The main objective of the present study is to detect manually individual anomaly trajectories
using filtered sea surface height (SSH) anomaly to the east of Okinawa Island and to understand
the influence of bathymetry on their life cycles. Initially seasonal monthly mean background is
computed using running mean in space (1000 km x 1000 km) and time (five months) using
composite sea level anomaly maps of TOPEX/POSEIDON (T/P) and ERS-1&2 altimeters. These
seasonal mean background is removed from each original Maps of Sea Level Anomaly (MSLA)
cycles to get the filtered SSH anomaly to detect individual mesoscale eddy trajectories based on
certain criteria. Spectral power in frequency domain along 25.875° N latitude using the filtered
SSH anomaly shows maximum peaks in 50, 136-150, and 365 day's periods. Trajectories of
ARGOS drifting buoys east of Okinawa Island show large meandering of a drifting buoy around
Daito Island. In this region high (＞0.8°C) variability is detected using root-mean-square (rms) of
sea surface temperature. The typical scale of the detected mesoscale eddy was 350 km and the
lifetime was 115 days. The total number of the detected warm and cold eddies is 94 in six
years period. More typical eddies are located in deeper region compared with shallow region.
Seasonal distribution of mesoscale eddies changes in space and time. The warm (cold) core ed
dies are coming from southeast (northeast) direction avoiding Daito Sea mount. Passage of tra
jectory of eddies are through deeper ocean. It has been found that along Ryukyu ridge region
eddies decay around shallow continental steep slope. It seems that the trajectory of eddies is
crucially affected by bathymetry. These bathymetric features may have an important role on the
generation and growth of mesoscale anomalies.
Sea-ice differs significantly in thickness, inner texture, thermal conductivity, snow-cover con
ditions, inner temperature gradient and so on according to its place of origin, consequently re
sulting in the great variability in inner heat flux. This work seeks to clarify the effect of the
above parameters on the heat conductive process through observations in the Sea of Okhotsk,
Lake Saroma, the Chukuchi Sea and the Arctic Gyre of the Arctic Ocean north to Canada.
The density and thermal conductivity of thicker sea-ice ike the Arctic sea-ice gradually in
crease from the bottom toward the surface,
and both parameters exhibit a positive correlation.
Sea-ice is more or less accompanied by snow cover everywhere. However, the inner tempera
ture of the upper ice layer is sensitive to the change in air temperature when the snow cover
is less than 20cm thick. However, the effects of changes in the air temperature is ameliorated
by snow cover, and the inner temperature does not exhibit a marked sensitivity when the snow
cover is thicker than 40cm.
The snow cover at all sites lacked uniform depth, but this fact is very significant in estimat
ing heat flux. We therefore observed the occurrence of snow-cover depth and derived a prob
ability density function that was applied to estimate the interface temperature between snowcover
and sea-ice. The heat flux of the sea-ice in the Chukchi Sea was finally estimesd to be
16% higher than that with a uniform depth of 20cm, the mean depth in the Chukchi Sea.
In order to estimate the heat flux correctly, we must know the heat process at the snow-cover
surface. We therefore compared an observation of surface temperature between a mercury ther
mometer and an infrared radiometer and discussed it in terms of the atmospheric stability. We
noticed that the latter exhibited 1.5°C lower temperature than the former, leading to the conclu
sion that the observation with the latter must be corrected by 1.5°C when used to construct an
atmospheric stability diagram.
The net sea-ice formation rate in the Arctic Ocean was estimated from actual observations of
sea-ice thickness by Romanov and sea-ice velocity by the International Arctic Buoy Program
(IABP). The net annual sea-ice melting is largest in the Fram Strait. However, the net annual
sea-ice formation is larger to the north of Beaufort Gyre (Canadian Gyre); the north coast of
Greenland; and north to the New Siberian Islands, East Siberian Sea. Sea-ice is formed over the
entire Arctic Ocean in winter-time, but the formation rate is higher in the east Kara Sea, East
Siberian Sea, Beaufort Gyre, north Chukchi Sea, and north to Barrow. Higher seasonal variabil
ity in the ice-formation rate, especially in marginal seas, is attributed to the effect of runoff
from rivers, i.e, the river runoff stays in the marginal seas for some years till it comes out of
the Arctic Ocean, flowing eastward and repeating cycles of freezing and melting. This analysis
also enables us to detect terrigenous ice that has its origin in the ice sheet, e.g., the Greenland
ice sheet, and to estimate the melting process, unlike in modeling.
This study develops a model that predicts environmental concentrations and ecosystem expo
sure to tributyltin (TBT), which is an organotin compound used as an antifouling paint on ship
hulls. In addition, using existing data, an average calculation for Tokyo Bay over four seasons
was carried out using a three-dimensional flow model and ecosystem model. The calculation re
sults for each season is stored in a database, and then combined with the chemical substance
fate prediction model for marine areas. This model can be adapted for operation in Windows
so that it is easy to use by other operators. Using the simple input of parameters, this prototype
modeling system enabled detailed prediction of environmental concentrations and ecosystem ex
posure in a marine area. In the future, it will be possible to conduct a risk assessment of human
exposure to TBT by improving the accuracy of parameters used in this modeling system.
The Arctic sea-ice field is known to slip along ice cracks (leads). Leads are not located on
certain steady lines, but generated on a certain frequency to disappear after slips. This iteration
is called 'shuffle.' In this paper, I estimated a diffusion coefficient from slip displacement and
frequency. The coefficient is 100 times smaller than that used for stabilizing numerical calcu
lation in basin-scale sea-ice models.