Dr. Neil Tindale, Texas A & M Univ.
JGOFS/Arabian Sea
Aerosols, short and long irradiation neutron activation analysis

JGOFS Arabian Sea Aerosol Data

This mineral aerosol concentration data set is from samples collected
during several cruises on the R/V Thomas Thompson during the JGOFS
Arabian Sea field program.

Air Sampling Pump System

The air sampling pumps, controller, electronics, and a clean bench were mounted in a 6-m shipping container located on the foredeck's container storage rack. The air sampler was a HiVol system loaded with a single unwashed Whatman 41 filter and with a flow rate of about 1.2-1.4 m3 min-1. Air flow rates were measured daily using an orifice-type flow tube that had been calibrated to a National Institute of Standards and Technology (NIST) certified EG&G FT32 turbine flowmeter. Air volume errors were between 5 and 10%; the total combined air volume and analytical error was about 10%. The experimental equipment was similar to that described in Schwartze et al. (1988) and used on previous atmospheric-oceanographic programs (Betzer et al., 1988).

References:

Betzer, P.R., Carder, K.L., Duce, R.A., Merrill, J.T., Tindale, N.W., Uematsu, M., Costello, D., Young, R.,
Feely, R.A., Breland, J.A., Bernstein, R., Greco, T., 1988. A pulse of Asian dust to the central North Pacific: long-range transport of giant mineral aerosols. Nature 336, pp. 568-571.

Schwartze, G., Boldi, R., Wasco, T., Duce, R., 1988. PASS: a portable atmospheric sampling system for chemical studies in the marine troposphere. Journal of Atmospheric and Oceanic Technology 5, pp. 561-570.

The data set includes the sampling period
for each sample; the "Day of Year", yrday, number for the start of the
sampling period for each sample; and the concentration of different
elements for each sample, in micrograms per cubic meter. For the
sampling period, "nd" is used as a filler to indicate "no data" gaps in
the data array. The dust values are estimated using aluminum
concentrations determined by neutron activation analysis. While most
samples cover a multi-day period, we only have data for about 150
days. We didn't participate in all of the cruises and, on the cruises
where samples were collected, often sampling conditions were less than
ideal (bad weather, ship maneuvering, relative wind from astern etc.).
A few samples that were collected showed obvious contamination from
local sources, presumably from material from the R/V Thompson or from
nearby fishing boats, and these samples were discarded and are not
included in the data set. While the sampling period represents the
period during which sampling occurred, sampling was usually not
continuous. Sampling was frequently stopped, whenever sampling
conditions were no longer suitable. Thus the concentration value at
any particular date represents a time integrated sample which is
usually non-continuous.

Cautionary comments:
1.  There may be a problem with the estimate for the mineral "dust"
concentration.  Most researchers use the average crustal ratio to estimate
mineral dust concentrations using elemental concentration data (Al, Fe,
etc.).  Surface sand and silt samples that were collected in Oman in the
Wahibah Sands region have distinct, non-crustal ratios.  If individual
aerosol samples are comprised of material from distinct sources, including
Oman, then it is not unreasonable for their elemental ratios to differ from
the published "average" crustal ratio that is used in most aerosol studies.

2.  The amount of Ti in all the aerosol samples was small, despite there
being a reasonable amount of dust material in most samples.  The peaks for
Cr and Ti overlap in the neutron activation short irradiation analysis and
if significant quantities of Cr are present, this will interfere with the Ti
analysis.  With the exception of one sample, the Ti values are at or below
the detection limit.  The sole value above the detection limit was corrected
for the Cr contribution using a correction based on the Cr values from the
long irradiations.  The correction changed the Ti value by less than 5%.
The Ti data flagged as being at or below the detection limit was not
corrected for possible Cr interference.

Data management office notes on supplementary fields - aerosols data

lat, lon

A nominal ship location is given in lat/lon.  The location is the noon
position most near the middle of the sampling period, e.g. for a sample
which was pumped intermittently from Jan. 3 - 5, the location is given for
noon, Jan. 4.  Intended as an aid to understanding, not a discrete location.

date_begin, date_end

We have included the start and stop day for each sample which we
believe constrains the sample time about as well as is useful for these
data.  We also have pump volume and thus a mean concentration of dust
per cubic meter of air for that time frame.  The actual number of hours
sampled during a time block (number of days) is complicated to
present.  Pumps were turned on and off repetitively depending upon ship
maneuvers and relative wind direction (to prevent ship exhaust
contamination).  Also, the total number of hours the pumps were on is a
less useful measure than pump volume, because of the variability in
pump efficiency due to changing barometric pressures. An hours worth of
pumping does not always yield the same volume of air.

aerosols_long.csv