| Summary
The
Oceanographic Component of the Transport Programme is directed towards
determining the oceanic transport pathways of contaminants from three
major geographic sources:
-
Local
sources in the Barents Sea, including the Pechora and White Seas
-
Inflow
of North Atlantic Water into the Barents Sea, and
-
The
Kara Sea, including the Ob and Yenisei Rivers
and
assessing the relative importance of these sources in producing
contaminant loadings in the rich Norwegian fishing grounds of the
Barents Sea.
Phase
1 of the programme is dedicated to development of an appropriate model
tool for transporting contaminants in the northern seas, implementing
the model, evaluating the model results and recommending a monitoring
strategy for the second phase of the programme.
Scientific
results
A
coupled ice-ocean dynamic-thermodynamic model was developed for the
programme to compute ocean currents on a 20-km grid for an area
encompassing the Arctic Ocean, northern North Atlantic and the Barents,
Kara, Greenland-Iceland-Norwegian Seas. The computed currents were
used to transport tracer fields in a series of simulations designed to
examine the sensitivity of computed contaminant distributions to
various physical and numerical factors. It was found that the size of
the model domain, dispersal by tidal currents, the time resolution of
the advecting ocean circulation (daily vs. monthly mean), the
atmospheric forcing used and the spatial resolution of the model had
significant influence on the computed tracer distributions.
It
is difficult to find data suitable for validating ocean transport
calculations, particularly at high latitudes. One such data set is
that provided by the EU VEINS (Variability of Exchanges in the
Northern Seas) programme. IMR has maintained a current meter mooring
section between Bjørnøya (midway between Norway and Spitzbergen) and
Fugløya (northern Norway) from July, 1997 until the present. This
mooring array is designed to measure exchanges between the Norwegian
Sea and the western Barents, in particular, the inflow of North
Atlantic Water into the Barents Sea. It was found that using the
large-area model, which included the Arctic Ocean, with NCEP
atmospheric forcing gave the best agreement with observations for the
annual mean net transport, but that using the DNMI hindcast
meteorological fields for forcing provided the best agreement with the
month-to-month variability in the observed transport.
Based
on the results of the sensitivity analysis in this project, it is
recommended that contaminant transport calculations in Phase 2 of the
programme be based upon three-dimensional hindcast simulations in
which realistic atmospheric forcing is used, tidal effects are
included, and high spatial resolution (20 km or better), high temporal
resolution (daily or better) and a large-area domain (Arctic+GIN
Seas+northern North Atlantic) are used.
The
project met its deliverables and is within budget. Two subcontracts
were let during the course of Phase 1: to PINRO for 50.000 kr and to
DNMI for 300.000 kr.
The
PINRO subcontract was for the validation of water mass distribution
computed by the limited-area model using PINRO’s archived
temperature and salinity fields for the eastern Barents Sea. The
results of that validation indicated the necessity of extending the
model domain to include the whole Arctic Ocean, which was subsequently
carried out.
Through
the DNMI subcontract, the project was supplied with ECMWF and DNMI
Hindcast atmospheric forcing fields, ice dynamics and thermodynamics
modules, wave hindcast fields and an ocean and ice hindcast for the
period 1990-1995 using the 30 km resolution DNMI ice-ocean model. The
wave hindcast fields will be used in sediment transport calculations
in Phase 2 of the programme. The DNMI 6-year ice-ocean hindcast was
used as a check on the NORWECOM results.
Other
work that was coordinated through the Oceanographic Component of the
programme included a rotating tank laboratory study by SINTEF in
Trondheim, in which validation data for the NP-AARI model of
freshwater transport in the Kara Sea was produced and the BarKode
database of temperature and salinity in the Barents and Kara Seas
which was produced by ACSYS through contracting to MMBI in Murmansk.
Relevance
for monitoring
It
is recommended that ice-ocean hindcasts with the large-area, 20 km
resolution model be extended to cover the period 1958-2000 using
hindcast or reanalysis meteorological forcing fields and that the
resulting velocity fields be archived as daily-mean values and tidal
contributions. These velocity fields should then be used in scenario
studies recommended by the Contaminants Component of the programme to
produce realistic distributions of the key contaminants. The study
would point to areas of increased contaminant loading that should be
monitored with observations. Furthermore, the 43-year hindcast will
contain a number of North Atlantic Oscillation (NAO) and Arctic
Oscillation (AO) cycles and would indicate the changes in monitoring
strategy needed, depending upon the stage of the NAO or AO cycle one
is in.
|
