RELATIONSHIP BETWEEN CARBON
CYCLE IN KARST AREAS AND CO₂
SOURCE-SINK OF ATMOSPHERE
-Case of Guizhou Karst

Li Bin      Yuan Daoxian

The influence of the increase of CO₂ in the atmosphere on global climate has aroused the attention of the scientific and political circles of today. A series of large research programmes, e. g. World Climate Research Programme (WCRP), the International Geosphere -Biosphere Programme (IGBP), Human Dimensions of Global Environmental Change Pro-gramme (HDGECP) and so on, has been operating^[1]. Meanwhile, in the various global monitoring programme of the atmospheric CO₂, each programme is aimed at different scientific purposes and adopts various monitoring and analytical methods to understand the global atmospheric CO₂ distribution and variations at different spatial and temporal scale, the ex-change characteristics and the flux at different interfaces, thus distinguishing the sources and the sinks of CO₂. But the estimations show that there isn't a balance between the source and the sink, that is, there is an unknown sink (Missing Sink). Based on the estimation from 1980 to 1989 of IPCC, the unknown sink is 1. 8*1. 4 Gt C/a. and Tans et al. give a result of 2. 0-4. 7 Gt C/a^[2] (from 1981 to 1987). The common carbon cycle models attribute the un-known sink to terrestrial vegetation or soil, but there have been no clear evidences. To counter this problem, this paper tries to put forward a new idea on the basis of discussion on the carbon cycle characteristics of the largest global carbon reservoir-carbonate rocks.

1 RELATIONSHIP BETWEEN THE CARBON CYCLE IN KARST SYSTEMS AND CO₂ CYCLE

Simply, karst dynamic system is a system taking the carbon cycle as the dominant process, and coupling with "CO₂-H₂O-CO₃^2-" triphase nonequilibrium system. Thus it can be seen that carbon cycle intensity is closely related to the CO₂ cycle. In order to study the cor-relation, seven sites in Guizhou karst area under different ecological, geological (especial lithological). conditions were selected for annual dynamic observation (Maolan of Lipo County) and different season observation (other 6 sites) of the soil CO₂ concentration, and pH , water temperature and temporary hardness of the epikarst springs so as to catch the carbon cycle trace and the correlation between various spheres.

Fig.1* shows the correlation between the annual soil CO₂ concentration and the HCO₃^-concentration of the epikarst spring at the Wangpaishan observation site of Maolan Karst Forest Reserve , that is, the HCO₃^- concentration increases with the CO₂ concentration in soil that is controlled by ecological environment. According to the basic chemical corrosion prin-ciple of carbonate rocks (Eqs. 1, 2), during the corrosion, atmospheric CO₂ is consumed. The field observation results above also show the correlation.

Fig 1. Relationship between soil CO₂ concentration and HCO₃^- concentration
in epikarst spring, Wanpaishan site, Maolan

      At the same time, the observation results under different ecological and geo-logical con-ditions show that because of thedifferent ecological condition the CO₂ concentrations in soil are different, and the intensity of the carbon cycle in the karst system different too (Fig. 2). Fig.2 also shows that under different lithologic conditions, even if the CO₂ cycle intensity is similar, the intensity of karst process is different. Usually, the carbon cycle in limestone is more intensive than that in dolomite.Therefore, in chemical corrosion, the CO₂ consumption is different for various carbonate rocks.

Fig 2. Relationship between soil CO₂ concentration and HCO₃^- concentration in epikarst springs under different ecologic environment.

   Meanwhile, it can be seen from the chemical models above that karstification is not only chemical corrosion, but epigenetic chemical sedimentation (e.g. speteothem, valley calcareous tufa, water-fall calcareous tufa, etc. ) may also take place once conditions (hydrochemical or hydrogeological) change. Accompanying the sedimentation, the CO₂ in the karst systems may be released into the at-mosphere, and becomes a source of the atmospheric CO₂. To sum up, the carbon cycle intensity is closely related to the CO₂ cycle intensity. And under different karst processes, the carbon cycle may result in either a source or a sink.

______________________________________________
* Surface water mixing up on 2nd of May. 29th of Jun. , 1st of Aug.

2 ESTIMATION OF ATMOSPHERIC C02 SINK IN THE CORROSION OF THE CARBONATE ROCKS IN GUIZHOU AREA

As mentioned above, a karst system is both a source and a sink, and after all, what a part do karst systems play in source-sink of the atmospheric CO₂? Which is greater? So far, there has not been a clear understanding or no final conclusion has yet been reached on this problem.Accordingly, this paper takes the Guizhou karst area as an example to estimate the source and the sink, and expect to get an initial conclusion.

2.1 The Basic Principles and Meth-ods of Estimation
     Based on the simple chemical model of carbonate rock corrosion and the correlation analysis above, during the corrosion of carbonate rocks, half of the HCO₃^- ions re-leased in solution comes from the at-mospheric CO₂, which provides a way to estimate the flux of atmospheric CO₂ consumed by rock corrosion. The second method is based on the hydrochemical principle, that is, to estimate the corrosion rate of the carbonate rocks by the runoff and concentrations of the major dissolved elements in a drainage basin. The third method is based on field dissolution test to get the corrosion rate in a region, then to estimate the flux of consumed atmospheric CO_2.

2.2 Estimations

2.2.1 Method based on simple chemical model
    Carbonate rocks are widely distributed in Guizhou, making up about 73% of the total area in the province. Karst springs and subterranean rivers are intensively developed. The karst water resource make up about 80% of the total water resources in the province*. Therefore, the area is one of the ideal areas for estimation of the flux of atmospheric CO₂ consumed in karst processes. As is known to all, karat water types are various, it may be difficult to estimate classi-fiedly one by one. But in Guizhou area, the subterranean rivers are the major discharge of the karat water resources. Most of karst water resources join with surface water systems by the subterranean rivers or the big karat springs, and join the global water cycle. According-ly, the subterranean river water resources may well reflect the processes and the quantity that the carbon in the karst area joins the global carbon cycle. Therefore, in the estimation, we adopt the subterranean river water resources and their average HCO₃^- concentration. According to Yang Lizheng's statistics^[3] the mean discharge of the subterranean rivers in Guizhou is about 572 m³/s, that is the annual total discharge is 1.8038 x10¹⁰ m³. Mean-while, we sampled and collected about 123 hydrochemical data of the subterranean rivers and got their average HCO₃^- concentration of 233. 89 mg/l.
    Based on the data and the simple chemical model above, the flux of atmospheric CO₂ consumed by the carbonate rock corrosion in Guizhou is:

where, [HCO₃^- ] is HCO₃^- concentration in karst water; V_H2O is the total water resources of the subterranean rivers in Guizhou.
      The distribution area of the carbonate rocks in Guizhou is about 1.3x10⁵ km², accord-ingly, the consumed annual specific area carbon flux is about 3.192 x l0⁶ g C/a .km².
2. 2. 2 Method based on hydrochemistry
      According to hydrochemical calculation of Fang Jinfu et al.^[4], the corrosion rate in Hongshui river valley and its adjacent regions is 40~90 mm/ka, that is, the unit annual cor-rosion is 40-90 m³/a .km².
   If the specific weight of the carbonate rocks is 2. 7, the corrosion of the carbonate rocks per year and per square kilometer in the area is: 108-243 t CaCO₃ or CaMg(CO₃)₂. Based on the chemical model above, to dissolve 1t CaCO₃ needs to consume 1.2 x 10⁵gC CO₂Accordingly, the flux of atmospheric CO₂ consumed by corrosion of the carbonate rocks is: 1. 296 x 10⁷ ~2.916 x 10⁷ gC/a. km². As to whole Guizhou karst area, the total flux is: 1.684 x 10¹²~
3. 791 x 10¹² gC/a.

2. 2. 3 Method based on field dissolution test
   The standard limestone tablets (diameter of 4 cm, thinkness of 0.1-0.3 cm) are usual-ly used for the field corrosion observation. Based on the multi-year observation results in
____________________________
* Yang Mingde, Deng Zimin. Correlation Analysis Between Guizhou Karat and Its Economic Development (in Chinese)

Guiyang City of Guizhou, the average corrosion quantity is 13. 9 x l0^-2mg/d on the ground,
9. 05 x 10^-2 mg/d at 20 cm beneath the ground surface and 7. 67 x 10^-2 mg/d at 50cm beneath the ground surface, averagely, 10.2067 x 10^-2 mg/d^[5]. Accordingly, the annual aver-age corrosion quantity is 37. 2543 mg/a.
    The total area of a standard tablet is 25. 12 cm². Therefore, the annual unit corrosion quantity is 1.483mg CaCO₃/a.cm², that is: 1.483 x 10⁷gCaCO₃/a . km². Accordingly, the flux of atmospheric CO₂ consumed by the corrosion of the carbonate rocks is: 1.7796 x 10⁶ gC/a .km².

2. 3 Assessment of Different Estimation Results

   In view of the different estimation methods above, the results aren't exactly the same. Some are similar, some have a great difference.
       As mentioned above, different methods are based on different principles, supposed conditions and influence factors. The result of the method based on the simple chemical model reflects a general information, and is not strictly restricted by lithology, climate, etc. , and may be objective. But this result is also influenced by the accuracy of the resources and the HCO₃^- concentration. Therefore, it is better to estimate the flux in each river basin or karst region. The method based on hydrochemistry is restricted by many factors including geologic conditions (physical, chemical characteristics of the rocks, thickness of strata, lithochemical composition and so on) , ecological and climatic conditions, etc. . The result of field corro-sion test is influenced by the limitations and the representativeness of the test conditions, as well as some undefined factors in the processes. Therefore, each method has its own limita-tion. If possible, it may be necessary to estimate the flux to each river basin under various environmental and lithologic conditions, and then to set up a global model.

3 ESTlMATION OF THE FLUX OF ATMOSPHERIC CO₂ RELEASED IN EPIGENETIC CHEMICAL SEDIMENTATION-CASE OF HUANGGUOSHU WATER FALL

   The epigenetic chemical sediments in karst area are common, e. g. speteothem, valley calcareous tufa, water-fall calcareous tuta, calcareous sinter and so on. These chemical sedi-mentation is often related to CO₂ release caused by the change in hydrodynamics or hydrochemistry. Therefore, accompanying these sedimentations, the CO₂ in karst system may be-come a source of atmospheric CO_2. But it is difficult to estimate the release quantity of each kind of sedimentation. So in order to show the characteristics and the quantity of the CO₂ release, we selected Huangguoshu Water Fall, and measured the hydrochemical data on the -spot and sampled the water for chemical analysis in laboratory at twice. The results are shown in Tab. 1.

Tab. 1 shows that the hydrochemical characteristics at the upper and the lower reaches of Huangguoshu Water Fall well reflect the formation mechanism of the waterfall calcareous tufa. Because of the intensive disturbance during water's falling, the CO₂ in karst water is quickly released, which results in the water to be saturated and the tufa to be formed. Therefore, the chemical process is the basic principle of the tufa formation. The various chemical indexes also reflect the process, that is, with the release of CO₂, pH of the water increases and the partical pressure of CO₂ decreases. Meanwhile, with the tufa formation, the HCO3^- , Ca²⁺, Mg²⁺ concentrations decrease too. Therefore, it is feasible to apply the simple chemical model to estimate the release flux of CO₂ in the formation process of valley and water fall calcareous tufa.

Tab. 1 Hydrochemical characteristics of Huangguosbu Water Fall

   According to Yu Jinbiao et at.^[6], the multi-year (1983~1986)mean total annual dis-charge is 5.7525 x 10¹¹l.
    Tab. 1shows that the mean difference in HCO3^- concentration between the water of the upper reaches and the water of the Shuilian Cave (middle of the water fall) is 15. 355 mg/ l.
   Based on the simple chemical model, the annual carbon flux released in the chemical sedimentation at Huangguoshu Water Fall is: 8.7 x 10⁸ gC/a.
   Tab. 1 also shows that the hydrochemical characteristics, e. g. HCO3^- , Ca²⁺, Mg²⁺ concentrations at Xiniutan Pond (water failing pond) are similar to those at the upper reaches. The fact shows that during water's failing, the intensive disturbance may result in the re-lease of CO₂ and the formation of tufa, but when the water is in unsaturated state, the corro-sion may happen again. Therefore, under such condition, the sedimentation may vary with the corrosion, and the source may be in balance with the sink.

4 POSITION AND ASSESSMENT OF THE CARBON CYCLE FLUX OF KARST AREA IN THE GLOBAL CO₂ CYCLE

    According to the estimation of the sink in the corrosion of the carbonate rocks and the source in the epigenetic chemical sedimentation in Guizhou karst area, the sink may be larger than the source. In addition, the carbonate rocks are widely distributed in the world, no matter how intensive the carbon cycle in biosphere and atmosphere is, the corrosion always happens, however the intensity is different. And the epigenetic chemical sedimentation may happen locally or under some special environmental conditions. Therefore, in the global viewpoint, the flux of atmospheric CO₂ consumed by the Corrosion may larger than that re-leased by the chemical sedimentation.

Tab. 2 list the flux of atmospheric CO₂ consumed by the corrosion in Guizhou karat area and on a global scale. It can be seen that there is a certain difference between various results, but they can show the role of the flux in the global carbon cycle.

Tab. 2 Flux of atmospheric CO₂ consumed by corrosion of carbonate rocks*

    As mentioned above, the role of the largest global carbon reservoir-carbonate rocks in the global carbon cycle can't be neglected. According to the field corrosion tests at different regions in China, the corrosion capacity in Guizhou is smaller than that in other areas of south China, e. g. the corrosion capacity is 10.2 x 10^-2 mg/d at Guiyang of Guizhou, 59.02 x 10^-2 mg/d at Guilin of Guangxi, 26.82 x 10^-2 mg/d at Kunming of Yunnan. In addition, based on hydrochemical estimation, the range of the corrosion capacity under different cli-matic zones in China is from 10 m³/a . km² to 200 m³/a . km², averagely 110 m3/a . km^{2 [7]}. Accordingly, the value in Guizhou is also lower than the average value. Therefore, if the flux of atmospheric CO₂ consumed in the corrosion of carbonate rocks under different climatic zones and environmental conditions can accurately be estimated, the unknown sink in the global carbon cycle may be evaluated more accurately.

5 CONCLUDING REMARKS

   Based on the geochemical tracing studies on the carbon cycle between the epikarst zone and its adjacent spheres under various ecological and geological environments, it is found that the carbon cycle intensity has a close relationship between the epikarst zone and its adjacent spheres. This good correlation shows the relationship between the corrosion of the carbonate rocks in karst area and the CO₂, cycle. Accordingly, the estimation of the flux of atmospheric CO₂ consumed in the corrosion of the carbonate rocks in Guizhou and in the world shows that
__________________________
*The global flux in estimated on the basis of the total area of the carbonate rocks in the world of 2.2 x 10⁷ km².

the sink in the global carbon cycle model can't be neglected. The sink may be one of the un-known sinks. Meanwhile, the sink estimation of the CO₂ flux released in the epigenetic chemical sedimentation of Huangguoshu Water Fall, and the related study show that the sink and the source under this kind of condition may be in dynamic balance.
   Of course,the study and estimation is initial. Both the representativeness and the exten-siveness lack depth. Because the global carbonate rocks are distributed in different climatic zones and under different geological and ecological environment and the estimation also re-lates to the rocks themselves, e. g. the physical and chemical characteristics, the texture and structure, the distribution area, the thickness and so on, it is necessary to cooperate on a global scale and to make systematic observation and estimation in connection with different regions (or drainage basins), different environmental conditions, so as to perfect the global carbon cycle model.

Acknowledgments

   This study was jointly founded by State Key Laboratory of Environmental Geochemistry and Ministry of Geology and Mineral Resources (Grant No. 8502218).

References

1 Wang Gengchen. Advances in global monitoring of CO₂ concentration in the atmosphere. Advance in Earth Sciences. 1994. 9(4),70-77(in Chinese)

2 Sarmiento.J.L.and Orr, J.C.. A perturbation simulation of CO₂ uptake in an ocean general circulation model. Journal of Geophysical Research, 97(C3), 3621-3645

3 Yang Lizheng. Distribution of subterranean rivers in south China. Carsologica Sinica, 1985, 4(l.2),92--100 (in Chinese with English abstract)

4 Fang Jingfu, Lin Junshu et al.. Study on present corrosion intensity and environment. Acta Geographica Sinica, 1993. 48(2), 122-130 (in Chinese)

5 Yuan Daoxian, Cai Guihong. The Science of Karst Environment. Chongqing: Chongqing Publishing House, 1988, 65 (in Chinese)

6 Yu Jinbiao, Zhang Haisheng. The Study on the Tourism Resources in Huangguoshu Area. Shanghai : Shanghai Scientific and Educational Publishing House, 1990, 173-179 (in Chinese)

7 Weng Jintao. The effect of carbonate rocks on global carbon cycle. Advance in Earth Sciences, 1995. 10 (2),154-158 (in Chinese)

Comments and suggestions to Webmaster,
Copyrights by Karst Dynamics Laboratory and Network Center of Guangxi Normal University,
1995-1999. All rights reserved.