Page 145 - Caldera Volcanism Analysis, Modelling and Response

120 A. Pittari et al.

3.4. Quantitative analysis of lithic clast populations
3.4.1. Method
Fifteen field lithic clast analyses within the Abrigo ignimbrite were conducted at 11
locations around Tenerife in order to gain an understanding of the proportions of
the different lithic clast types and their spatial variation. Sites of individual analyses
were chosen so as to obtain the vertical, down-slope and lateral variations of lithic
clast populations within the ignimbrite. For each analysis, the number of each lithic
clast type larger than a threshold grainsize (0.5 cm apparent maximum diameter on
outcrop surface, for 14 analyses) was counted within a box drawn over the
2
ignimbrite surface. The area of the box varied (0.2–0.36 m , for 14 analyses) so that
a minimum of 400 (up to 700) lithic clasts were counted in each case. In the case
of a cobble-sized lithic concentration zone in the upper Sur-A unit (Locality 4),
2
a minimum clast size of 1 cm and a box-size of 1 m was used. It was assumed that
lithic clasts were thoroughly mixed within the pyroclastic flow prior to deposition
so that the square represents the bulk lithic clast distribution of the ignimbrite
deposit at the specified locality and stratigraphic level.
Different lithic clast types are sometimes difficult to distinguish from each other
in the field, especially for small clast sizes or clasts that are masked by surface
weathering. Clast types that are categorised by dispersed phenocryst assemblages
may be mistaken for aphyric varieties where the clast size is small if there are no
distinguishing groundmass features (e.g. small clasts of MV3 or MV2 may be
mistaken for MV1, see Table 2). Gradational boundaries exist between the different
compositional types (i.e. mafic, intermediate and felsic class) and between fresh and
pervasively altered types, and textural and appearance (e.g. colour) descriptors may
overlap. In addition to the large sample sizes and assignment of a minimum
threshold grainsize (see above), the error in clast identification was reduced through
the use of simple, visually identifiable descriptors to classify clast types, which are
highlighted in Tables 1–4. Given these constraints, it is estimated that the error in
assigning a clast to one of the eight major groups is insignificant. The error in
assigning different subgroups is dependant on how distinct a particular subgroup’s
distinguishing features are, but, for each subgroup, an estimated 90% or more clasts
were identified correctly.

3.4.2. Results
Figures 7–12 show bar charts representing the proportions of different lithic clast
groups, and the position of each lithic clast analysis within each stratigraphic log.
The geographic locations for each analysis are shown in Figure 2. The main results
of the quantitative analysis are summarised below, and definitions for abbreviated
lithic clast types are shown in Figure 7 and Tables 1–4. Quoted percentages are
relative abundances within the total lithic clast population.
The range in relative abundances of each of the major lithic groups is: 26–59%
mafic crystalline volcanic clasts (MV1, MV3 and lesser MV4 are most common,
MV2 and MV5 are minor, see Table 2); 14–40% altered lithic clasts (A1 and A3
are more abundant than A2, A4 is rare, see Table 4); 3–35% felsic crystalline