South Africa - National IMAGES report for 1998-1999
REPORT ON RESEARCH ON IMAGES-96 CORES IN SOUTH AFRICA

J. ROGERS

Most of the research on IMAGES-96 (NAUSICAA) cores has been carried
out on cores MD962080 from the tip of the Agulhas Bank and MD962084
from the slope off Namaqualand, NW of Cape Town. In addition, work
is in progress on MD962091 from the slope off southern Angola and
some work has been done on core MD962077 from the Natal Valley off
the SE coast of South Africa.

Core 2080, the most southerly of the NAUSICAA cores has been studied
by Dr John Rogers (JR) of the Department of Geological Sciences
(Marine Geoscience) at the University of Cape Town (UCT) in
collaboration with his PhD student, Mrs Amanda J. Rau (AJR), who is
cosupervised by Dr Julia Lee-Thorp of UCT's Department of Archaeology
(Archaeometry). Working on separate subsamples, taken at
10cm-intervals from the cores by JR and AJR in October, 1997 in
Bordeaux, both 2080 and 2084 are in an advanced state of study.
Assisted by undergraduate students Ansa Lindeque and Samantha
Govender during the '98/'99 austral summer vacation, by a UCT
graduate student, Kerry Lynham, during the austral autumn semester of
1999 and by two undergraduate students, Jovind Hanerjee and Ajay
Trikam in the austral winter vacation of 1999, the entire 2080 core
has been studied for its content of Ice-Rafted Detritus (IRD). Each
of the 220 samples has been separated into its sand- and
mud-fractions and each fraction has then been leached of its
carbonate components.

Focussing on the acid-insoluble sand-fraction, up to 15% of the total
unleached sample, a semi-quantitative determination of the
acid-insoluble component proportions was made using a binocular
microscope. This showed that the dominant component was quartzose
very fine sand with traces of angular, quartzose and lithic very
coarse sand. Ruddiman (1977) mapped the distribution of ice-rafted
sand, including very fine sand, in the North Atlantic, his most
equatorward IRD sand being south of Casablanca, which has a latitude
(35N) similar to that of Cape Town (34S).

The acid-insoluble components included siliceous spicules
from benthic sponges and siliceous tests of planktonic radiolaria. In
addition there were foraminiferal infillings (clay minerals?) and
glauconite pellets. Initial experiments to purify the
acid-insoluble sand of its non-terrigenous components involved
dissolution of the siliceous microfossils in boiling sodium
carbonate, a relatively dangerous method. Instead, the safer method
of floating off the low-S.G. (2.2) sponge spicules and radiolaria
using sodium polytungstate was used successfully by mixing the
solution to an S.G. of 2.5. The "lights" consisted of spicules and
radiolaria, as expected, but also included foraminiferal infillings
and minor amounts of plastic of several colours (not just the white
plastic of the core-liners). The "heavies" consisted of quartz and
glauconite. The contrast between the "lights" and the "heavies" has
been documented with photomicrographs, the traces of quartzose
and lithic very coarse sand contrasting clearly with the dominant
quartzose and glauconitic very fine sand, the grains of which are 10
times finer.

A nest of calibrated 100mm-diameter sieves will now be used to
separate the traces of quartzose and lithic very coarse sand. These
will be counted and their numbers per gram of total unleached
sediment calculated (Elliot, personal communication, 1998). The
"heavies" are being purified of slightly magnetic glauconite pellets
using a Franz Isodynamic Magnetic Separator to provide a percentage
of acid-insoluble terrigenous sand, purified of extraneous biogenic
and authigenic components.

Amanda Rau has analysed both cores MD962080 and MD962084 for
nitrogen isotopes using bulk samples. The resultant isotopic record
is interesting in terms of the range of isotopic values (some very
depleted N-15 values are recorded) and the variability. The isotopes
appear to fluctuate with some sort of cyclicity, particularly in the
lower third of the cores. Spectral analysis will determine the
periodicity of the fluctuations. The majority of work has been
carried out on core 2080. Calcium carbonate analyses reveal that
the coarse fraction (>125 microns) consists predominantly of
carbonate in the form of foraminifera. Carbon isotopes were
determined from the organic fraction, after leaching with HCl.
Carbon isotope ranges are more constrained than the nitrogen
isotopes, but there is evidence of cyclicity reflecting variations in
productivity.

An age model has been determined from the oxygen isotope data that
was provided by the CNRS-CEA at Gif-sur-Yvette on the benthic
foraminifera Cibicides wuellerstorfi and by Min-Te Chen of the
National Taiwan Ocean University in Keelung on the planktonic
foraminifera Globorotalia inflata for the top 14m and top 11m,
respectively.

Counts of 21 planktonic foraminiferal species have been completed for
the top 11m using the >125 micron fraction, in order to establish
foraminiferal assemblages and variability in surface water masses.
The core is dominated by two species, Globorotalia inflata and
Neogloboquadrina pachyderma (dextral) with lesser abundances of the
eastern boundary current species Neogloboquadrina dutertrei. Limited
abundances of the polar species Neogloboquadrina pachyderma
(sinistral) and the subtropical and tropical species Globigerinoides
ruber and Globorotalia menardii occur sporadically upcore.
Statistical analysis will determine the significance of the upcore
variation in species abundance and assemblages.

Little information is available for work on the terrigenous fine
fraction of core 2091 off Angola, which is being supervised by dr P
Zawada in Pretoria, the work being done near Cape Town by the Council
for Geoscience.

Professor Louis Scott has found abundant pollen in core 2084 and has
been sent all the subsamples by John Rogers. he is currently swamped
with INQUA work, the conference being in Durban, South Africa in
August, 1999.

Dr John Rogers
Marine Geoscience, Department of Geological Sciences,
University of Cape Town, Private Bag, RONDEBOSCH, 7701
SOUTH AFRICA. Ph: 27-21-650-3165 Fax: 27-21-650-3783