RICERCA ITALIANA     

Research program

CHAOTIC DYNAMICS AND FRACTAL GEOMETRY IN THE GENESIS AND MIXING OF MAGMAS

University Co-ordinator

Università di PISA - SCIENZE DELLA TERRA - PISA(PI)

Research Unit Leader

Pietro ARMIENTI

Description

With this project we aim to evaluate the nature and length scale of the distribution of heterogeneities in the mantle source of oceanic island basalts (OIB). In order to achieve our goal we will be going to document the compositional variability observed in different regions characterized by OIB magmatsm: i) basalts from chains of oceanic islands (e.g. Lanzarote, Canary); ii) recent alkaline basalts from Iblean Basin; iii) basalts from a single volcano (e.g. Etna). For the latter, we will focus on the evaluation of isotopic disequilibria as observed between whole rocks and pyroxenes in two lava flows, e.g. 1669 and 1981, peculiarly variable because of the alternation of magmas with different geochemical characteristics in the former, and for mingling evidence in the latter.

We will differently approach each scale of observation because we suppose the presence of heterogeneities induced by chaotic processes which produced fractal structures. In order to test these hypotheses, we will use geochemical-statistical techniques to evaluate the heterogeneities at the scale of oceans, whereas a petrological-geochemical approach is suitable at the scale of single ocean islands and volcanoes.
We will refer to the SUMA model (Statistical Upper Mantle Assemblege) as proposed by Meibom and Anderson (2004). According to this model, the geochemical peculiarities of OIB would strongly depend on the heterogeneities introduced into the upper mantle by the occurrence of various processes, such as subduction (fluids, melts), degassing, dehydration, metasomatism and different degrees of sampling and homogenization by partial melting. In this frame, the chaotic distribution of the above-mentioned processes, as the result of the tectonic and thermal history of the mantle volume that has been sampled, can account for OIB heterogeneity without the need of any specific geochemical reservoir. The evidence that mantle heterogeneities are the products of chaotic and fractal dynamics imply that the variables of the system are related each other in a strictly not-linear way. Thus the number, the largest as possible, of variables that are taken into account is fundamental not to loose crucial information to understand the evolution of the system in space and time. For this reason, the statistical analysis will be carried out on OIB literature data at the length scale of oceans, thanks to an innovative method of plotting that allows to display the compositional variability of a magmatic series in a multi-dimensional space by using a high number of geochemical parameters at once and a limited number of end-members, on the basis of which the geochemical affinity of each basaltic sample is represented. As far as the study at the length scale of single oceanic island and single volcanic centre, we will perform new geochemical (major and trace element) and isotopic (Pb, Hf) analyses on basaltic rocks from Lanzarote (Canary Islands) and Iblean Basin to highlight the compositional variability of their mantle sources at the scale of both tens and hundreds of kilometers. Moreover, geochemical and isotopic analyses will be carried out on minerals of early generation (e.g. pyroxenes) in selected lavas from Etna (1669 and 1981 flows).
The activity of the Pisa Unit will constrain the nature of OIB mantle source(s), focusing on partial melting processes and mechanisms responsible for small- to moderate- scale mantle heterogeneities in time. The cross use of both chemical and isotopic tracers is fundamental in understanding of magma petrogenesis and crucial to identify any possible component in mantle source characterization.

In order to understand what is the characteristic length scale of mantle heterogeneity, the mineralogical features and pressure conditions during partial melting of the source region should be taken into account together with clues from geochemistry. Recent experimental studies on partial melting processes of pyroxenitic sources recognized the value of 2 GPa as the lowest pressure to obtain a typical thermal boundary between two peridotite-derived melts, in the projection of diopside on the CaTs-Ol-Q plane in the basaltic system. Consequently, we could systematically identify the nature of the basalt sources (pyroxenites/peridotites) and pressure conditions in which partial melting occurred (Kogiso et al., 2004). As far as Etna, we will make the most of the happening of high rate of magma production in a relatively wide range of geochemical variability to discriminate among the possible causes for heterogeneity, such as (a) small-scale compositional variability of the mantle source, (b) progressive fall of partial melting degree, (c) progressive increase of the effect of a fluid component, eventually deriving from dehydration of the subducting Ionic crust.

We will deal with the statistical study of OIB compositional variability at the length scale of oceans by using geochemical and isotopic data available in the literature, relative to basaltic "plume"-related magmas. We will combine radiogenic isotope (Sr, Nd, Pb and Hf) and trace element (highly incompatible element ratios) geochemistry to better constrain the geochemical nature of mantle sources, eventually highlighting any metasomatic enrichment due to melt percolation into the mantle and/or interaction with outside subduction-related material, e.g. sediments from slabs of different age.

As far as Lanzarote island, new Pb and Hf isotope analyses will be performed on samples for which major and trace element and Sr and Nd isotope data already exist. Pb and Hf isotope analyses will be carried out at the laboratory of Isotope Geochemistry of the Ecole Normale Superieure in Lyon (France) by the fellow (assegnista di ricerca) we requested, under Dr. Janne Blichert-Toft supervision (collaborator together on this project). The geochemical heterogeneity of Lanzarote samples will be compared to the geochemical features of the remaining Canary lavas to define the nature and spatial/temporal scale of mantle heterogeneity beneath the central-eastern Atlantic Ocean, in comparison with existent models on the supposed chaotic and fractal nature of mantle dynamics .

Tertiary and Quaternary basalts from Eastern Sicily (Iblean) show striking HIMU (High-m) affinity, the origin of which we will investigate through new Pb and Hf isotope analyses, by checking eventual isotope disequilibria between whole rock and pyroxenes, to document the small-scale geochemical heterogeneity in the Iblean area as well. Parallel we will study the degree of compositional uniformity for the most bulky flows.

The 1669 and 1981 lavas from Etna show a compositional variability which manifests itself through a change in the modal and chemical composition in the former, and evidence of mingling between basalts characterized by diverse chemistry and crystallinity in the latter. We will carry out a detailed chemical and isotopic (Sr, Nd, Pb, Hf) study both on whole-rocks and separated minerals (pyroxenes) from the two lava flows, to quantify the extent of heterogeneity in the source region which sets off to produce interacting magmas (in collaboration with the Perugia unit).

We will constrain the genesis condition for each sample through simulations carried out by using the P-melts code. Besides, using the Wij parameters of binary interaction as proposed by Ghiorso et al. (2002) to calculate the Gibbs energy of a silicate melt at a given P-T condition, we will calculate the SiO2 activity to estimate the likely depth of the mantle source.

In order to understand in detail what are the causes that induce compositional heterogeneity in the mantle, we will perform numerical simulations of mixing in domains with typical rheologic and physical characteristics of the Earth's mantle, by taking into account fields of chaotic flux and fractal geometries (in collaboration with the Perugia unit).

Steps:

First year.
Phase #1-months1-3: Sampling integration of Etna lavas of 1669 and 1981, in collaboration with the Unit of Perugia, and relative isotope and trace element analyses ( ICP-MS and MC-ICP-MS). Study of the isotopic and geochemical variability of the selected lavas and reconstruction of the geochemical features of their mantle source(s).

Phase #2-months 4-8: Sampling integration of Lanzarote and Iblean lavas, and relative isotope and trace element analyses ( ICP-MS and MC-ICP-MS). Development of new methods to data displaying which takes into account a large number of geochemical parameters.

Phase #3-months 9-12: Study on mixing processes among basaltic magmas at Etna and constraints on their compositional "end-members", in collaboration with the Unit of Perugia. Updating of data collection on OIB magmatism, both from available databases and current literature data. Analytical data will include major and trace elements and Sr, Nd and Pb isotope ratios.

Second year.
Phase #4-months 13-16: Identification of the effects of sampling volumes and scale invariance at the length scale of oceanic islands, by statistical analysis of the data collected in the previous phase, and reconstruction of the geochemical variability of the upper mantle at large length scale.

Phase #5-months 17-20: Evaluation of the mineralogical nature of the mantle source, estimate of the depth of origin and source volumes involved, with respect to the degree of partial melting and trace element enrichments in Lanzarote and Iblean lavas. Identification of the magma source and its geochemical variability.

Phase #6-months 21-24: Reconstruction of the general outline and possible causes which concur to generate the geochemical variability of the upper mantle as a source of OIB magmas, both at the length scale of oceans and single volcanic apparatus. Numerical modelling (in cooperation with the Perugia Unit).

Data diffusion
Publication and presentation of results on national and international journals - at least four papers on international journals- and at national and international meetings (FIST , EGS-AGU), also in collaboration with the Perugia Unit.