Soil temperature and CO₂ degassing, SO₂ fluxes and field observations before and after the February 29, 2016 new vent inside Nyiragongo crater

Abstract

Nyiragongo volcano threatens ∼1.5 million inhabitants of Goma (DR Congo) and Gisenyi (Rwanda) cities and people living in the surrounding villages. In 2002, the volcano produced lava flows which invaded Goma and destroyed the economic district of the city, forced a mass exodus of the population and caused the loss of several lives. Nyiragongo volcanic activity is therefore closely followed by the inhabitants, and any news related to increased activity agitates people in the area, especially those in Goma. Here, we report a short time series of soil temperature and carbon dioxide degassing for four locations, and plume sulphur dioxide fluxes preceding and following the opening of a new vent inside the main Nyiragongo crater on February 29, 2016. The observed sudden and unexpected changes in Nyiragongo activity raised the fear of a new volcanic eruption and led to panic in Goma and the surroundings, inducing some people to leaving the city. We use the dataset and field observations before and after the opening of the new vent, in conjunction with published information about Nyiragongo’s eruptive mechanism and of the volcano’s plumbing system geometry (mainly the crater), to show that the new vent was fed by magma intruded from the lava lake or the upper conduit.

Appendices

Appendix 1: Evolution of the topography of Nyiragongo crater from 2002 to present

Nyiragongo’s pit crater was empty after the drainage of the lava lake through the Shaheru vent on January 22, 2002 (panel a, Fig. 5). The lava lake activity resumed in May 2002 but no major variation of the topography was observed until at least August 2005 (panels b and c, Fig. 5) when the lava lake started a sudden upward which lasted almost 4 months. Since January 2006 (panel d, Fig. 5) to present (panels d to f, Fig. 5; and Fig. 2a–c), the third platform has remained almost at the same level. On the other hand, the lava level has undergone several increase/decrease periods which sometimes may occur within the interval of minutes to hours.

Fig. 5

Photo time series showing the evolution of Nyiragongo main crater topography from the 2002 eruption to present

Appendix 2: Nyiragongo crater about a year after the lava lake activity resumed

During the first year of the lava lake, there were at least two separated active points inside Nyiragongo crater. These points gradually mixed as the lava lake level was increasing and yielded to the present lava lake. On May 6, 2003, another active point was observed just below the second platform level and remained active for ∼48 h.

Fig. 6

Nyiragongo activity characteristics of March–May 2003 a few months after the activity resumed in the pit crater

Appendix 3: Material and methods

Soil temperature was measured at Mugunga, Heal Africa (HA), Bugarura, Shaheru and Nyiragongo summit sites (see Fig. 1). For temperature measurement, we used a Tinytag Plus data logger (HYDREKA SAS, France) provided with a probe which was buried down to 1.5 m into the soil and was thus not or (at least poorly) influenced by weather conditions. The data logger was left in the field during the measurement time as it records data continuously with logging interval of 10 min. The daily mean values shown in Fig. 3b were then calculated from the 144 daily records. The probe functional range is between −50 and +600 °C with a resolution of ±1 °C at 0 °C. CO₂ soil degassing was measured by a GA2000 portable landfill gas analyser (Geotechnical Instruments, UK) which gives the concentration in percent per volume (%v) with typical accuracy of ±0.5 %v. CO₂ is measured with a wavelength infrared detector (WIRD) with reference channel. The GA2000 is provided with a pump and gas inlet connected to a plastic tube, the later fixed to a ∼1.2-m probe which is buried into the soil (at least 60 cm below) during the measurements to avoid atmospheric contamination. The SO₂ flux quantification in the Nyiragongo plume (from the lava lake and later as a mixture of both the lava lake and the new vent, with possible contribution from Nyamulagira plume) was made by continuous (each 15 min) scanning remote sensing UV absorption spectroscopy (280–420 nm) with telemetric data transmission. The wind speed is set at 10 m/s and the plume height at 2000 m above the spectrometer. A full description of the technique can be found in Galle et al. (2003; 2007). Data used in this study were obtained from the Rusayo station (Fig. 1) of which SO₂ flux is influenced by Nyamulagira plume depending on wind direction.