Depth of Melt Segregation Below the Nyos Maar-Diatreme Volcano (Cameroon, West Africa): Major-Trace Element Evidence and Their Bearing on the Origin of CO2 in Lake Nyos

Abstract

The Nyos maar-diatreme volcano on the Oku Volcanic Group (OVG) in NW Cameroon carries yet the most infamous maar lake in the world because the lake exploded in 1986 releasing CO 2 that killed *1,750 people and over 3,000 livestock. A process of safely getting rid of accumulated gas from the lake started in 2001. Even though *33 % of it has been removed, gas continues to seep into the lake from the mantle, so the lake still poses a thread. Available data on basaltic lava from the maar-diatreme volcano and other volcanoes of the OVG are used here to determine the depth and location where the magmas are produced, and to make inferences on the generation of CO 2 in the Nyos mantle. Fractionation-corrected major element data agree well with experimental data on mantle peridotite and suggest that Lake Nyos magmas formed at pressures of 2-3 GPa in the garnet stability field. This inference is corroborated by trace element models that indicate small degree (1-2 %) partial melting in the presence of residual garnet (2-3 %). The basalts have elevated High Field Strength Element (HFSE) ratios (Zr/Hf = 48.5 ± 1.2 and Ti/Eu = 5,606 ± 224) which cannot be explained by any reasonable fractional crystallization model. A viable mechanism would be melting of a mantle that was previously spiked by percolating carbonatitic melts. It is suggested that small degree partial melting of this metasomatised mantle produces the lavas with super chondritic HFSE ratios, and is generating the CO 2 that seeps into and accumulates in the lake, and which asphyxiated people and animals during the 1986 gas disaster. This finding requires that current efforts to degas Lake

Towards the reconstruction of the shallow plumbing system of the Barombi Mbo Maar (Cameroon) Implications for diatreme growth processes of a polygenetic maar volcano

“…In particular, looking at the Lake Nyos and Monoun disasters (e.g., Kling et al, 1986;Kusakabe et al, 1989;Le Guern et al, 1992), the scientific community is still debating the type of eruption involved. These two disasters prompted research and there is a permanent interest in studying maar volcanism in the area though most investigations are concentrated on Nyos and Monoun maars (e.g., Issa et al, 2013;Aka, 2015;Kusakabe, 2015).…”

Upper Mantle Dynamic and Evolution beneath Lake Nyos Volcano (Cameroon Volcanic Line), Documented by Peridotite Xenoliths Enclosed in Alkali Basalts

“…In addition, the report of the temperatures obtained in this work in the P-T diagram of Mercier and Carter [47] applicable for rifting context, we estimate the pressures from 8 to 18 Kbar corresponding to depths of 50 to 30 km (Figure 11 and see text for explanation), pressures in which studied xenoliths were incorporated in the host lavas, except sample NK14. [48]. The curves correspond to the different possible experimental evolutions depending on the depth.…”

“…Aka [48] indicates that the depth of melt segregation below the Nyos maardiatreme volcano is shallower. The supposed shallow depth of the asthenophere beneath the LNVU indicates that the deepest xenoliths brought up by the basaltic eruption originate from a mantle region not much shallower than the source regions of basaltic magma.…”

The Debris Flow of 1<sup>st</sup> August 2012 in Kakpenyi-Tinta (Akwaya Sub Division) Southwest Cameroon—I: Event Description, Causes and Impacts

“…Lake monitoring studies showed that the killer gas in both lakes was magmatic carbon dioxide. The lakes are currently being degassed artificially [8] [9]. Floods are common in urban and rural areas and are on the rise in major cities.…”

“…Compared to large-scale infrequent disasters like volcanic eruptions, earthquakes, and gas explosions from volcanic crater (maar) lakes ( [9], most small-scale everyday disasters are not well reported in Cameroon, despite the fact that cumulatively, they cause the most casualties and distress to the often very poor people affected [13] [14]. This can flaw the inventory and evaluation of economic impacts of disasters in the country particularly for landslides found in remote areas.…”