Volume 48 (1996) Number 2 
157 
6 
mixed, the lower part of the curve would be a ver 
tical line. However, because potential temperature 
and salinity are not homogeneously mixed the ver 
tical gradients are significantly reduced. Possibly, 
we observe an eroded BML which was established 
during a period of higher velocity. At station 9 a 
small arrow indicates a second change of the po 
tential temperature gradient which possibly marks 
the top of the present BML. 
Light transmission data from the CTD probe 
give no hint of the existence of a bottom nepheloid 
layer. The height of this layer of enhanced turbidity 
frequently exceeds that of the BML significantly 
(Nyffeler and Godet [1986]). Therefore, the height 
of the BML may be regarded as the upper limit of 
the BBL according to the above definition. 
Discussion 
The current data confirm earlier observations of 
Warren [1981] and Lonsdale [1976]: The Peru 
Basin is an area of very weak bottom currents, and 
sometimes the flow is not ‘current-like’. The mean 
currents are comparable to those in the Clarion- 
Clipperton province in the northeast Pacific (Kontar 
and Sokov [1994]) but high-energy events like 
benthic storms have not been observed in the Peru 
Basin. Benthic storms are geographically related to 
sea-surface height variability, and the Peru Basin is 
an area of extremely low sea-surface height variabi 
lity (0-4 cm, Hollister and Nowell [1991]). 
There are two types of erosional, nodule-free 
areas. Some of them are down-slope orientated, 
where the nodules are removed or buried by down- 
slope sediment slides. Other patches are orientated 
parallel to the slope (about 0.1 x 1 km), which might 
point to bottom currents (Wiedecke and Weber 
[1996]). However, the observed bottom currents are 
not capable of causing erosion. Considering the ex 
tremely slow growth of nodules on the order of se 
veral hundred thousands of years, the cause may 
have been highly energetic bottom currents in ear 
lier periods of the basin history. 
The currents at MK1 and MK2 are not corre 
lated. The distance between MK1 and MK2 is 117 
Table 3 
Lagrangian statistics, low-passed data (48 hours) 
mab 
T, 
T 
y 
L y 
k x 
k, 
m 
days 
km 
cm 2 /s x 10 5 
D1: 
200 
9.3 
11.5 
14.7 
17.3 
27.07 
30.20 
50 
6.7 
14.8 
11.1 
27.2 
21.44 
57.98 
30 
5.7 
13.8 
10.4 
26.2 
22.10 
57.25 
15 
5.8 
15.4 
12.4 
40.7 
26.34 
71.80 
MK1: 
503 
6.0 
0.7 
5.6 
0.5 
5.99 
0.43 
202 
4.3 
0.6 
3.6 
0.5 
3-58 
0.40 
50 
0.8 
1.6 
0.6 
1.9 
0.55 
2.61 
MK2: 
503 
0.7 
2.5 
0.5 
3.2 
0.34 
4.86 
202 
2.0 
1.7 
1.7 
1.3 
1.68 
1.12 
50 
3.6 
0.5 
4.0 
0.4 
5.16 
0.40 
mab = metres above bottom, x= zonal, y= meridional 
T,,, = Lagrangian integral time scale 
L,= Lagrangian length scale, k v = eddy diffusivity 
nm, i. e. even with a mean speed of 5 cm/s a signal 
has a transit time of about 50 days from one moo 
ring to the other. However, according to Robinson 
and Kupferman [1985], current measurements in 
the deep central Pacific which are only 10 km apart 
may be distinctly different. This is confirmed by the 
small Lagrangian length and time scales (Taylor 
[1921]) ranging between 0.4 and 5.6 km (respec 
tively 0.5 to 6.0 days) for MK1 and MK2 (see 
Table 3). The values are calculated from low- 
passed data, i. e. tidal and inertial motions are re 
moved. If periods of higher kinetic energy occur in 
the records, as at long-term mooring D1, the values 
increase by about one order of magnitude. This 
holds also true for the eddy diffusivities k r v . Due to 
the large number of rotor stalls at MK1 and MK2, the 
values cannot be calculated for the deeper current 
meters. This strong local variability makes it difficult 
to assess the impact of mining operations, because