2004 and 2011 in the North Pacific, P06 in 2003 and 2009 in the South Pacific. It is
expected that in most of these cases there was a long duration of measurements over
the sections (more than 20 days). However, there is an example of faster measurements (9 days) in January 2010 in the Drake Passage. There are also several
examples of “rapid” sections, when deviation Q t was smaller, but comparable to the
error ⟨δQ⟩: the October 2010 section in the Drake Passage (8 days), SR03 in 2008
(18 days) and 2011 (14 days) the south of Australia (Table 2).
Figures 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and 15 show the distributions of
cumulative deviations Q t over the 49 sections studied here (Table 1). The accumulation of the deviation occurs from north to south for quasi-meridional, and from
east to west for quasi-zonal sections. Configurations of level deviations ∑ Δζ t are
almost similar to these distributions. The coefficient of similarity between them can
be estimated from Table 2. In addition, the intervals of the sections with sharp local
deviation Q t , which are not less than the FBTE ⟨δQ⟩, are highlighted in some of
Figs. 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and 15. These sharp deviations can be
associated both with a significant time pause in the measurements and a rapid
change in the current field in short time intervals.
Segments with sharp deviation Q t are found in the western parts of the oceans
over quasi-zonal sections: A05 in 2010 and 2015 close to the continental slope of
North America in the North Atlantic (Fig. 11c, e); P02 in 2004 and 2013 crossing
the Kuroshio zone and the Kuroshio Recirculation in the North Pacific (Fig. 14b, c),
P06 in 2009 crossing the retroflection of the East Australian Current in the South
Pacific (Fig. 15b). In the latter case, significant deviation Q t compared with ⟨δQ⟩
was accumulated just in 3 days. It should be noted that in the P02 section in 2004 a
deviation of 57 Sv, which occurred only in 5 days, is comparable to the total
transport of the Kuroshio estimated at 52–57 Sv [7, 14]. Two segments with a
significant sharp deviation are located in the western part of section A10 in 2011
(Fig. 11d).
In the Southern Ocean, the segments of sections Drake_2010 in October 2010
(Fig. 4d), and P11A in 1993 (Fig. 9a) with sharp deviations Q t corresponded to the
northern periphery of the Antarctic Circumpolar Current (ACC). Of particular note
is the P11A section south of Australia, where about 47 Sv were accumulated just in
5 days, which is more than a quarter of the average total transport over this section
in the Southern Ocean, consisting of the transport of the ACC (173 Sv according to
[3], transport compensating the flow of water from the Pacific into the Indian Ocean
through the Indonesian Straits (14–18 Sv) (e.g., [4, 18]), and transport compensating the flow from the Pacific to the Arctic Ocean through the Bering Strait
(0.8 Sv) [16]. A sharp jump of Q t in the northern part of section P19 (Fig. 14c) is
associated with a long time break in the measurements.
The above examples show that the non-stationarity of the current field in
combination with non-simultaneous measurements at stations of a hydrographic
section can lead to significant distortions in the transport estimates over the sections
based on the ADCP measurements, i.e. “instantaneous” measurements of the
velocity of currents at stations. Often, the accumulation of these distortions along
Influence of the Current Field Non-stationarity …
393
expected that in most of these cases there was a long duration of measurements over
the sections (more than 20 days). However, there is an example of faster measurements (9 days) in January 2010 in the Drake Passage. There are also several
examples of “rapid” sections, when deviation Q t was smaller, but comparable to the
error ⟨δQ⟩: the October 2010 section in the Drake Passage (8 days), SR03 in 2008
(18 days) and 2011 (14 days) the south of Australia (Table 2).
Figures 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and 15 show the distributions of
cumulative deviations Q t over the 49 sections studied here (Table 1). The accumulation of the deviation occurs from north to south for quasi-meridional, and from
east to west for quasi-zonal sections. Configurations of level deviations ∑ Δζ t are
almost similar to these distributions. The coefficient of similarity between them can
be estimated from Table 2. In addition, the intervals of the sections with sharp local
deviation Q t , which are not less than the FBTE ⟨δQ⟩, are highlighted in some of
Figs. 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 and 15. These sharp deviations can be
associated both with a significant time pause in the measurements and a rapid
change in the current field in short time intervals.
Segments with sharp deviation Q t are found in the western parts of the oceans
over quasi-zonal sections: A05 in 2010 and 2015 close to the continental slope of
North America in the North Atlantic (Fig. 11c, e); P02 in 2004 and 2013 crossing
the Kuroshio zone and the Kuroshio Recirculation in the North Pacific (Fig. 14b, c),
P06 in 2009 crossing the retroflection of the East Australian Current in the South
Pacific (Fig. 15b). In the latter case, significant deviation Q t compared with ⟨δQ⟩
was accumulated just in 3 days. It should be noted that in the P02 section in 2004 a
deviation of 57 Sv, which occurred only in 5 days, is comparable to the total
transport of the Kuroshio estimated at 52–57 Sv [7, 14]. Two segments with a
significant sharp deviation are located in the western part of section A10 in 2011
(Fig. 11d).
In the Southern Ocean, the segments of sections Drake_2010 in October 2010
(Fig. 4d), and P11A in 1993 (Fig. 9a) with sharp deviations Q t corresponded to the
northern periphery of the Antarctic Circumpolar Current (ACC). Of particular note
is the P11A section south of Australia, where about 47 Sv were accumulated just in
5 days, which is more than a quarter of the average total transport over this section
in the Southern Ocean, consisting of the transport of the ACC (173 Sv according to
[3], transport compensating the flow of water from the Pacific into the Indian Ocean
through the Indonesian Straits (14–18 Sv) (e.g., [4, 18]), and transport compensating the flow from the Pacific to the Arctic Ocean through the Bering Strait
(0.8 Sv) [16]. A sharp jump of Q t in the northern part of section P19 (Fig. 14c) is
associated with a long time break in the measurements.
The above examples show that the non-stationarity of the current field in
combination with non-simultaneous measurements at stations of a hydrographic
section can lead to significant distortions in the transport estimates over the sections
based on the ADCP measurements, i.e. “instantaneous” measurements of the
velocity of currents at stations. Often, the accumulation of these distortions along
Influence of the Current Field Non-stationarity …
393