Abstract
The degassing systems installed in the early 2000s at lakes Nyos and Monoun, Cameroon, have been working well, resulting in significant removal of dissolved gas. However, the systems of both lakes started losing their capability due to a reduced CO2 partial pressure in the bottom waters, especially after installation of additional pipes in 2011. After initiation of a degassing operation, changes in CO2 profiles in the bottom layer of Lake Nyos over time showed that gas-poor shallow water has descended to the bottom, leaving little CO2 in the bottom water. The degassing system at Lake Monoun has completely lost its gas self-lift capability as the reduced CO2 partial pressure in the bottom water is too low to sustain the gas-lift system, and all the degassing pipes stopped working in 2010. This situation led to the accumulation of CO2 due to continued natural recharge of magmatic CO2. To compensate this recharge of gas, we installed a new deep water removal system that is driven by solar power at Lake Monoun in December 2013. This system does not need power lines, fuel or complicated maintenance, thus it is convenient for remote lakes such as lakes Monoun and Nyos.
In the mid-1980s, the sudden release of a huge amount of CO2 from lakes Nyos and Monoun in NW Cameroon killed approximately 1800 people and an uncountable number of cattle around the lakes. This phenomenon was later called a ‘limnic eruption’ by J.C. Sabroux (Halbwachs et al. 2004). The follow-up geochemical, geological and limnological surveys clarified a possible mechanism for the limnic eruption, although the trigger of the eruption is still under debate (Giggenbach 1990; Evans et al. 1993, 1994; Kusakabe et al. 2000, 2008). The results of geochemical monitoring of the lakes indicated re-accumulation of CO2 in bottom waters even after the loss of a large amount of gas at the time of the limnic eruption, suggesting the possibility of further eruptions at both lakes (Kusakabe et al. 2008; Kusakabe 2015).
Degassing systems were installed in the early 2000s to make the lakes safe. In these systems, a plastic pipe was extended close to the bottom. Upon removal of shallow water inside the pipe, deep CO2-rich water rises, becomes CO2-saturated and starts bubbling, thus gas self-lift is primed, leading to CO2-rich bottom water being pumped out. The advantage of gas self-lift is that no additional energy is required. The first pipes were installed at Lake Nyos in 2001 and at Lake Monoun in 2003 (Halbwachs et al. 2004). Two additional degassing pipes were installed at lakes Monoun and Nyos in 2006 and 2010, respectively. The degassing went well and the amount of dissolved CO2 in both lakes decreased with time. As degassing proceeded, CO2-rich bottom water was removed, leading to subsidence of shallow water with a lower CO2 concentration and partial pressure of CO2 at the intake depth was also lowered. This resulted in weakening of the degassing power and the height of the fountain decreased, as shown in Figure 1 (Lake Nyos). At Lake Monoun degassing finally ceased in March 2012.
Comparison of fountain height in 2001 and 2013 at Lake Nyos.
The purpose of this study is to show the relationship between the recent changes in CO2 concentration in deep water and decreasing activity of the degassing system. We describe a new solar powered deep water removal system that was installed at Lake Monoun where gas self-lift capability of the degassing pipes was lost and gas recharge resumed. Installation of this system was intended to make Lake Monoun completely gas-free.
Recent changes in CO2 concentration in deep water at Lake Nyos
Monitoring of deep water CO2 concentrations is essential to evaluate the risk of recurrence of limnic eruptions (Kling et al. 2005; Kusakabe et al. 2008; Yoshida et al. 2010). The CO2 concentration along the vertical water column of Lake Nyos was measured using the method described in Yoshida et al. (2010). We collected deep water through a long plastic hose, a miniature version of the degassing pipe, which was deployed to the bottom as described in Nagao et al. (2002) and Tassi et al. (2009). The two-phase flow in the hose was separated into gas and liquid using a separator. The volume of each phase was accurately measured with a gas flow meter and a glass measuring cylinder. This method has two advantages: easy handling and data acquisition on site. The main disadvantage of this method is that it is difficult to apply to water in which the CO2 concentration is not high enough to keep continuous gas self-lifting. It is possible to measure the CO2 concentration of shallower lake water by using a potable pump; however, monitoring at the deep layer is required to evaluate the effect of the degassing system.
We started the CO2 measurement from the deepest water, moving gradually to shallower levels by pulling the hose upward. We took six measurements from January 2009 to March 2014 at Lake Nyos. In a deep layer, where the chemocline develops (c. 180 m), we measured the CO2 concentrations at an interval of 0.5 or 1.0 m, and at an interval of 5 or 10 m in water shallower than 170 m. Table 1 shows the results. The CO2 profiles are shown in Figure 2, which indicates a sharp decrease in CO2 concentrations, especially in the deepest water. Since the bottom water has been continuously transported to the surface, the CO2 profiles subsided with time. The CO2 concentrations shallower than 170 m were also reduced, but the shape of the CO2 profiles has not changed with time, suggesting that a stable structure of mid-depth water was maintained during degassing. From the changes in the profiles over time we can estimate a rate of profile subsidence. As shown in Figure 3, the CO2-rich water in the deepest zone subsided by 1–3 m between January 2009 and January 2011, by 5–8 m between January 2011 and December 2011, and by 3 to 4 m between December 2011 and March 2012. Based on these results, we obtained the average rates of subsidence for those three periods to be 1–3 m a−1, 5–8 m a−1 and 12–16 m a−1, respectively. Installation of additional degassing pipes in early 2011 and adjustment of location and intake depth of the pipes (IRGM, unpublished report) contributed to the increase in the rate of subsidence. The rate, however, will not continue in the future because the bottom water removal rate will become low following the reduction of CO2 concentration. This phenomenon has already been observed in the decreasing height of the degassing fountains of both lakes (Figs 1 & 4).
Changes with time in the CO2 profiles at Lake Nyos.
Enlargement of the CO2 profiles of the bottom layer at Lake Nyos.
Comparison of fountain height in 2005 and 2007 at Lake Monoun.
CO2 concentration in deep water of Lake Nyos between January 2009 and March 2014
The effect of degassing can be evaluated from the change in the amount of remaining CO2 in the bottom water. We calculated the amount of CO2 removed during the above period using the respective profiles and the volume of water below 180 m. We used the bathymetry of Kling et al. (2005) for this calculation. Table 2 shows the amount of CO2 removed and the annual removal rate of CO2 between 180 and 210 m depth. The amount of dissolved CO2 decreased from 2.97 Gmol in 2009 to 1.78 Gmol in 2014. The average removal rate decreased sharply after March 2012. This resulted from decrease in CO2 concentration of the bottom water as shown in Figure 2.
The amount of CO2 dissolved in the bottom water at Lake Nyos in 2009–13 and the change of CO2 removal rate with time
Decreasing activity of degassing systems
Decreasing activity of the degassing systems can be understood visually from decreasing fountain height. Figures 1 and 4 compare the fountain height at different times at lakes Nyos and Monoun, respectively. The fountain height of Lake Nyos was 45 m at the beginning of the degassing operation (31 January 2001), then gradually lowered year by year, and came to only 4.6 m in March 2014. We checked inside all the degassing pipes and confirmed that the lowering of fountains was caused not by obstacles such as scaling or collapse of the pipes. This suggests that the CO2 concentration was effectively decreasing. Figure 5 shows the changes with time in the CO2 concentration at 206 m (intake depth of degassing pipes) and the amount of CO2 remaining in the bottom water (180–210 m deep). During degassing with a single pipe (before 2011), the amount and concentration of CO2 decreased slowly. The decrease rate was naturally accelerated after installation of two additional degassing pipes in early 2011. Figure 5 also suggests that the decreasing rate of the amount of the remaining CO2 will gradually slow down in harmony with the decreased CO2 concentration, and eventually the degassing at Lake Nyos will cease in the near future as was the case at Lake Monoun (Kusakabe 2014). Once the degassing fountains have ceased, they cannot restart spontaneously. This means CO2 will buildup again at Lake Nyos due to continued natural recharge of magmatic CO2, implying an ongoing risk of recurrence of limnic eruption. The best way to avoid this risk is to remove the bottom water artificially. Since a power line is unavailable and not easy to access, especially in the rainy season at Lake Nyos, a system that works without external power is required, preferably a solar powered pump system, as proposed by Yoshida et al. (2010, 2013). This system needs to be less expensive and free of human intervention.
Changes over time in the amount and concentration of CO2 at Lake Nyos. Open circles and open triangles denote CO2 concentration at 206 m depth (left y axis) and the amount of CO2 below 180 m (right y axis), respectively.
Solar powered pumping system
Figure 6 shows the solar powered pumping system (maximum output of the solar power system is 320 W and maximum flow rate of the pump is 2.8 m3 h−1) installed at Lake Monoun. In order to reduce the total cost of installation of the system, we used an existing degassing pipe which had already lost its function. The intake depth of the pipe was 99 m, very close to the bottom. As its inner diameter is only 100 mm, we chose a small rotary water pump with an outer diameter of 74 mm. Since the pump does not need high electric power, we used four small solar modules. The size of each module was 120 cm by 54 cm in dimension and the maximum output of each module was 80 W. We set up the solar modules on an aluminum frame that sits on a new raft. The new raft was connected to the old raft which had been used for the degassing pipe. Naturally, electric power is supplied to the pump only in daytime. A gas self-lift was induced by the up flow of bottom water in the pipe after the system started working, and the outflow did continue even at night. With this system CO2-rich bottom water can be efficiently uplifted to the surface. In March 2014, we measured the daily removal rate of deep water. It was 25 m3 d−1. Flow rate of a gas self-lift observed by eye-measurement is about five times as much as the flow through the pump. The overall flow rate was estimated to be more than 100 m3 d−1 if water removal by gas self-lift is included. Based on this rate and the CO2 concentration of deep water measured in March 2014 (90 mmol kg−1 in bottom water, as shown in Fig. 7), the annual removal rate of CO2 was calculated to be 3.3 Mmol a−1. Since this removal rate is half of the natural recharge rate (8.2–8.4 Mmol a−1, Kling et al. 2005; Kusakabe et al. 2008), it is advisable to install 2–3 additional systems at Lake Monoun to compensate the natural CO2 recharge. We confirmed that the output of the solar modules, pumping rate, conditions of electric controller and other parts were all satisfactory in December 2014, indicating the perfect performance of the system. The water removal rate also stayed unchanged. This means that the system is robust and can keep working for a long time. Another important point is that the system does not need complicated maintenance and fuel, an important factor that any system should have in a remote area in Cameroon like lakes Nyos and Monoun.
Solar-powered deep-water removal system installed at Lake Monoun. (photo: M. Kusakabe.)
CO2 profile of the bottom water at Lake Monoun in March 2014. The CO2 concentration of the bottom water was too low to induce a spontaneous self-lifting flow before 2013.
Conclusions
The amount of dissolved CO2 in lakes Nyos and Monoun was significantly reduced as a result of the degassing operation since the early 2000s. Detailed measurement of CO2 concentrations at Lake Nyos by Yoshida et al. (2010) showed that the CO2 concentration of the deepest water is decreasing to less than half of the maximum value of 350 mmol kg−1. In the near future, however, the current degassing pipes are expected to stop functioning due to lowered gas self-lift capability as seen at Lake Monoun. This would lead to the renewed buildup of CO2 as long as the natural recharge of CO2 continues, and to a potential risk of recurrence of limnic eruptions. We have proven that the solar-powered deep water removal system installed at Lake Monoun is robust and keeps working without any problem. This means that the risk of limnic eruptions can be avoided if we install a deep water removal system at Lake Nyos after cessation of self-degassing.
Acknowledgments
Fieldwork and installation of a solar-powered deep water removal system at Lake Monoun were supported by JICA and JST under the SATREPS project ‘Magmatic fluid supply into lakes Nyos and Monoun, and mitigation of natural disasters through capacity building’, headed by T. Ohba. We are grateful to Yoichi Yoshida and Aya Inaba, JICA coordinators at SATREPS. Logistic support by the IRGM is acknowledged. S. Djomou of the IRGM is thanked for his help during our work. Critical comments by two anonymous referees helped to improve an earlier version of the manuscript.
- © 2017 The Author(s). Published by The Geological Society of London. All rights reserved