Published in C.R.Acad. Sci. Paris, t. 324, serie II a, p. 17 a 23, 1997.
Philippe Gombert

1. The Problem of Karstic Denudation

At the beginning of the century, karstologists thought higher karstic denudation rates took place under the most humid climates, i. e. tropical climates. Then Corbel (1957) proved that cold climates also have a certain influence and he proposed a formula for specific dissolution calculation X (mm/millenary): X= 0.4× E× T/N where E is annual rainfall (m), T the carbonate concentration of karstic water (mg/1) and N die proportion of limestone in the catchment basin.

Nevertheless, rainfall was always considered as the major karstic denudation agent and several relationships between karstic denudation and rainfall have been proposed (Pulina, 1974; Lang, 1977). More recently, karstic hydrogeochemistry showed the importance of water inflow enriched with dissolved CO ₂ (Nicod, 1975; Bakalowicz, 1979 a; Ford, 1985; Maire, 1990). However up to now Corbel's formula remains the most usual method for calculating the karstic denudation rate.

Corbel's formula was later used by Williams (1963) who integrated spring flow, catchment basin area and limestone density. But all these parameters need a sound knowledge of the karstic aquifer system and regular hydrochemical analysis and spring flow measurements over several hydrogeological cycles (Bakalowicz, 1979 b). This is generally not the case in karstic countries developed under extreme climate, for example in tropic monsoon or polar regions.

During an international congress in 1978, limestone tablets were distributed to karstologists from the world over for them to measure the loss of weight after exposure to climatic or pedologic agents. These measures, however gave only 26% of the calculated specific dissolution (Gams, 1985). With a multili-near regression analysis, Gerome-Kupper (1984) noted that rainfall and temperature effectively explained together most of the variance: but only 32% of the total variance of the statistical sample could be so explained. This method cannot be used for comparison of karstogenesis in different climates.

Geochemical balance is certainly the most precise method for karstic denudation measurement. However it needs a perfect knowledge of karstic aquifer and requires daily or weekly water analysis and flow measurements for each hydrogeological cycle (Marjolet and Salado, 1975; Bakalowicz, 1979b; Blavoux and Mudry,1988).

A mathematical approach called "Maximum Potential Dissolution" (DMP in French) is presented here. Taking into account infiltration, CO₂ soil productivity and water concentration in CaCO₃ resulting from calcocarbonic equilibrium, the calculation is performed at an annual scale and based upon the following assumptions: (1) rainfall water enters the calcareous karstic system with an inflow corresponding to the infiltration, (2) water immediately dissolves CO₂ from the soil according to its average temperature, (3) water immediately dissolves CaCO₃ up to saturation (according to Dubreucq et al., 1988). This model leads to the DMP formula: DMP = 400× Ie× C where Ie is infiltration (mm) given by Coutagne's formula (in Remenieras, 1986) and C is water concentration of calcite at saturation (mg/1) give by calcocarbonic reactions (Trombe, 1952) all of whose equilibrium constants K_i. can be approximated by an equation involving absolute temperature (T, ° K) and a set of tabulated parameters a_i, b_i and c_i. (tableI):

log K_i = a_i + b_i / T + c_i× T

CO₂ partial pressure of the phase associated with water solution is given by Brook's (1983) formula as a function of Annual Evapo-Transpiration (AET, mm):

DMP final expression is:

This results are in good agreement with the values of specific dissolution and geochemical balance known all over the world (table II). This expression is not easy to handle because of the complexity of development parameters, but all of them depend only on average rainfall and temperature values. Consequently, we tried to simplify DMP by definiing it in relation to a climatic index. Statistical calculation is performed for 266 meteorological stations all around the world (Miller, 1966; Maire, 1990) and the best adjustment is obtained for the index i₂₀ = R / (T+ 20) where R is the annual rainfall (mm) and T the mean annual temperature (° C). The DMP equation corresponds to the following 3rd degree polynomial:

DMP = - 4.14´ 10^-5 i₂₀³ +1.93´ 10^-2 i₂₀² + 8.02´ 10^-1 i₂₀ - 10.5

We can now suggest an average karstic denudation rate value for each climate (table 11) and calculate that higher DMP values correspond well to warm humid climates as observed all around the world.

DMP calculation includes current knowledge on main agents of karstic denudation:

infiltration, calcocarbonic equilibrium and CO₂ productivity of the soil. It can be calculated everywhere in the world using only annual rainfall and average temperature. Moreover, its application to paleoclimates has been successfully tried by Gombert (1988).

J. Speleol. Soc. Japan, 20: 27-37, March 30, 1996

Kazuhisa YOSHIMURA¹⁾ , Mari IWAYA-INOUE²⁾, Takashi SOMEYA³⁾,
Kentaro MATSUURA⁴⁾ , Mihoko HASHIGUCHI²⁾,
Tomoko HARA⁵⁾, Hisaki OHTSUBo⁶⁾ and
Shiro MATSUOKA¹⁾

1) Department of Chemistry, Faculty of Science, Kyushu University Ropponmatsu,
    Ropponmatsu, Chuo-ku, Fukuoka, 810 Japan
2) Department of Agronomy, Faculty of Agriculture, Kyushu University Ropponmatsu,
    Ropponmatsu, Chuo-ku, Fukuoka, 810 Japan
3) Department of Applied Biology, Department of Agriculture,
    Saga University, Honjo, Saga, 840 Japan
4) Department of Agricultural Chemistry, Faculty, of Agriculture, Kyushu University,
    Hakozaki, Higashi-ku. Fukuoka, 812 Japan
5) Department of Pharmaceutical Sciences, Faculty of Pharmaceutical Sciences,
    Kyushu University, Maidashi, Higashi-ku, Fukuoka, 812-82 Japan
6)Department of Materials Science and Engineering, Faculty of Engineering,
    Kyushu University, Hakozaki, Iligaslii-ku, Fukuoka, 812 Japan

Abstract The cascade at the entrance to Akiysohi-do Cave, Yamaguchi, is made of biologically induced carbonate, tufa. The groundwater issuing from the cave was supersaturated for calcite and the degree of supersaturation was the highest at the entrance to the cave. The pH of the tufa suspension increased with the light exposure. Photosynthetic cyanobacteria were predominant among algal flora on the surface of the tufa deposits distributed on sites of lower light intensity, while not on sites of higher light intensity. The CaCO₃ local supersaturation on the surface of tufa is caused by the local increase in the CO₃^2- concentration because the bacteria consume the dissolved CO₂ and/or HCO₃^- and the pH of the solution is increased. Tufa deposition at the entrance to the cave is, therefore, related to the activity of cyanobacteria having a sticky mucous sheath and to the resulting CaCO₃ precipitation around the bacteria.

Tufa in Limestone Areas in Southwest Japan
J.Speleol. Soc. Japan, 20:19-26, March 30, 1996

Kazuhisa YOSHIMURA¹⁾, Kensaku URATA²⁾, Akiniro KANO³⁾,
Youji INOKURA⁴⁾ and Yukimasa HONDA⁵⁾

1)Department of Chemistry, Faculty of Science, Kyushu University Ropponmatsu,
Ropponmatsu, Chuo-ku, Fukuoka, 810 Japan
2)Department of Geography), Faculty of Science, Tokyo Metropolitan University,
Hachioji, Tokyo, 192-03 Japan
3)Department of Earth and Planetary System, Faculty of Science,
Hiroshima University, Higashihiroshima, Hiroshima, 724 Japan
4)Research Division of University Forests, Faculty of Agriculture,
Kyushu University, Sasaguri, Fukuoka, 811-24 Japan
5)Edogawa Caving Club, Flat A, 15/F, Hoilee Bdg., 58-60 Saiwanho St.,
Shaukeiwan, Hong Kong

Abstract Tufa is biologically induced carbonate deposited in freshwater. Tufa deposition is related to the growth of photosynthesizing organisms such as cyanobacteria and also to the rapid CaCO₃ precipitation around them: the CaCO₃ local supersaturation on the surface of tufa is one of the most important factors. Tufa deposits in some limestone areas in Southwest Japan: Akiyoslii-dai Plateau (Yamaguchi), Hirao-dai Plateau (Fukuoka), Shirokawa (Ehime) and the Gizabanta Coast (Okinawa), are here reported for the first time. The widest tufa deposit occurs in Shirokawa: it is 380 m long, 10 m wide, and over 1 m thick. The depositional rate is so high that even a mislaid cotton glove was encrusted with tufa. Tufa deposits exhibit annual layers due to seasonal variations in biological activities. In Akiyoshi-do Cave, Yamaguchi, the groundwater flows above the groundwater level, because the hydraulic gradient profile from Kotogafuchi (the groundwater located at the farthest place from the entrance) to the entrance to the cave, has been maintained constant by the tufa cascade located at the entrance to the cave. The CaCO₃ supersaturation in the stream water, the absence of big storm runoffs, and the lack of human interference, may be critical factors in tufa formation. Further research on tufa is now continuing.