The Cenozoic East African rift system (EARS) is a sequence of grabens, intraplate transforms, accommodation zones, and abundant magmatism. The Central Kenya Peralkaline Province (CKPP) includes a variety of volcanic rocks—trachyte, pantellerite, comendite, phonolite, trachy-basalt, basaltic trachy-andesite, and basalt—which erupted from volcanoes including: Mount Kenya; Eburru and the Greater Olkaria volcanic complexes; Longonot, Menengai, and Suswa volcanoes; and Ndabibi, Elmenteita, and Tandamara mafic volcanic fields. Only two of the volcanoes, Mount Kenya (on the flank of the central graben) and Suswa (the southernmost of the CKPP volcanoes), have phonolites as part of their assemblage.
The evolution of Suswa volcano is divided into three major stages separated by caldera collapse events: (1) pre-caldera, (2) syn-caldera, and (3) post-caldera. Rock compositions are grouped into two sets: pre-/syn-caldera and post-caldera. Both sets range from trachyte to phonolite, but are distinguished by the amount of SiO2. The pre-/syn-caldera rocks have 60% to 62% SiO2, while post-caldera rocks have 57% to 59% SiO2. Each set shows a trachyte to phonolite trend that results from increasing Na2O, accompanied by the increase of a number of trace elements (Be, Cs, Hf, Nb, Rb, Ta, Th, Y, Zn, Zr, and REE, except Eu).
I use whole rock data, mineral chemistry and glass analyses to evaluate magmatic processes, such as magma mixing, Na-F complexing, magma recharge, fractional crystallization and feldspar accumulation. Alkali feldspar compositions support magma mixing process. High fluorine contents (0.01–2.0%) in matrix and melt inclusion glass, and the presence of fluorite in matrix and F-apatite (in groundmass, and as daughter minerals in melt inclusions) provide key evidence to sustain Na-F complexing process. Positive correlations between fluorine and Na2O in matrix glass imply complexing between F and Na. Since fluorine forms stable complexes with several elements, including REE, we attribute the trachyte-phonolite trends in the Suswa rocks to be the result of Na-F complexing.
In addition to F-Na complexing, three other possible processes (not necessarily mutually exclusive) could have produced Suswa post-caldera rocks. They include: (1) a liquid line of descent where the post-caldera rocks are less evolved than the pre-/syn-caldera trachytes; (2) magma mixing where post-caldera magmas are a well-mixed hybrid of Elmenteita/Tandamara type and pre-/syn-caldera magmas; and (3) alkali feldspar fractionation of pre-/syn-caldera magma. The mineralogy in the Suswa rocks is the key evidence to argue against the liquid line of descent process. Instead, the mineralogy supports magma mixing. For example, the pre-caldera unit has Ca-poor anorthoclase (An0-5), while syn-caldera units contain similar anorthoclase, xenocrystic plagioclase with resorbed textures. This suggests that the syn-caldera plagioclase belonged to a mafic magma similar to Tandamara/Elmenteita. In addition, the syn-caldera units contain alkali feldspar with higher anorthite contents (15 to 20%) that resembles post-caldera feldspars. Post-caldera rocks have feldspars with resorbed/sieve textures, and those from the youngest lavas (the Ring Trench Group) are zoned with euhedral overgrowths. Alkali feldspar fractionation is also potentially important because of the abundance of large anorthoclase phenocrysts in post-caldera rocks. If alkali feldspar fractionation were the case, post-caldera lavas (Early Post-caldera Lava Group) are derivative liquids from pre-/syn-caldera magma, via precipitation of stable (equilibrium) alkali feldspar compositions. A key aspect of arguing the importance of mixing is that the observed chemical trends can only be produced by alkali feldspars with higher anorthite contents (An10-15), such as those precipitated in the mixed magma chambers. Therefore, a mixing process existed prior to feldspar fractionation.