<strong>at</strong>ron and Deep<br />
Spr<strong>in</strong>gs are given. The calcul<strong>at</strong>ion of s<strong>at</strong>ur<strong>at</strong>ion <strong>in</strong>dex versus different<br />
carbon<strong>at</strong>e and/or bicarbon<strong>at</strong>e sodium m<strong>in</strong>erals shows th<strong>at</strong> the borehole<br />
w<strong>at</strong>ers are s<strong>at</strong>ur<strong>at</strong>ed <strong>in</strong> calcium m<strong>in</strong>erals. The lake w<strong>at</strong>ers are s<strong>at</strong>ur<strong>at</strong>ed<br />
<strong>in</strong> Na/Ca carbon<strong>at</strong>e m<strong>in</strong>erals and only the lake br<strong>in</strong>es are s<strong>at</strong>ur<strong>at</strong>ed also<br />
<strong>in</strong> Na–carbon<strong>at</strong>e and bicarbon<strong>at</strong>e, Na–sulf<strong>at</strong>e and Na–chloride m<strong>in</strong>erals.<br />
Indeed, the br<strong>in</strong>es of PL-8/2002 and PL-9/2006 are s<strong>at</strong>ur<strong>at</strong>ed <strong>in</strong><br />
sodium carbon<strong>at</strong>e–bicarbon<strong>at</strong>e m<strong>in</strong>erals like nahcolite, trona and <strong>n<strong>at</strong>ron</strong>.<br />
Τhe calcul<strong>at</strong>ion of ionic strengths versus different sodium–sulf<strong>at</strong>e<br />
and sodium–chloride m<strong>in</strong>erals show th<strong>at</strong> the br<strong>in</strong>es with Ι>6 are<br />
s<strong>at</strong>ur<strong>at</strong>ed <strong>in</strong> halite (NaCl), thenardite (Na 2 SO 4 ) and mirabilite (Na 2-<br />
SO 4 ⁎10H 2 O) confirm<strong>in</strong>g th<strong>at</strong> <strong>in</strong> such lakes, the Ca–Na–HCO 3 salts<br />
precipit<strong>at</strong>e first and the SO 4 –HCO 3 salts follow. The results of<br />
measurement of Na + and Cl − show <strong>in</strong>deed th<strong>at</strong> only the s<strong>at</strong>ur<strong>at</strong>ion<br />
<strong>in</strong> halite, thenardite and mirabilite is achieved.<br />
6.2. Simul<strong>at</strong>ion of evapor<strong>at</strong>ion of groundw<strong>at</strong>ers<br />
Us<strong>in</strong>g geochemical program (EQL/EVP by Risacher and Clement,<br />
2001), it is possible to <strong>in</strong>vestig<strong>at</strong>e the m<strong>in</strong>eralogical composition of<br />
the m<strong>in</strong>erals th<strong>at</strong> precipit<strong>at</strong>e from the w<strong>at</strong>er, with different <strong>in</strong>itial<br />
chemical compositions, as they evapor<strong>at</strong>e. All sampled w<strong>at</strong>ers from<br />
boreholes and the spr<strong>in</strong>g were evapor<strong>at</strong>ed <strong>in</strong> order to check if sodium<br />
carbon<strong>at</strong>es could precipit<strong>at</strong>e by evapor<strong>at</strong>ion of these w<strong>at</strong>ers theoretically.<br />
As shown <strong>in</strong> Table 4, most w<strong>at</strong>ers can give nahcolite (NaHCO 3 )<br />
and thenardite (Na 2 SO 4 ) precipit<strong>at</strong>es <strong>in</strong> the last phase of evapor<strong>at</strong>ion.<br />
Ιn Table 4, the evapor<strong>at</strong>ed samples th<strong>at</strong> can give trona and <strong>n<strong>at</strong>ron</strong><br />
precipit<strong>at</strong>es are shown. In the same table, a diluted w<strong>at</strong>er from <strong>Lake</strong><br />
<strong>Pikrolimni</strong> (PL-5/2003) is also shown. It is observed th<strong>at</strong> three<br />
samples, S/1/2002; b/2/2002; PL-5/2003, give Na–carbon<strong>at</strong>e–bicarbon<strong>at</strong>e<br />
m<strong>in</strong>erals [gaylussite Na 2 Ca⁎(CO 3 ) 2 ⁎5H 2 O; pirssonite Na 2-<br />
Ca⁎ (CO 3 ) 2 ⁎ 2H 2 O; trona Na 2 CO 3 ⁎ NaHCO 3 ⁎ 2H 2 O and <strong>n<strong>at</strong>ron</strong><br />
Na 2 CO 3 ⁎10H 2 O], Na–sulf<strong>at</strong>e m<strong>in</strong>erals (burkeite Na 2 CO 3 ⁎2Na 2 SO 4 ;<br />
thenardite Na 2 SO 4 ) and Na–chloride m<strong>in</strong>erals [halite NaCl]. W<strong>at</strong>er<br />
samples S/1/2002, b/2/2002 and PL-5/2003 give similar m<strong>in</strong>erals<br />
with those found by XRD analysis <strong>in</strong> collected salt samples but also<br />
similar m<strong>in</strong>erals referred <strong>in</strong> liter<strong>at</strong>ure (Shortland et al., 2006; Taher,<br />
1999; Gueddari, 1984). However, only the w<strong>at</strong>er sample coded S/1/<br />
2002 and b/2/2002, Na–ΗCO 3 type, could give <strong>n<strong>at</strong>ron</strong> and trona.<br />
Fig. 6 shows the results of S/1/2002 sample when the <strong>in</strong>itial<br />
chemical composition of this w<strong>at</strong>er was modeled. As it can be seen, as<br />
the w<strong>at</strong>er is concentr<strong>at</strong>ed about 250 times of the <strong>in</strong>itial chemical<br />
composition, the first Na–carbon<strong>at</strong>e m<strong>in</strong>eral, gaylussite, deigns to<br />
precipit<strong>at</strong>e. After a concentr<strong>at</strong>ion of 395 times of the <strong>in</strong>itial chemical<br />
composition, <strong>n<strong>at</strong>ron</strong>, pirssonite, and mirabilite precipit<strong>at</strong>e. The gaylussite<br />
and the pirssonite are among the first salts th<strong>at</strong> precipit<strong>at</strong>eand their<br />
form<strong>at</strong>ion is rel<strong>at</strong>ed to calcite. The gaylussite [Na 2 Ca⁎(CO 3 ) 2 ⁎5H 2 O]<br />
generally forms by the reaction of primary calcite (CaCO 3 )andthesodic<br />
carbon<strong>at</strong>e solution, accord<strong>in</strong>g to the reaction<br />
CaCO 3 þ 2Na þ CO 3 þ 5H 2 O↔Na 2 Ca⁎ðCO 3 Þ 2<br />
⁎5H 2 O:<br />
The pirssonite is usually found <strong>in</strong> the presence of gaylussite or of<br />
calcite crystals. The direct precipit<strong>at</strong>ion of pirssonite takes place<br />
accord<strong>in</strong>g to the reaction<br />
Na 2 Ca⁎ðCO 3 Þ 2<br />
⁎2H 2 O↔2Na þ Ca þ 2CO 3 þ 2H 2 O:<br />
Secondarily, pirssonite forms either by the dehydr<strong>at</strong>ion of<br />
gaylussite or by the reaction of Ca–Na solutions with calcite accord<strong>in</strong>g<br />
to the follow<strong>in</strong>g equ<strong>at</strong>ions<br />
Na 2 Ca⁎ðCO 3 Þ 2<br />
⁎5H 2 O↔Na 2 Ca⁎ðCO 3 Þ 2<br />
⁎2H 2 O þ 3H 2 O<br />
Table 2<br />
XRD results on evaporites from different lakes.<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> 2002 Wadi N<strong>at</strong>run a <strong>Lake</strong> N<strong>at</strong>ron, Tanzania b <strong>Lake</strong> Magadi, Kenya b The Deep Spr<strong>in</strong>gs <strong>Lake</strong>, California b<br />
Calcite (CaCO 3 )<br />
Dolomite c Calcite (CaCO 3 ) Calcite (CaCO 3 )<br />
Dolomite [Ca(Mg)CO 3 )<br />
Pirsonnite Na 2 Ca⁎(CO 3 ) 2 ⁎2H 2 O Pirsonnite Na 2 Ca⁎(CO 3 ) 2 ⁎2H 2 O<br />
Nahcolite (NaHCO 3 )<br />
Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O) Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O) Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O) Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O) Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O)<br />
Thermon<strong>at</strong>rite (Na 2 CO 3 ⁎H 2 O), Thermon<strong>at</strong>rite (Na 2 CO 3 ⁎H 2 O),<br />
Burkeite (Na 2 CO 3 ⁎2Na 2 SO 4 ) Burkeite (Na 2 CO 3 ⁎2Na 2 SO 4 Burkeite (Na 2 CO 3 ⁎2Na 2 SO 4 )<br />
Kogarkoite (Na 2 SO 4 ⁎NaF)<br />
Thenardite (Na 2 SO 4 ) not detected Thenardite (Na 2 SO 4 ) Thenardite (Na 2 SO 4<br />
<strong>in</strong> 2006 salt<br />
Halite (NaCl) Halite (NaCl) Halite (NaCl) Halite (NaCl) Halite (NaCl)<br />
a Shortland A.J. (2004).<br />
b Gueddari M. (1984).<br />
c Wenigswieser (1992).
E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
141<br />
64% of all salts th<strong>at</strong> precipit<strong>at</strong>e, while halite only 10%. On the contrary,<br />
Fig. 6 shows the results of PL-5/2003 sample when the <strong>in</strong>itial chemical<br />
composition of this w<strong>at</strong>er was modeled. It seems th<strong>at</strong> when the model<br />
stops halite is more abundant <strong>in</strong> the precipit<strong>at</strong>es than trona: trona<br />
represents only 0.5% of all salts th<strong>at</strong> precipit<strong>at</strong>e, while halite 60%.<br />
These two models show the change which may have happened <strong>in</strong> the<br />
w<strong>at</strong>er chemistry of the lake's spr<strong>in</strong>gs and why trona is so hard to f<strong>in</strong>d<br />
<strong>in</strong> <strong>Lake</strong> <strong>Pikrolimni</strong> when the models reaches their lowest po<strong>in</strong>t: trona<br />
salts are hoarded by the deposition of large quantities of halite, as<br />
shown by Shortland et al. (2006) <strong>in</strong> their study about Wadi N<strong>at</strong>run.<br />
The aliment<strong>at</strong>ion of the lake <strong>in</strong> the past had happened ma<strong>in</strong>ly from<br />
deep spr<strong>in</strong>gs with similar chemistry while today is arisen ma<strong>in</strong>ly from<br />
meteoric w<strong>at</strong>ers and irrig<strong>at</strong>ion groundw<strong>at</strong>ers (Pl<strong>in</strong>y; Many public<br />
testimonies vouch th<strong>at</strong> there were a lot of spr<strong>in</strong>gs <strong>in</strong> the middle of<br />
the lake, but after the big earthquake of 1981 these spr<strong>in</strong>gs have<br />
disappeared).<br />
The chemical composition of the residual w<strong>at</strong>er and of three samples<br />
(S/1/2002; b/2/2002; PL-5/2003) after the simul<strong>at</strong>ion of evapor<strong>at</strong>ion<br />
was similar with the chemical composition of the lake's w<strong>at</strong>er <strong>in</strong><br />
conditions of <strong>in</strong>tense evapor<strong>at</strong>ion (Table 5).<br />
Especially the chemical composition of the residual w<strong>at</strong>er, coded<br />
S/1/2002, after the evapor<strong>at</strong>ion simul<strong>at</strong>ion, is similar to th<strong>at</strong> of the<br />
lake br<strong>in</strong>e PL-9/2006 (Table 5). Furthermore, the model shows th<strong>at</strong><br />
<strong>n<strong>at</strong>ron</strong> starts to precipit<strong>at</strong>e when Cl − content is 1417 mmol/L [when<br />
the concentr<strong>at</strong>ion factor (CF) is about 395 times: 395⁎3.8 mmol/L<br />
(<strong>in</strong>itial Cl of s/1/2002)=1500 mmol/L] while trona starts to precipit<strong>at</strong>e<br />
when Cl − content is approxim<strong>at</strong>ely 2500 mmol/L [when the<br />
concentr<strong>at</strong>ion factor (CF) is about 720 times: 700⁎3.8 mmol/L (<strong>in</strong>itial Cl<br />
of s/1/2002)=2651 mmol/L]. The lake w<strong>at</strong>er reaches similar concentr<strong>at</strong>ions<br />
only dur<strong>in</strong>g the summer, August (but not every summer), as<br />
observed by sampl<strong>in</strong>g. Accord<strong>in</strong>g to Pl<strong>in</strong>y, <strong>n<strong>at</strong>ron</strong> was collected <strong>in</strong><br />
Macedonia only <strong>in</strong> the ris<strong>in</strong>g of the Dog Star for n<strong>in</strong>e days. This tim<strong>in</strong>g<br />
places this event <strong>at</strong> the beg<strong>in</strong>n<strong>in</strong>g of the hottest part of the summer,<br />
when the lake is found <strong>at</strong> the peak of evapor<strong>at</strong>ion and trona starts to<br />
precipit<strong>at</strong>e. However, the lake does not reach such high concentr<strong>at</strong>ions<br />
every year, a fact th<strong>at</strong> is due to the we<strong>at</strong>her (temper<strong>at</strong>ure,<br />
humidity, precipit<strong>at</strong>ion). Therefore, the collection would have been<br />
tak<strong>in</strong>g place only dur<strong>in</strong>g particularly hot and dry summers.<br />
7. Conceptual model<br />
Fig. 5. S<strong>at</strong>ur<strong>at</strong>ion Index (Log s<strong>at</strong>ur<strong>at</strong>ion r<strong>at</strong>io) versus ionic strength (logarithmic). White<br />
triangles: nahcolite; x: gaylussite; –: pirssonite; white diamonds: <strong>n<strong>at</strong>ron</strong>; black<br />
squares: trona; white circles: halite; crosses: mirabilite; white squares: thenardite;<br />
crossed x: thermon<strong>at</strong>rite.<br />
CaCO 3 þ 2Na þ CO 3 þ 2H 2 O↔Na 2 Ca⁎ðCO 3 Þ 2<br />
⁎2H 2 O:<br />
Also when the concentr<strong>at</strong>ion factor (CF) is about 700 times of the<br />
<strong>in</strong>itial chemical composition, burkeite and trona precipit<strong>at</strong>ion <strong>in</strong>iti<strong>at</strong>es.<br />
Much l<strong>at</strong>er, when the w<strong>at</strong>er is concentr<strong>at</strong>ed by about a factor of<br />
1200, halite beg<strong>in</strong>s to precipit<strong>at</strong>e. When the models stop, halite and<br />
trona are the more abundant precipit<strong>at</strong>es. Trona is more abundant <strong>in</strong><br />
the precipit<strong>at</strong>es than halite: trona (50%) and <strong>n<strong>at</strong>ron</strong> (14%) represent<br />
The Na/Cl r<strong>at</strong>io <strong>in</strong> <strong>Lake</strong> <strong>Pikrolimni</strong> supports evapor<strong>at</strong>ive concentr<strong>at</strong>ions<br />
as the dom<strong>in</strong>ant process <strong>in</strong> the chemical evolution of these lake<br />
w<strong>at</strong>ers th<strong>at</strong> have high proportions of Na + and HCO 3 − . The derived high<br />
Na + and HCO 3 − contents are due to silic<strong>at</strong>e hydrolysis. The comput<strong>in</strong>g<br />
program has shown th<strong>at</strong> the dilute spr<strong>in</strong>g <strong>in</strong>flow was concentr<strong>at</strong>ed (by<br />
the assumption of Cl − conserv<strong>at</strong>ion <strong>in</strong> solution throughout evapor<strong>at</strong>ion)<br />
about 12000 times to obta<strong>in</strong> the most sal<strong>in</strong>e residual actual lake w<strong>at</strong>er,<br />
PL-8/2002; PL-9/2006). However, this evapor<strong>at</strong>ion budget calcul<strong>at</strong>ion<br />
(Table 5), if compared with the actual analysis of the most sal<strong>in</strong>e<br />
<strong>Pikrolimni</strong> br<strong>in</strong>e, shows th<strong>at</strong> some constituents are not conserv<strong>at</strong>ive,<br />
due to different removal mechanisms. Therefore, it is believed th<strong>at</strong> <strong>in</strong> the<br />
past, the solute composition of such w<strong>at</strong>ers, which are controlled by<br />
silic<strong>at</strong>e hydrolysis and evapor<strong>at</strong>ive concentr<strong>at</strong>ion, gave the <strong>n<strong>at</strong>ron</strong> salts<br />
of <strong>Lake</strong> <strong>Pikrolimni</strong>. Namely, <strong>in</strong> the past, after acquisition of solutes from<br />
<strong>at</strong>mospheric precipit<strong>at</strong>ion, hydrolysis of silic<strong>at</strong>e and possibly CO 2 of<br />
deep orig<strong>in</strong>, the w<strong>at</strong>ers are subjected to evapor<strong>at</strong>ion (<strong>at</strong> the surface or by<br />
capillarity). Initially, the concentr<strong>at</strong>ions of all constituents <strong>in</strong>crease until<br />
the m<strong>in</strong>eral precipit<strong>at</strong>ion occurs. The precipit<strong>at</strong>ion may take the form of<br />
efflorescent crusts on top of the surface or of <strong>in</strong>tergranural caliche-type<br />
films and cements. The efflorescent crusts are the product of complete<br />
desicc<strong>at</strong>ion and they are subjected to complete or differential solution<br />
byra<strong>in</strong>.Thisrecycl<strong>in</strong>gofsaltsisobservedactually<strong>at</strong><strong>Lake</strong><strong>Pikrolimni</strong>.<br />
Thus, dur<strong>in</strong>g summer when the w<strong>at</strong>er level of the lake is low or there is<br />
no w<strong>at</strong>er <strong>at</strong> all (total desicc<strong>at</strong>ion), the first ra<strong>in</strong>s dissolve the residual<br />
br<strong>in</strong>es (or salts) and get charged by solutes. This recycl<strong>in</strong>g of salts is also
142 E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
Table 3<br />
Thermodynamic calcul<strong>at</strong>ion model: sequence of m<strong>in</strong>erals precipit<strong>at</strong>ion.<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> 8/2002 <strong>Lake</strong> <strong>Pikrolimni</strong> 9/2006 <strong>Lake</strong> Magadi, Kenya a <strong>Lake</strong> N<strong>at</strong>ron, Tanzania a The Deep Spr<strong>in</strong>gs <strong>Lake</strong>, California a<br />
Calcite (CaCO 3 )<br />
Dolomite [Ca(Mg)CO 3 )]<br />
Calcite (CaCO 3 )<br />
Dolomite [Ca(Mg)CO 3 )]<br />
Calcite (CaCO 3 ) Calcite–Mg [Ca(Mg)CO 3 ]<br />
Dolomite (MgCO 3 )<br />
Gaylussite [Na 2 Ca⁎<br />
(CO 3 ) 2 ⁎5H 2 O]<br />
Gaylussite [Na 2 Ca⁎<br />
(CO 3 ) 2 ⁎5H 2 O]<br />
Gaylussite [Na 2 Ca⁎<br />
(CO 3 ) 2 ⁎5H 2 O]<br />
Pirssonite [Na 2 Ca⁎<br />
(CO 3 ) 2 ⁎2H 2 O]<br />
Pirssonite [Na 2 Ca⁎<br />
(CO 3 ) 2 ⁎2H 2 O]<br />
Pirssonite [Na 2 Ca⁎<br />
(CO 3 ) 2 ⁎2H 2 O]<br />
Nahcolite (NaHCO 3 ) Nacholite (NaHCO 3 ) Nahcolite (NaHCO 3 ) Nahcolite (NaHCO 3 ) Nahcolite (NaHCO 3 )<br />
N<strong>at</strong>ron (Na 2 CO 3 ⁎10H 2 O) N<strong>at</strong>ron (Na 2 CO 3 ⁎10H 2 O)<br />
Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O) Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O Trona (Na 2 CO 3 ⁎NaHCO 3 ⁎2H 2 O<br />
Thermon<strong>at</strong>rite (Na 2 CO 3 ⁎H 2 O) Thermon<strong>at</strong>rite (Na 2 CO 3 ⁎H 2 O)<br />
Gypsum (CaSO 4 ⁎2H 2 O)<br />
Glauberite [Na 2 Ca(SO 4 )]<br />
Burkeite (Na 2 CO 3 ⁎2Na 2 SO 4 )<br />
Mirabilite (Na 2 SO 4 ⁎10H 2 O) Mirabilite (Na 2 SO 4 ⁎10H 2 O)<br />
Thenardite (Na 2 SO 4 ) Thenardite (Na 2 SO 4 )) Thenardite (Na 2 SO 4 )<br />
Halite (NaCl) Halite (NaCl) Halite (NaCl) Halite (NaCl) Halite (NaCl)<br />
a Gueddari M. (1984).<br />
Table 4<br />
Simul<strong>at</strong>ion of evapor<strong>at</strong>ion of groundw<strong>at</strong>er.<br />
Label<br />
S/1_/2002<br />
b/2/2002<br />
b/3/2002<br />
b/4/2002<br />
b/A/2002<br />
wt/B/2002<br />
b/G/2003<br />
b/G/2004<br />
b/P1/7/2004<br />
b/G/7/2004<br />
b/B/8/2004<br />
b/SPA<br />
wt/SPA<br />
MWP-1<br />
MWPh-1<br />
Pl-5/2003<br />
M<strong>in</strong>erals<br />
BURKEITE_CALCITE_GAYLUSSITE_GLASERITE_HALITE_MAGNESITE_MIRABILITE_NATRON_PIRSSONITE_TRONA<br />
CALCITE_GAYLUSSITE_GLASERITE_HALITE_MAGNESITE_PIRSSONITE_THENARDITE_TRONA<br />
CALCITE_GAYLUSSITE_GLASERITE_HALITE_MAGNESITE_MIRABILITE_NAHCOLITE_PIRSSONITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_SYLVITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_GLAUBERITE_GYPSUM_HALITE_MAGNESITE_MIRABILITE_SYNGENITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
CALCITE_GLASERITE_HALITE_MAGNESITE_NAHCOLITE_THENARDITE<br />
HALITE_PIRSSONITE_THENARDITE_TRONA<br />
In each l<strong>in</strong>e, the list of m<strong>in</strong>erals corresponds to the whole sequence of precipit<strong>at</strong>ed m<strong>in</strong>erals dur<strong>in</strong>g concentr<strong>at</strong>ive evapor<strong>at</strong>ion.<br />
the way of <strong>in</strong>creas<strong>in</strong>g the solute load of the groundw<strong>at</strong>er. Precipit<strong>at</strong>ion<br />
and re-solution <strong>in</strong>evitably lead to strong segreg<strong>at</strong>ion of the <strong>in</strong>itial<br />
solutes. Therefore, sorption ion exchange and reduction reaction on<br />
m<strong>in</strong>eral surface may also remove certa<strong>in</strong> solute.<br />
8. Conclusions<br />
<strong>Lake</strong> <strong>Pikrolimni</strong> is a sal<strong>in</strong>e lake th<strong>at</strong> is characterized by alkal<strong>in</strong>e<br />
br<strong>in</strong>e, poor <strong>in</strong> Ca 2+ and Mg 2+ . The dilute HCO 3 − spr<strong>in</strong>g fresh w<strong>at</strong>er<br />
(350 g/L dissolved solid). Such br<strong>in</strong>es show a considerable<br />
range <strong>in</strong> ionic composition and concentr<strong>at</strong>ion. High solute concentr<strong>at</strong>ion<br />
due to solar evapor<strong>at</strong>ion of w<strong>at</strong>er, m<strong>in</strong>eral precipit<strong>at</strong>ion,<br />
fractional dissolution and solute recycl<strong>in</strong>g are the ma<strong>in</strong> processes<br />
responsible for these br<strong>in</strong>es form<strong>at</strong>ion. In particular, these evapor<strong>at</strong><strong>in</strong>g<br />
conditions were such th<strong>at</strong> <strong>in</strong>curred a hydrogeochemical environment<br />
th<strong>at</strong> was responsible for the lake to provide, <strong>in</strong> dry periods,<br />
“nitrum chalestricum” (trona). Besides, the progressive concentr<strong>at</strong>ion<br />
of br<strong>in</strong>es <strong>in</strong> alkal<strong>in</strong>e lakes leads to a preferential precipit<strong>at</strong>ion of<br />
sodium carbon<strong>at</strong>e followed by sulf<strong>at</strong>es and chlorides. This conclusion<br />
comes <strong>in</strong> agreement with the results. Also, the m<strong>in</strong>eralogical analysis<br />
(X-Ray Diffraction of salts), the evapor<strong>at</strong>ion simul<strong>at</strong>ion and the<br />
thermodynamic model showed th<strong>at</strong> salts of trona, burkeite and halite,<br />
deposit from these br<strong>in</strong>es. So, the conditions th<strong>at</strong> are responsible for<br />
the form<strong>at</strong>ion of soda seem to be present <strong>in</strong> the bas<strong>in</strong> and confirm<br />
Pl<strong>in</strong>y's description.<br />
Fig. 6. M<strong>in</strong>erals molarity versus concentr<strong>at</strong>ion factor.
E. Dotsika et al. / Journal of <strong>Geochemical</strong> Explor<strong>at</strong>ion 103 (2009) 133–143<br />
143<br />
Table 5<br />
Chemical composition (mmol/L) of the residual w<strong>at</strong>er, after evapor<strong>at</strong>ion, when the models stop.<br />
Label CF alk Cl SO 4 Na K Ca Mg TDS<br />
S/1/2002 12,300 1913 4300 610 6590 845 0 0.02 455,699<br />
b/2/2002 6850 1057 4510 780 6320 820 0 0.02 448,892<br />
PL-5/2003 2700 182 4830 840 5870 830 0.2 0.45 431,224<br />
S/1/2002 chemical composition of residual w<strong>at</strong>er when <strong>n<strong>at</strong>ron</strong> starts to precipit<strong>at</strong>e 395 3143 1417 459 5448 30 0.004 0.03 318,000<br />
S/1/2002 chemical composition of residual w<strong>at</strong>er when trona starts to precipit<strong>at</strong>e 720 3246 2500 790 7273 53 0.001 0.02 434,000<br />
Testimonia<br />
Testimonium 1. Pl<strong>at</strong>o, 5th century BC, Republic 430a (ed. I. Burnet,<br />
1963).<br />
Testimonium 2. Scolia <strong>in</strong> Pl<strong>at</strong>onem, 159 (ed. Ruhnken).<br />
Testimonium 3. Etymologicum Magnum, Halastri.<br />
Testimonium 4. Stephanus Byzantius, Halastra.<br />
Testimonium 5. Pl<strong>in</strong>y, N<strong>at</strong>uralis Historiae 31, 46 (ed. W.H.S. Jones,<br />
Loeb 1963). Chapter on soda, esp. 106–109.<br />
Acknowledgements<br />
The authors would like to express their gr<strong>at</strong>itude to Mr. Aggelakis<br />
for his help dur<strong>in</strong>g sampl<strong>in</strong>g sessions as well as to Dr. Risacher F. and<br />
Dr. Fritz B. for their contribution of the geochemical program EQL/<br />
EVP. F<strong>in</strong>ally, special thanks go to the two unknown reviewers for their<br />
really contributory remarks and suggestions of the manuscript and to<br />
the editor of this journal for his helpful guidance.<br />
References<br />
An<strong>at</strong>oliki, A.E., 2006. Public Particip<strong>at</strong>ion <strong>in</strong> W<strong>at</strong>er Re<strong>source</strong>s' Management <strong>at</strong> a Greek<br />
River Bas<strong>in</strong>. Project W<strong>at</strong>er Agenda, LIFE04/ENV/GR/000099.<br />
Apha, 1989. Standard Methods for the Exam<strong>in</strong><strong>at</strong>ion of W<strong>at</strong>er and Wastew<strong>at</strong>er, 19th<br />
end. American Public Health Associ<strong>at</strong>ion, Wash<strong>in</strong>gton, DC.<br />
Coulson, W.D.E., Leonard Jr., A., 1979. A prelim<strong>in</strong>ary study of the Naukr<strong>at</strong>is region <strong>in</strong> the<br />
western Nile Delta. Journal of Field Archaeology 6.2, 151–168.<br />
Dalezios, N., Loukas, A., Bampzelis, D., 2002. The role of agrometeorological and<br />
agrohydrological <strong>in</strong>dices <strong>in</strong> the phenology of whe<strong>at</strong> <strong>in</strong> central <strong>Greece</strong>. Physics and<br />
Chemistry of the Earth 27, 1031–1038.<br />
De Cosson, F.C., 1936. El Barnugi. Bullet<strong>in</strong> de Societe Royale d'Archeologie — Alexandrie<br />
30 (N.S. IX-1), 113–116.<br />
Eugster, H.P., 1966. Sodium carbon<strong>at</strong>e–bicarbon<strong>at</strong>e m<strong>in</strong>erals as <strong>in</strong>dic<strong>at</strong>ors of pCO 2 .<br />
Journal of Geophysical Research 71, 3369–3377.<br />
Eugster, H.P., 1980. Geochemistry of evaporitic lacustr<strong>in</strong>e deposits. Annual Review of<br />
Earth and Planetary Sciences 8, 35–63.<br />
Eugster, H.P., Jones, B.F., 1979. Behavior of major solutes dur<strong>in</strong>g closed-bas<strong>in</strong> br<strong>in</strong>e<br />
evolution. American Journal of Science 279 (6), 609–631.<br />
Evelyn-White, H.G., Hauser, W., 1926–1933. The Monasteries of the Wadi 'n N<strong>at</strong>run, 3<br />
volumes, Public<strong>at</strong>ions of the Metropolitan Museum of Art Egyptian Expedition 2, 4,<br />
8, New York.<br />
Gueddari, M., 1984. Geochimie et thermodynamique des evaporites cont<strong>in</strong>entales, etude<br />
du lac N<strong>at</strong>ron en Tanzanie et du chott El Jerid en Tunisie. Sciences Geologiques<br />
Universite Louis Pasteur de Strasbourg, Institute de Geologie, Mémoire 76, 144.<br />
Hardie, L., Eugster, H., 1970. The evolution of closed-bas<strong>in</strong> br<strong>in</strong>es. M<strong>in</strong>eral. Soc. Amer.<br />
Spec. Pap. 3, 273–290.<br />
H<strong>at</strong>zopoulos, M.B., Loukopoulou, L.D., 1989. Morrylos. Athenes, Centre de Recherches<br />
de l'Antiquite Grecque et Roma<strong>in</strong>e.<br />
Henderson, J., 1985. The raw m<strong>at</strong>erials of early glass production. Oxford Journal of<br />
Archaeology 4 (3), 267–291.<br />
Ign<strong>at</strong>iadou, D., 2002. Research on the Nitrum Chalestricum. Proceed<strong>in</strong>gs of Archaeological<br />
Work <strong>in</strong> Macedonia and Thrace 16, Thessaloniki, <strong>Greece</strong>.<br />
Jones, B.F., Eugster, H.P., Rettig, S.L., 1977. Hydrochemistry of the <strong>Lake</strong> Magadi bas<strong>in</strong>,<br />
Kenya. Geochimica et Cosmochimica Acta 41, 53–72.<br />
Mart<strong>in</strong>, M., Sauneron, S. (Eds.), 1982. Claude Sicard. In: Bibliotheque d'etudes, vols. 3. Le<br />
Caire, pp. 83–85.<br />
Mercier, J., 1966. Etude géologique des zones <strong>in</strong>ternes des Hellénides en Macédo<strong>in</strong>e<br />
centrale (Grèce). Contribution a l'étude du métamorphisme et de l'évolution<br />
magm<strong>at</strong>ique des zones <strong>in</strong>ternes des Hellénides, Thèses, Paris 1966, Ann. geol. Pays.<br />
Hellen., 20, 1–792.<br />
Mountrakis, D., 1985. Geology of <strong>Greece</strong>. InUniversity Studio Press, pp. 50–72.<br />
Thessaloniki.<br />
Parkhurst, D.L., Appelo, C.A.J., 1999. User's guide to PHREEQC—a computer program for<br />
speci<strong>at</strong>ion, reaction-p<strong>at</strong>h, 1D-ransport, and <strong>in</strong>verse geochemical calcul<strong>at</strong>ion: US<br />
Geological Survey. W<strong>at</strong>er-re<strong>source</strong>s Investig<strong>at</strong>ions Report 99, 4259–4312.<br />
Reade, W., Freestone, I.C., St Simpson, J., 2005. Innov<strong>at</strong>ion or cont<strong>in</strong>uity Early first<br />
millennium BCE glass <strong>in</strong> the Near East: the cobalt blue glasses from Assyrian<br />
Nimrud. In: Cool, H. (Ed.), Annales du 16e Congres de l'Associ<strong>at</strong>ion Interntionale<br />
pour l'Histoire du Verre-London, 2003, pp. 23–27.<br />
Risacher, F., 1992. Geochimie. Sciences Géologiques Bullet<strong>in</strong> 45 (3–4), 135–214.<br />
Risacher, F., Clement, Α., 2001. A computer program for the simul<strong>at</strong>ion of evapor<strong>at</strong>ion of<br />
n<strong>at</strong>ural w<strong>at</strong>ers to high concentr<strong>at</strong>ion. Computers and Geosciences 27, 191–201.<br />
Schlick-Nolte, L.A., Werthmann, R., 2003. Glass vessels from the Burial of Nesikhons.<br />
Journal of Glass Studies 45, 11–34.<br />
Shortland, A., 2004. Evaporites of the Wadi N<strong>at</strong>run: seasonal and annual vari<strong>at</strong>ion and<br />
its implic<strong>at</strong>ion for ancient exploit<strong>at</strong>ion. Archaeometry 46, 497–516.<br />
Shortland, A., Schachner, L., Freestone, I., Tite, M., 2006. N<strong>at</strong>ron as a flux <strong>in</strong> the early<br />
vitreous m<strong>at</strong>erials <strong>in</strong>dustry: <strong>source</strong>s, beg<strong>in</strong>n<strong>in</strong>gs and reasons for decl<strong>in</strong>e. Journal of<br />
Archaeological Science 33, 521–530.<br />
Taher, A.G., 1999. Inland sal<strong>in</strong>e lakes of Wadi el N<strong>at</strong>run depression, Egypt. Intern<strong>at</strong>ional<br />
Journal of Salt <strong>Lake</strong> Research 8 (2), 149–169.<br />
Wenigswieser, S., 1992. M<strong>in</strong>eralogische untersuchungen an den Evaporiten und Tonen<br />
des Wadi El-N<strong>at</strong>run, Fakul<strong>at</strong> fur Bio- und Geowissenschaften, Universit<strong>at</strong> Karlsruhe,<br />
Unpublished thesis.