PARTIAL PRESSURES OF CO2 IN EPIKARSTIC ZONE DEDUCED FROM HYDROGEOCHEMISTRY OF PERMANENT DRIPS, THE MORAVIAN KARST, CZECH REPUBLIC
DELNI TLAK CO2 V EPIKRAŠKI CONI, KOT GA RAZKRIVAJO HIDROKEMIčNE RAZISKAVE STALNIH VODNIH CURKOV
Jiri FAIMON1, Monika LIČBINSKA1-2, Petr ZAJIČEK3 & Ondra SRACEK4-5
Abstract	UDC 551.444:550.4(437.32)
Jift Faimon, Monika Ličbinska, Petr Zajtček & Ondra Sracek: Partial pressures of CO2 in epikarstic zone deduced from hy-drogeochemistry of permanent drips, the Moravian Karst, Czech Republic
Permanent drips from straw stalactites of selected caves of the Moravian Karst were studied during one-year period. A hypothetical partial pressure of CO2 that has participated in lime-
stone dissolution, Pro2(H)=10-'
was calculated from the
dripwater chemistry. The value significantly exceeds the partial pressures generally measured in relevant shallow karst soils, PCO2(soii)=10-2'72±002. This finding may have important implications for karst/cave conservation and paleoenvironmental reconstructions.
Keywords: cave, carbon dioxide, dripwater, hydrogeochemis-try, hypothetical partial pressure, karst processes, karstification model.
Izvleček	UDK 551.444:550.4(437.32)
Jift Faimon, Monika Ličbinska, Petr Zajtček & Ondra Sracek: Delni tlak CO2 v epikraški coni, kot ga razkrivajo hidrokemične raziskave stalnih vodnih curkov
V obdobju enega leta smo analizirali preniklo vodo izpod sta-laktitnih cevčic v izbranih jamah Moravskega krasa. Iz analiz sklepamo, da je hipotetični delni tlak CO2, pri katerem poteka raztapljanje apnenca PCO2(H)=10-1'53±0 04. Ta vrednost znatno presega vrednosti meritev delnega tlaka CO2 v kraški prsteh PCO2(soil)=10-2'72±002 Dobljeni rezultati so potencialno pomembni za zaščito jam in krasa ter rekonstrukcijo paleoklime. Ključne besede: jama, ogljikov dioksid, prenikle vode, hidro-geokemija, hipotetični delni tlak, kraški procesi, model zakra-sevanja.
INTRODUCTION
Currently, scientific effort focuses on karst processes for two main reasons: (1) karst systems require better conservation because they are widely impacted by anthropogenic activities (Fernandez et al. 1986; Dragovich & Grose 1990; Pulido-Bosch et al. 1997; Baker & Genty 1998; Hoyos et al. 1998; Balak et al. 1999; Sanchez-Moral et al. 1999; Song et al. 2000, Carrasco et al. 2002; Faimon
et al. 2004b, 2006; Beach et al. 2008; Russell & MacLean 2008) and (2) terrestrial speleothems are increasingly used as archives of paleoclimate data (see McDermott 2004, or Fairchild et al. 2006, for a review). Basic geo-chemical interactions in carbonate karst are summarized in box model in Fig. 1. It comprises five elementary processes:
1	Masaryk University, Department of Geological Sciences, Kotlarska 2, 611 37 Brno, Czech Republic, e-mail: faimon@sci.muni.cz
2	Institute of Geological Engineering, Faculty of Mining and Geology, VŠB - Technical University of Ostrava, 17.listopadu 15, 708 33 Ostrava - Poruba, Czech Republic
3	Agency for Nature Conservation and Landscape Protection of the Czech Republic, Caves Administration of the Moravian Karst, Svitavska 11-13; 678 01 Blansko, Czech Republic
4	Department of Geology, Faculty of Science, Palacky University, 17. listopadu 12, 771 46 Olomouc, Czech Republic
5	OPV s.r.o. (Protection of Groundwater Ltd), Belohorska 31, 169 00 Praha 6, Czech Republic Received/Prejeto: 14.8.2011
ACTA CARSOLOGICA 41/1, 47-57, POSTOJNA 2012
H2O
CO
2(g)
CO
CO2(aq) + H2O
2(aq),
H2CO3,
H2CO3
H+ + HCO-,
HCO-- ^^_H+ + CO32-,
and
CaCO3(caic) ^_Ca3+ + CO3-.
(1) (3)
(3)
(4)
(5)
Details on the processes (Eq. 1-5) can be found elsewhere (e.g., Plummer & Busenberg 1983; Dreybrodt 1988; Stumm & Morgan 1996). All the processes in solution appear to be at equilibrium due to high rate of hydrolytical/dissociation reactions (Stumm & Morgan 1996, pp. 193-194), even though, the conversion of CO3(aq) into aqueous carbonate species could be somewhat slower (Dreybrodt et al. 1996, 1997). In contrast, partial disequilibrium is frequently observed at the both AS-boundary (atmosphere-solution boundary) and SC-boundary (solution-calcite boundary) because of relatively slow dissolution/degassing of gaseous CO3 and calcite dissolution/precipitation. Disequilibrium extent is quantified by saturation index, SI(calc), in the case of the SC-boundary and by difference between the partial pressure in atmosphere, PCO3(g), and the partial pressure related to the activity of aqueous carbonate species, Pco3(w), in the case of the AS-boundary. Generally, total equilibrium is conditioned by partial equilibria of all elementary processes (Eq. 1-5). This means that all fluxes (depicted by arrows in the Eq. 1-5) must be balanced by relevant counter-fluxes. Total equilibrium can be attained relatively quickly (in days or tens of days) at various levels in a karst profile (soils, epikarst, vadose zone), but it can be disturbed by changing conditions. Changes in solution composition caused by water mixing or change of CO3 partial pressure along water flow path are the most important factors. Whereas the problem of water mixing was originally discussed in Bögli (1964, 1980), the issue of CO3 partial pressure changes along the flow path of the water is addressed in this article. As a karst profile is not accessible easily, caves represent a cross-section into vadose zone and cave dripwaters carry a record of preceding karst processes. The karst waters entering a cave are mostly saturated at higher CO3 partial pressure (typical for karst CO3 sources) in comparison with instantaneous partial pressure of CO3 in a cave. As a response, dripwater degasses (the process Eq. 1 is running from right to left), CO3(aq) activity decreases, and all additional processes (Eq. 3-5) reach a new
Fig. 1: Box model of basic interactions in carbonate karst.
equilibrium (Holland et al. 1964). As a result, pH and CO33- activity increase and water becomes supersaturated with respect to calcite. Eventually, the excess of calcium and carbonate ions is expelled from the solution as calcite (speleothem growth) (see - e.g., Dreybrodt 1988, for a review). In this article, we demonstrate that there is a possibility to reconstruct the original partial CO3 pressure by geochemical modeling. It is believed that biogenic CO3 produced in karst soils is a main source of karst CO3. It is derived mainly from (1) autotrophic and (3) heterotrophic respiration (see - Kuzyakov & Lari-onova 3005; Kuzyakov 3006 - for review). The CO3 concentrations show high seasonality with highest values in summer and lower values in autumn/winter (Uchida et al. 1997; Moncrieff & Fang 1999; Pilegaard et al. 3001; Faimon et al. 3004b, 3010, 3013). Little is known about the epikarstic sources of CO3. Up to now, it has rather been matter of hypotheses and speculations (Atkinson 1977; Fairchild et al. 3000, 3006; Spötl et al. 3005; Faimon et al. 3010, 3013). However, the recent results of Benavente et al. (3010) who measured up to 6 vol. % of
CO2 in karst vadose zone indicate that epikarstic CO2 sources could be significant. The aim of the present work is (1) to specify partial pressure of CO2 participating in
dripwater chemistry formation, (2) to try distinguishing individual CO2 sources, and (3) to evaluate the role of seasonal variations.
RESEARCH SITE AND METHODOLOGY
SITE CHARACTERIZATION The Moravian Karst (see - e.g., Absolon 1970; Balak et al. 1999; Otava & Balak 2002) is the largest karst area in the Bohemian Massif. It is situated north of Brno (49°13' to 49°25' N, 16°38' to 16°47' E) as a part of the Drahany Highlands. Altitude varies between 244 m and 613 m; average value is 447.5 m. Karst rocks cover an area of 94 km2 forming a belt 3-5 km wide and 25 km long. Total rock thickness is estimated to be 500-1,000 m. Besides some Jurassic sandstones/limestones and Cretaceous sediments, the Moravian Karst is formed mainly by Middle and Upper Devonian limestones. They are divided into two formations, the Macocha Fmt. (Vavrinec Lmst., Josefov Lmst., Lažanky Lmst., Vilemovice Lmst.) and the Lišen Fmt. (Krtiny Lmst., Hady-Rlčka Lmst.). The largest cave system is the Amaterska Cave (over 15 km of corridors). At present, four caves are open to the public, the Punkevni Caves, the Sloupsko-Šošuvske Caves, the Katerinska Cave, and the Balcarka Cave. The whole karst area is geologically very stable and any endogenous carbon dioxide emanations were not reported.
The study was performed in the northern part of the Moravian Karst, in the Punkevni Cave, the Balcarka Cave, and the Amaterska Cave (Fig. 2). The parent limestones (Vilemovice Lmst.) are composed of calcite as a predominant component. Dripwaters are of Ca-HCO3 type. Overlying vegetation comprises a mix of dispersed deciduous woodland (beech dominates), conifer woodland (spruce dominates), grasses with a thin brown
rendzina soil cover and a part of vegetation is composed of agriculture plants, e.g. wheat, oilseed rape, and mustard plant. Mean annual rainfall in the study area is about 700 mm and mean annual temperature is about 10°C.
SAMPLING AND ANALYTICS Totally 88 dripwater samples were studied during one-year-period study (2002-2003). From 11 drips, 4 drip rates were relatively slow (28.1±6.6 ml/hour), 3 drip rates were quick (1070.6±427.4 ml/hour), and 4 drip rates were moderate (194.4±48.4 ml/hour) (the confidence intervals a=0.05). The samples of dripwater were collected from speleothems into polyethylene vessels of volume 50-100 ml. Time of sampling varied in the range of tens minutes to two hours. Immediately in the cave environment, pH (WTW pH 330i, WTW pH-electrode SetTix 22), alkalinity (microtitration by 0.05 mol l-1 HCl (evaluated as a Gran plot, Stumm & Morgan 1996, pp. 179-186, or Appelo & Postma 2005, pp. 183-186), and aqueous calcium (complexometric microtitration by 0.01 mol l-1 EDTA, 10% KOH, calcein as inner indicator) were determined. The waters were then analyzed in the laboratory for K, Mg, Na (AAS), NHJ, NO-, POt-, NO-(spectrophotometry), SOl, and Cl- (microtitration). The estimated analytical errors were below 5%. Cave CO2 concentrations were measured in situ by 2-channel IR-detector FT A600-CO2H linked with ALMEMO 2290-4 V5, Ahlborn, Germany (measuring range: 0 to 10,000 ppmv; accuracy: ± 50 ppmv + 2 vol. % of measured value in the range of 0 to 5000 ppmv; resolution: 1 ppmv or 0.0001 vol %). Data on shallow soil CO2 concentrations were taken from Faimon et al. (2010, 2012).
DATA PROCESSING The speciation, saturation indices, and hypothetical PCO2 values were computed by the PHREEQC code (Parkhurst & Appello 1999). Statistical analysis was performed in the STA-TISTICA code (StatSoft, Inc., www.statsoft.cz).
Fig. 2: Sketch map of the Moravian Karst and the caves of interest.
Fig. 3: Dripwater composition, monitored cave , saturation indices, and calculated and	(88 samples).
RESULTS
dripwater composition
The hydrochemistry of 88 dripwater samples is summarized in the box-plot in Fig. 3. The pH values varied in the range from 7.38 to 8.33 (7.88 median; 7.90±0.05 confidence interval). The calcium concentration varied in the range (2.42-5.52)x10-3 mol L-1 with 3.58x10-3 mol L-1 median and (3.61±0.14)x10-3 mol L-1 confidence interval. Alkalinity varied in the range (0.41-1.05)x10-2 eq L-1 (6.30x10-3 eq L-1 medians, (6.26±0.27)x10-3 eq L-1 confidence interval). The concentrations of K, Mg, and Na were in (1.53±0.11)x10-5, (7.60±1.40)x10-5, and (1.18±0.15)x10-4 mol L-1 confidence intervals, respectively. The concentrations of NO-, Cl-, and SOt- were in (1.63±0.72)x10-4, (8.92±1.60)x10-5, and (4.07±0.32)x10-4 mol L-1 confidence intervals, respectively. The given composition roughly falls into the range reported for dripwa-ters by McDonald et al. (2007), Jimenez-Sanchez et al.
(2008), or Baker et al. (2000). On the other hand, drip-waters collected in the study are somewhat more mineralized than dripwaters reported by, e.g., Covelli et al. (1998), Vocal et al. (1999), Musgrove & Banner (2004), Spötl et al. (2005), or Cai et al. (2011). Partial pressures of
CO2 in cave atmosphere varied in the range 10-3.35-10-2.15 (10-2.81 median, 10-2.83±0.07 confidence interval), which is consistent with the values reported by Ek & Gewelt (1985), Spötl et al. (2005), Baldini et al. (2006) or Faimon et al. (2010, 2012). All confidence interval are for a=0.05 (see Davies 2002).
SATURATION INDICES, SI(calc) AND pc02(w) Saturation index with respect to calcite, S7(calcite) = log (Q/K,) (where Q=aCa. aCO2- and Kc is calcite solubility product), and partial pressure of gaseous CO2 that would be at equilibrium with aqueous CO2, pc02(w), were calculated using PHREEQC code. The SIcalcite values varied in the range 0.22-1.38 (0.92 median, 0.89±0.05 confidence interval) (see Fig. 3). These values somewhat exceed those presented by Covelli et al. (1998), Tooth & Fairch-ild (2003), or Spötl et al. (2005). The PCO2(W) values were in the range 10-2.98-10-1.95 (10-2.43 median, 10-2.44±».»5 confidence interval) (see Fig. 3), which is consistent with Fai-mon et al. (2006, 2012). Note that the values significantly exceeded the PCO2(g) monitored in cave air. All confidence interval are calculated for a=0.05.
DATA analysis
Correlations of all variables (concentrations of aqueous species, saturation indices, and partial CO2 pressures) in form of a nonparametric Spearman's correlation coefficient, p, are given in the matrix in Tab. 1. The statistically significant correlations at a=0.05 (see Davies 2002) are highlighted (p > 0.22). The couples of variables Ca/logPCO2(H), alk/logPCO2(H), Mg/Cl, and logPCO2(g)/ logPcO2(w) show strong positive correlations (p > 0.60). In contrast, strong correlations of the couples pH/logPCO2(w), pH/logPCO2(g), and SI(calc)/logPCO2(w) are negative.
Weaker correlations (0.40 < p < 0.60) were found for the couples Ca/alk, alk/Mg, alk/logPCO2(w), Mg/Na, Mg/NOj, Mg/SO4, Mg/logPCO2(H), Na/NOj, Na/Cl, NO3/a, a/SO4, logPCO2(w)/logPCO2(H) (positive) and K/Na, IC/NO3, SI(calc)/logPCO2(g) (negative).
Very weak correlations (0.22 < p0 < 0.40) were found for the couples Ca/K, Ca/Mg, Ca/SO4, Ca/SI(calc), Ca/logPCO2(w), alk/SO4, alk/SI(calc), alk/logPCO2(g), K/SI(calc), K/logPCO2(H), Mg/logPCO2(g), Mg/logPCO2(w), Na/SO4, Cl/logPCO2(w), SO4/logPCO2(H), SI(calc)/logPCO2(H), logPCO2(g)/ logPCO2(H), (positive), and pH/Cl, K/Cl (negative).
tab. 1: Spearman Rank Order Correlations. Statistically significant correlations (a=0.05) are highlighted.
	pH	Ca	alk	K	Mg	Na	NO3-	Cl-	SO4-	Sl(ca,c,	logPCO2(g)	logPCO2(w)	l°gPCO2(H)
pH	1.00												
Ca	-0.20	1.00											
alk	-0.16	0.56	1.00										
K	0.09	0.23	0.15	1.00									
Mg	-0.19	0.36	0.49	-0.21	1.00								
Na	-0.15	-0.09	0.02	-0.57	0.58	1.00							
NO-	-0.12	-0.14	-0.13	-0.45	0.43	0.51	1.00						
Cl-	-0.25	0.17	0.12	-0.25	0.61	0.54	0.56	1.00					
SO4-	-0.03	0.31	0.30	0.01	0.43	0.32	0.14	0.48	1.00				
S'(calc)	0.83	0.28	0.30	0.23	0.07	-0.15	-0.19	-0.12	0.17	1.00			
lOgPcO2(g)	-0.61	0.10	0.39	0.10	0.30	0.14	0.08	0.18	0.05	-0.41	1.00		
l°gPcO2(w)	-0.95	0.35	0.43	-0.02	0.32	0.14	0.08	0.28	0.14	-0.66	0.67	1.00	
l°gPcO2(H)	-0.18	0.77	0.94	0.22	0.49	-0.02	-0.16	0.14	0.30	0.32	0.33	0.43	1.00
alk - alkalinity
CALCULATION OF Pco2(h) The calculation focuses on the determination of a hypothetical partial CO2 pressure, PCO2(H), that has participated in both limestone dissolution and development of resulting water chemistry. The PCO2(H) was found as the partial CO2 pressure, at which degassed dripwa-ter of a given composition would return to the equilibrium with calcite. By PHREEQC code, Pco2(h) were determined by "adjusting" pH so that solution was at equilibrium with calcite, see the input file in Appendix.
Fig. 4: Frequency diagrams of partial pressures of CO2 in (a) shallow karst soils (286 samples) (Faimon et al. 2010, 2012) and (b) hypothetical epikarst CO., source (88 samples).
Alternatively, the logPCO2(H) may be estimated from the equation
log PcO2(H) = log PcO2(w) + SI,
(calcite)>
(6)
that is result of combining the expressions for equilibrium constants of the processes given by eqns. 1-5 with the simplified electric charge balance, 2[Ca2+^[HCO-]. Eqn. (6) is valid close to calcite-water-CO2 equilibrium. With SI(cajc,te) increasing up to ~ 1.5, the logPCO2(H) relative error increases exponentially up to ~10%.
The found PCO2(H) values ranged from IO-1.92 to 10-a96 (10-i.si median, lO-1.53^0.04 (a=0.05) (see Fig. 3). Distribution of the calculated PCO2(H) is given in the frequency diagram in Fig. 4 and is compared with PCO2(soil) values found in shallow karst soils by Faimon et al. (2010, 2012). Note that the logPco2(soil) and logPC02(H) show very different modes, -2.75 and -1.45, respectively. Seasonal variation of PCO2(H) are small; the summer values, 10-1.48±0.06 (May to September), only slightly exceeded the winter values, 10-1.59±0.05 (a=0.05) (October to March).
DISCUSSION
INTERPRETATION OF CORRELATIONS The strong positive correlations of Ca/logPCo2(H) and alk/logPCo2(H) reflect stoichiometry of limestone dissolution (see Eq. 1-5). Note that logPcO2(H) must fit the dominant aqueous species that control its calculation. The dissolution stoichiometry is also mirrored by the weaker positive correlations of Ca/alk, alk/Mg, Mg/logPCO2(H) and Ca/Mg. Some positive correlations (Ca/SI(calc), alk/SI(calc), Ca/logPCO2(w), Mg/logPCO2(w), SI(calc)/logPCO2(H), logPCO2(w)/logPCO2(H)) indicate dependence of variables on the extent of water mineralization. The positive correlations of pH/SI(calc), logPCO2(g)/logPCO2(w,), alk/logPCO2(w), alk/logPCO2(g), together with negative correlations of pH/logPCO2(w), pH/logPCO2(g), SI(calc)/logPCO2(w), or SI(calc)/logPCO2(g), reflect dripwater degassing. The weak positive correlation logPCO2(g)/logPCO2(H) indicates interrelationship between source CO2 and cave CO2. The positive correlations of variables such as Na/Mg, Na/NO3, NO3/a, Na/Cl, a/SO4, Na/SO4, Mg/Cl, Mg/NO3 and Mg/SO4 are probably consequence of rain water stoi-chiometry and/or agriculture activities. The reasons for other weak correlations as Ca/K, Ca/SO4, SO4/logPCO2(H), alk/SO4, K/SI(calc), K/logPCO2(H), Mg/logPCO2(g), Cl/logPCO2(w) (positive) and K/Na, pH/Cl, K/Cl, K/NO3 (negative) are less comprehensible.
SATURATION INDICES The dripwater data have been plotted as the graph of pH vs. Ca concentration (Fig. 5). As can be seen, all the experimental points are above the equilibrium line. They are roughly arranged into a horizontal line that corresponds to degassing process. Extrapolation of the line to the left towards the equilibrium line shows the composition corresponding to PCO2(H) value. The summer and winter data are partly separated, but they both show the similar PCO2(H) values. The shift of the winter data to the
right (to higher supersaturation) mirrors higher degassing because of increasing difference between (1) the initial PCO2(g) participating on dripwater formation (corresponding to PCO2(H)) and (2) the cave air PCO2(g) reduced by stronger winter ventilation. Besides, this seasonality indicates that the studied dripwaters were degassed as far as in cave environment.
Fig. 5: The plot of Ca concentration against pH. The bold line indicates total equilibrium in pure calcite-water-CO^^^ system. The intersections of perpendicular dashed lines with the equilibrium line denote equilibrium composition at different PCO2(g, 10-' ' (left) and 10-3-5 (right). Evolution in horizontal direction corresponds to water degassing.
PCO2(H) AND KARSTIFICATION MECHANISMS Principally, two different karstification models can be distinguished: (1) closed system model (CSM), where the concentrations of aqueous carbonate species are not replenished from surrounding environment and (2) open system model (OSM), where the water dissolving limestone is in contact with the source of gaseous CO2. The
Fig. 6: Dependence of equilibrium PCO2(g) on initial P^ CSM conditions (A) and OSM conditions (B).
real karstification process is probably a compromise between both models.
Pco2(h) in closed system
Based on the CSM, the seepage water is initially saturated by CO2 at given PCO2(g) and equilibrates with calcite in the zone that is without a contact with original gaseous CO2. Such situation could be conceivable, e.g., if the movement of seepage water were faster than CO2 diffusion from solution/atmosphere boundary and calcite dissolution. In this case, a certain portion of CO2(aq) is "consumed" by dissolution and equilibrium PCO2(g) (corresponding to PCO2(w) and also to calculated PCO2(H)) is lower than the initial one. It is documented in Fig. 6 by deviation of the curve (A) representing the CSM from the curve (B) valid for the OSM. As follows from Fig. 6, the deviation is smaller when the initial PCO2(g) is higher. In case that water does not achieve equilibrium with calcite, the deviation is proportionally smaller. To summarize, actual initial PCO2(g) participating on dripwater chemistry under the CSM conditions could even be higher than that calculated as PcO2(H).
Another theoretical possibility to disturb PCO2(H) under closed system conditions is an addition of acidi-
Fig. 7: Mixing model - evolution of hydrogeochemical variables during mixing of two waters of different hydrochemistry. (A) Non-reactive mixing model of the water#1 and water#2. (B) Reactive mixing model; mix of the waters#1 and water#2 subsequently equilibrated with calcite (SI^c^c==0).
ty from an external source to the initially saturated water. As result, PCO2(H) shifts to higher value than that, at which the water has actually been saturated. In fact, it is hardly conceivable idea under real karst conditions.
Pco2(H) IN OPEN SYSTEM Under the OSM conditions, seepage water is saturated by both CO2 and calcite to equilibrium at constant PCO2. Therefore, the initial and equilibrium PCO2 are the same (see the curve A in Fig. 6). It should be noted that constant PCO2 requires an extensive CO2 reservoir with strong input fluxes. As the seepage water can equilibrate with both
CO2(g) and calcite already at the base of soil profile or in epikarst, we believe that real karst conditions are generally shifted closer to OSM. The relatively high values of pCO2(H) calculated from our data set are consistent with this idea. In case that water does not achieve equilibrium with calcite under OSM conditions, the hypothetical partial pressure PCO2(H) is simply PCO2(w). In case of pre-
precipitation calcite along transport path, the calculated PCO2(H) would be lower than that actual initial PCO2(g).
A problem of both former models is potential mixing of waters in vadose zone. In Fig. 7, mixing model is given for two different initial waters: the water #1 which has been saturated with respect to calcite at PCO2(g)~10-3.5 (see SI(calcite)=0, and Pco2(w)=10-3.5) and the water #2
influenced to some extent by karstification mechanism (closed vs. open system) and by various processes (water mixing, precipitation, acidification). However, most of these possibilities lead to PCO2(H) values lower than that initial PCO2(g) actually participating on dripwater chemistry. If water interacts with limestone under open conditions with respect to CO2(g) (as probably in this study),
which has been saturated with respect to calcite at PCO2(H) may be a good estimator of initial PCO2(g). In all
PCO2(g)~10-1.5 (see SI(calcite)=0, and Pco2(w)=10-1.5). The first mixing model is non-reactive; it includes simple mixing without any other processes (Fig. 7a). The resulting mix (calculated as water#1/(water#1+water#2) is unsaturated with respect to calcite, SI(calcite)<0, despite equilibrium of both initial waters. It is consistent with the assumptions of Bögli (1964, 1980), Gabrovšek & Dreybrodt (2000), Dreybrodt et al. (2010), or Qian & Peiyue (2011). The PCO2(H) values deduced from such mix can be somewhat overrated with respect to the actual PCO2 values controlling calcite saturation of the initial waters. The reason is in the interpretation of the SI(calc) decrement as a consequence of PCO2(H) enhancement. It indicates that the non-reactive mixing model produces a positive error for water unsaturated with respect to calcite. It should be noted that it is not the case of the studied dripwaters. The second model is reactive; water mixing is followed by subsequent calcite dissolution up to saturation, which probably better corresponds to real conditions. The calculated PCO2(H) (equivalent to PCO2(w)) is always between the initial PCO2(w) values for water#1 and water #2 (see Fig. 7b). It means that the calculated PCO2(H) is located between the lowest and the highest PCO2(g) values, with which the waters were in contact.
IMPLICATIONS The reconstruction of initial PCO2(g) from dripwater chemistry is not trivial. The calculated PCO2(H) values can be
cases, PCO2(H) could be understood as an important dripwater parameter that allows a better estimation of the conditions of limestone dissolution and dripwater formation.
The values derived from dripwater hydrochemistry, PCO2(H)~10-1.53±"."4, and the partial pressures in relevant karst soils, PCO2(soii) ~ 10-2.72±0.02, show a clear inconsistence, predicted already by Bourges et al. 2001 or Fai-mon et al. 2010, 2012. As the percolating waters were in a contact with CO2 of higher partial pressure than those in soils, it can be deduced that the CO2 source is situated deeper in karst profile. we believe that this CO2 is a product of decomposition of organic matter flushed and buried together with limestone weathering products in karrens of deeper epikarstic zone. Such CO2 source located deeper below ground surface in epikarst zone is consistent with the idea of Atkinson (1977) about "ground air CO2" and also with the results of direct measurements in epikarst zone by Benavente et al. (2010). The small difference between summer and winter PcO2(H) values indicates that the source could be less dependent on external conditions than generally expected. The CO2 production could be also less sensitive to short-term climatic changes, which is important information for the researchers dealing with paleoenvironmental reconstruction. In addition, such CO2 source could also be less vulnerable by processes in shallow karst soils such as, e.g., changes of vegetation cover or agriculture activities.
CONCLUSIONS
Hydrochemistry of permanent drips in selected caves of the Moravian Karst was studied. All dripwaters show supersaturation with respect to calcite. Correlations of analyzed variables indicate that it is a result of CO2 degassing from dripwater. Based on a simplified model, a hypothetical partial pressure of the CO2 that had participated in dripwater chemistry formation was calculated. The resulting values, PCo2(H)=10-1.53±"."4, exceed substantially
those directly measured in shallow karst soils. These findings question the widely accepted idea that shallow soils are a dominant source of karst CO2. As an alternative source, it offers CO2 produced in deeper epikarstic zone. In such case, water hydrogeochemistry would not necessarily reflect short-term climatic changes and processes in soils. This conclusion may play a role in paleoclimatic reconstructions and karst protection.
ACKNOWLEDGMENTS
tte authors wish to thank Wofgang Dreybrodt and Simone Milanolo for their comments that helped to im-
prove the article. tte study was supported by the fund GA205/03/1128 of the Grant Agency of Czech Republic.
REFERENCES
Absolon, K., 1970: Moravsky kras (in Czech).- Academia, pp. 416, Prague.
Appelo, C.A.J. & D. Postma, 2005: Geochemistry, groundwater and pollution, second edition.- A.A. Balkema Publishers, pp. 536, Leiden.
Atkinson, T.C., 1977: Carbon dioxide in the atmosphere of the unsaturated zone: an important control of groundwater hardness in limestones.- Journal of Hydrology, 35, 111-123.
Baker, A. & D. Genty, 1998: Environmental pressures on conserving cave speleothems: effects of changing surface land use and increased cave tourism.- Journal of Environmental Management, 53, 165-175.
Baker, A., Genty, D. & I.J. Fairchild, 2000: Hydrological characterisation of stalagmite dripwaters at Grotte de Villars, Dordogne, by the analysis of inorganic species and luminescent organic matter.- Hydrology and Earth System Sciences, 4, 3, 439-449.
Balak, I., Jančo, J., Štefka, L. & P. Bosak, 1999: Agriculture and nature conservation in the Moravian Karst (Czech Republic).- International Journal of Speleology, 28B, 1/4, 71-88.
Baldini, J.U.L., Baldini, L.M., McDermott, F. & N. Clip-son, 2006: Carbon dioxide sources, sinks, and spatial variability in shallow temperate zone caves: Evidence from Ballynamintra Cave, Ireland.- Journal of Cave and Karst Studies, 68, 1, 4-11.
Beach, T., Luzzadder-Beach, S., Dunning, N. & D. Cook, 2008: Human and natural impacts on fluvial and karst depressions of the Maya Lowlands.- Geomor-phology, 101, 308-331.
Benavente, J., Vadillo, I., Carrasco, F., Soler, A., Lian, C. & F. Moral, 2010: Air carbon dioxide contents in the vadose zone of a Mediterranean karst.- Vadose Zone Journal, 9, 1, 126-136.
Bögli, A., 1964: Mischungkorrosion - ein Beitrag zum Verstärkungsproblem.- Erdkunde, 18, 2, 83-92.
Bogli, A., 1980: Karst hydrology and physical speleology.-Springer-Verlag, pp. 284, New York.
Bourges F., Mangin A. & D. d'Hulst, 2001: Carbon dioxide in karst cavity atmosphere dynamics: the example of the Aven d'Orgnac (Ardeche).- Comptes Rendus de l'Academie des Sciences - Series IIA -Earth and Planetary Science, 333, 685-692.
Cai, B., Zhu, J., Ban, F. & M. Tan, 2011: Intra-annual variation of the calcite deposition rate of drip water in Shihua Cave, Beijing, China and its implications for palaeoclimatic reconstructions.- Boreas, 40, 525-535.
Carrasco, F., Vadillo, I., Lian, C., Andreo B. & J.J. Du-ran, 2002: Control of environmental parameters for management and conservation of Nerja Cave (Malaga, Spain).- Acta Carsologica, 31, 1, 105-122.
Covelli, S., Cucchi, F. & R. Mosca, 1998: Monitoring of percolation water to discriminate surficial inputs in a karst aquifer.- Environmental Geology, 36, 296-304.
Davies, J.C., 2002: Statistics and data analysis in geology, third edition.- John Wiley & Sons, pp. 638, New York.
Dragovich, D. & J. Grose, 1990: Impact of tourists on carbon dioxide levels at Jenolan Caves, Australia: an examination of microclimatic constraints on tourist cave management.- Geoforum, 21, 111-120.
Dreybrodt, W., 1988: Processes in Karst Systems. Physics, Chemistry and Geology, Springer Series in Physical Environment, No 4.- Springer-Verlag, pp. 288, Berlin.
Dreybrodt, W., Lauckner, J., Zaihua, L., Svensson, U. & D. Buhmann, 1996: fke kinetics of the reaction CO2 + H2O ® H+ + HCO3, as one of the rate limiting steps for the dissolution of calcite in the system H2O-CO2-CaCO3.- Geochimica et Cosmochimica Acta, 60, 18, 3375-3381.
Dreybrodt, W., Eisenlohr, L., Madry, B. & S. Ringer, 1997: Precipitation kinetics of calcite in the system CaCO3-H2O-CO2: fke conversion to CO2 by the slow process H+ + HCO3- ® CO2 + H2O as a rate limiting step.- Geochimica et Cosmochimica Acta, 61, 18, 3897-3904.
Dreybrodt, W., Romanov, D. & G. Kaufmann, 2010: Evolution of caves in porous limestones by mixing corrosion: A model approach.- Geologia Croatica, 63, 2, 129-135.
Ek, C. & M. Gewelt, 1985: Carbon dioxide in cave atmospheres. New results in belgium and comparison with some other countries.- Earth Surface Processes and Landforms, 10, 173-187.
Faimon, J., Zajiček, P. & V. Baldik, 2004a: Correlation of soil and cave carbon dioxide.- In: Bella, P. (ed.) Research, Utilization and Conservation of Caves. Proc-ceedings of 4th scientific conference, 5'h-8'h October 2003, Tale, Slovak Republic. Dugaprint, 136-137, Žilina.
Faimon, J., Zimak, J., Zajiček, P., Schwarzova, M. & J. Štelcl, 2004b: tte study of recent destructive processes in the Moravian Karst caves (Czech Republic).- Acta Universitatis Palackianae Olomucensis, Facultas Rerum Naturalium, Geographica, 38, 9-13.
Faimon, J., Štelcl, J. & D. Sas, 2006: Anthropogenic CO2-flux into cave atmosphere and its environmental impact: A case study in the Cisarska Cave (Moravian Karst, Czech Republic).- Science of the Total Environment, 369, 231-245.
Faimon, J. & M. Ličbinska, 2010: Carbon dioxide in the soils and adjacent caves of the Moravian Karst.-Acta Carsologica, 39, 3, 463-475.
Faimon, J., Ličbinska, M. & P. Zajiček, 2012: Relationship between carbon dioxide in Balcarka Cave and adjacent soils in the Moravian Karst region of the Czech Republic.- International Journal of Speleology, 41, 1, 17-28.
Fairchild, I.J., Borsato, A., Tooth, A.F., Frisia, S., Hawkes-worth, C.J., Huang, Y., McDermott, F. & B. Spiro, 2000: Controls on trace element (Sr-Mg) compositions of carbonate cave waters: implications for spe-leothem climatic records.- Chemical Geology, 166, 255-269.
Fairchild, I.J., Smith, C.L., Baker, A., Fuller, L., Spötl, C., Mattey, D., McDermott, F. & E.I.M.F., 2006: Modification and preservation of environmental signals in speleothems.- Earth-Science Reviews, 75, 105-153.
Fernandez, P.L., Gutierrez, I., Quindos, L.S., Soto, J. & E. Villar, 1986: Natural ventilation of the Paintings Room in the Altamira Cave.- Nature, 321, 586-588.
Gabrovšek, F. & W. Dreybrodt, 2000: Role of mixing corrosion in calcite-aggressive H2O-CO2-CaCO3 solutions in the early evolution of Karst Aquifers in limestones.- Water Resources Research, 36, 5, 1179-1188.
Holland, H.D., Kirsipu, T.V., Huebner, J.S. & U.M. Ox-burgh, 1964: On some aspects of the chemical evolution of cave water.- Journal of Geology, 72, 36-67.
Hoyos, M., Soler, V., Canaveras, J.C., Sanchez-Moral, S. & E. Sanz-Rubio, 1998: Microclimatic characterization of a karstic cave: human impact on microenvi-ronmental parameters of a prehistoric rock art cave (Candamo Cave, northern Spain).- Environmental Geology, 33, 231-242.
Jimenez-Sanchez, M., Stoll, H., Vadillo, I., Lopez-Chica-no, M., Dominguez-Cuesta, M., Martin-Rosales, W. & M. Melendez-Asensio, 2008: Groundwater contamination in caves: four case studies in Spain.- International Journal of Speleology, 37, 1, 53-66.
Kuzyakov, Y., 2006: Sources of CO2 efflux from soil and review of partitioning methods.- Soil Biology & Biochemistry, 38, 425-448.
Kuzyakov, Y. & A.A. Larionova, 2005: Root and rhizomi-crobial respiration: A review of approaches to estimate respiration by autotrophic and heterotrophic organisms in soil.- J. Plant Nutr. Soil Sci., 168, 503-520.
McDermott, F., 2004: Palaeo-climate reconstruction from stable isotope variations in speleothems: a review.- Quaternary Science Reviews, 23, 901-918.
McDonald, J., Drysdale, R., Hill, D., Chisari, R. & H. Wong, 2007: tte hydrochemical response of cave drip waters to sub-annual and inter-annual climate variability, Wombeyan Caves, SE Australia.- Chemical Geology, 244, 605-623.
Moncrieff, J.B. & C. Fang, 1999: A model for soil CO2 production and transport 2: Application to a florida Pinus elliotte plantation.- Agricultural and Forest Meteorology, 95, 237-256.
Musgrove, M. & J.L. Banner, 2004: Controls on the spatial and temporal variability of vadose dripwater geochemistry: Edwards Aquifer, central Texas.- Geo-chimica et Cosmochimica Acta, 68, 1007-1020.
Otava, J. & I. Balak, 2002: Moravian Karst. Geology of the protected landscape area in the ČR.- ČGS, pp. 5, Prague.
Parkhurst, D.L. & C.A.J. Appelo, 1999: User's guide to PHREEQC (Version 2) a computer program for spe-ciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations.- U.S. Geol. Surv. Water-Res. Investig. Report 99-4259, pp. 312, Denver, Colorado.
Pilegaard, K., Hummelsh0j, P., Jensen, N.O. & Z. Chen, 2001: Two years of continuous CO2 eddy-flux measurements over a Danish beech forest.- Agriculture and Forest Meteorology, 107, 29-41.
Plummer, L.N. & E. Busenberg, 1983: ^e solubilities of calcite, aragonite and vaterite in CO3-H3O solutions between 0 and 90°C, and an evaluation of the aqueous model for the system CaCO3-CO3-H3O.- Geo-chimica et Cosmochimica Acta, 46, 1011-1040.
Pulido-Bosch, A., Martin-Rosales, W., Lopez-Chicano, M., Rodriguez-Navarro, C.M. & A. Vallejos, 1997: Human impact in a tourist karstic cave (Aracena, Spain).- Environmental Geology, 31, 143-149.
Qian, H. & L. Peiyue, 3011: Mixing Corrosion of CaCO3 in Natural Waters.- E-Journal of Chemistry. 8, 3, 1134-1131.
Russell, M.J. & V.L. MacLean, 3008: Management issues in a Tasmanian tourist cave: Potential microclimatic impacts of cave modifications.- Journal of Environmental Management, 87, 474-483.
Sanchez-Moral, S., Soler, V., Canaveras, J.C., Sanz-Rubio, E., Van Grieken, R. & K. Gysels, 1999: Inorganic deterioration affecting the Altamira Cave, N Spain: quantitative approach to wall-corrosion (solutional etching) processes induced by visitors.- Science of the Total Environment, 343/344, 67-84.
Song, L., Wei, X. & F. Liang, 3000: tte influence of cave tourisms on CO3 and temperature in Baiyun Cave, Hebei, China.- Internationl Journal of Speleology, 39B, 1/4, 77-87.
Spötl, C., Fairchild, I.J. & A.F. Tooth, 3005: Cave air control on dripwater geochemistry, Obir Caves (Austria): Implications for speleothem deposition in dynamically ventilated caves.- Geochimica et Cosmochimica Acta, 69, 3451-3468.
Stumm, W. & J.J. Morgan, 1996: Aquatic chemistry (3'd edition).- John Wiley and Sons, pp. 1040, New York.
Tooth, A.F. & I.J. Fairchild, 3003: Soil and karst aquifer hydrological controls on the geochemical evolution of speleothem-forming drip waters, Crag Cave, southwest Ireland.- Journal of Hydrology, 373, 51-68.
Uchida, M., Nojiri, Y., Saigusa, N. & T. Oikawa, 1997: Calculation of CO3 flux from forest soil using 333Rn calibrated method.- Agricultural and Forest Meteorology, 87, 301-311.
Vokal, B., Obelic, B., Genty, D. & I. Kobal, 1999: Chemistry measurements of dripping water in Postojna Cave.- Acta Carsologica, 38, 1, 305-331.