The age and petrogenesis of alkaline magmatism in the Ampasindava Peninsula and Nosy Be archipelago, northern Madagascar

Abstract

The Ampasindava alkaline province consists of a series of circular and elliptical intrusions, lava flows, dyke swarms and plugs of Cenozoic age emplaced into the Mesozoic-Cenozoic sedimentary rocks of the Antsiranana basin (NW Madagascar) and above the crystalline basement. The magmatism in the Ampasindava region is linked to a NW-SE trending extensional tectonic setting. New ⁴⁰Ar/³⁹Ar age determinations on feldspar separate of alkali granites and basaltic dykes yielded ages of 18.01 ± 0.36 Ma and 26 ± 7 Ma, respectively. Alkali basalts and basanites, nepheline syenites and phonolites, and silica saturated-to-oversaturated syenites, trachytes, granites and rhyolites are the main outcropping lithologies. These rocks have sodic affinity. The felsic rocks are dominant, and range from peraluminous to peralkaline. The mantle-normalized incompatible element patterns of the mafic lavas match those of Na-alkaline lavas in within-plate rift settings. The patterns are identical in shape and absolute concentrations to those of the Bobaomby (Cap d’Ambre) and Massif d’Ambre primitive volcanic rocks. These geochemical features are broadly compatible with variable degrees of partial melting of incompatible element-enriched mantle sources. The mineralogical and geochemical variations are consistent with fractional crystallization processes involving removal of olivine, feldspar, clinopyroxene, amphibole, Fe-Ti oxides and apatite. Removal of small amount of titanite explains the concave upward lanthanide pattern in the evolved nepheline syenites and phonolites, which are additionally rich in exotic silicates typical of agpaitic magmas (eudialyte, F-disilicates).

Appendix 1. Analytical methods

Samples for this study were collected from several localities of Ampasindava province (e.g., Sambirano, Ankify, Ambato, Ankaramy, Maromandia, Nosy Be archipelago). The bulk rock chemical analyses were made by X-ray fluorescence (XRF) (Axios Panalytical instrument) at the University of Naples. The analytical uncertainties are in the order of 1–2 % for major elements and 5–10 % for trace elements. LOI (weight loss on ignition) was determined with standard gravimetric techniques, after igniting the powders at 1000 °C. Other XRF analyses were performed at Washington University. Concentrations of major element oxides (Na, Mg, Al, Si, K, Ca, Ti, Mn, Fe, P) were determined on fused glass discs prepared from pre-ignited sample powders using procedures described by Couture (1993).Values of loss-on-ignition (LOI) represent weight loss measured on sample powders ignited for 50 min at 950 ° C in a muffle furnace. Concentrations of 18 trace elements (As, V, Cr, Co, Ni, Zn, Ga, Rb, Sr, Zr, Nb, Ba, Sn, Y, Mo, Br, Ce and Pb) were determined by XRF analysis of pressed powder pellets, using a combination of the modified Compton and fundamental-parameters methods, as discussed by Couture and Dymek (1996). The methods and accuracy for major-element analysis have been reported in Couture (1989) and Couture (1993). Couture (1993) have adopted several refinements of the usual dilution-fusion methods to assure homogeneity and gravimetric accuracy of the prepared samples. For example, the melt is continuously mixed during fusion, to assure homogeneity; the sample and flux are mixed in the platinum crucible used for fusion, thereby eliminating large, selective loss of sample material to extraneous containers; the samples are oxidized before and during fusion, to eliminate exchange of iron with platinum crucibles, which is commonly a very large source of error in XRF analysis. Lanthanides (REE) and other trace elements for selected samples were determined by inductively coupled plasma mass spectrometry (ICP-MS) at ACTLABS, Ancaster, Ontario. Mineral chemical data (more than 600 analyses) were obtained at Centro Interdipartimentale Strumentazioni per Analisi Geomineralogiche (CISAG), University of Naples, utilizing an Oxford Instruments Microanalysis Unit equipped with an INCA X-act detector and a JEOL JSM-5310 microscope. Further analytical details can be found in Melluso et al. (2010, 2014).

For ⁴⁰Ar/³⁹Ar dating, we separated unaltered, optically transparent plagioclase and alkali feldspar using a Frantz magnetic separator and than carefully hand-picked grains under a binocular microscope. The unaltered and optically transparent grains were then leached in dilute HF for one minute and then thoroughly rinsed with distilled water in an ultrasonic cleaner. Mineral separates were irradiated for 25 h in the Hamilton McMaster University (Canada) nuclear reactor. The samples were loaded into two large wells of aluminium disc. The wells were bracketed by small wells that included FCs used as a neutron fluence monitor, for which an age of 28.294 (±0.10 %) Ma was adopted (Renne et al. 2010) and a good in-between grain reproducibility has been demonstrated (Renne et al. 1998). The discs were Cd-shielded (to minimize undesirable nuclear interference reactions) and irradiated. The mean J-values computed from standard grains within the small pits range from 0.007054 (±0.21 %) to 0.006880 (±0.34 %) determined as the average and standard deviation of J-values of the small wells for each irradiation disc. Mass discrimination was monitored using an automated air pipette and provided mean values ranging from 1.001286 (±0.36 %) to 1.001553 (±0.30 %) per dalton (atomic mass unit). The correction factors for interfering isotopes were (³⁹Ar/³⁷Ar)_Ca = 7.30 × 10⁻⁴ (±11 %), (³⁶Ar/³⁷Ar)_Ca = 2.82 × 10⁻⁴ (±1 %) and (⁴⁰Ar/³⁹Ar)_K = 6.76 × 10⁻⁴ (±32 %).

The ⁴⁰Ar-³⁹Ar analyses were performed at the Western Australian Argon Isotope Facility at Curtin University. The sample was loaded in 0-blank Cu-foil packages and was step-heated using a Pond Engineering® double vacuum resistance furnace. The gas was purified in a stainless steel extraction line using a GP50 and two AP10 SAES getters and a liquid nitrogen condensation trap. Argon isotopes were measured in static mode using a MAP 215–50 mass spectrometer (resolution of ~450; sensitivity of 4 × 10⁻¹⁴ mol/V) with a Balzers SEV 217 electron multiplier, using 9 to 10 cycles of peak-hopping. Data acquisition was performed with the Argus program written by M.O. McWilliams; the program ran in a LabView environment. The raw data were processed using the ArArCALC software (Koppers 2002). Plateau and isochron ages were calculated using the new decay constants determined by Renne et al. (2010). Argon isotopic data corrected for blank, mass discrimination and radioactive decay are given in supplementary Table 1. Uncertainties on the plateau and isochron ages include all sources of errors, including error on the decay constant and uncertainty on age of the monitor. Blanks were monitored every three to four steps and typical ⁴⁰Ar blanks ranged from 1 × 10⁻¹⁶ to 2 × 10⁻¹⁶ mol. Our criteria for the determination of a plateau are as follows: plateaus must include at least 70 % of the ³⁹Ar, and a plateau should be distributed over a minimum of three consecutive steps agreeing at the 95 % confidence level and satisfying a probability of fit (P) of at least 0.05. Plateau age is given at the 2σ level and has been calculated using the mean of all the plateau steps, each weighted by the inverse variance of the individual analytical error. Integrated ages (2σ) are calculated using the total gas released for each Ar isotope. Inverse isochrons include the maximum number of steps with a probability of fit ≥ 0.05.