Volume 48 (1996) Number 2 
) 
159 
Portions of a well-mixed layer can also be advected 
away. The daily means at 202 and 503 mab range 
between 0 and 4 cm/s with a great day-to-day varia 
bility. The BML height varies according to the mean 
values and stabilizes only during periods of stronger 
flow which last 4-5 days at MK1, or several weeks 
at D1. Numerical estimates of BML height h„, for ex 
ample those given by Richards [1990] for modelling 
purposes, provide too small values compared to the 
observations presented above (of the order of /i c ). 
But /¡n is also directly related to i7 E and will therefore 
vary in the same way as lt c (see Fig. 8). Armi and 
Millard [1976] estimated h a to be approximately 6 
times the Ekman layer height, which agrees fairly 
well with the above observations. 
Observations of the temporal variability of 
BMLs are well documented from different Atlantic 
sites, e.g. by Armi and Millard [1976] and Klein 
[1987]. Pulsations of the deep flow like those ob 
served at D1 are also known from the deep NE-At- 
lantic (Klein and Mittelstaedt [1992]), where pe 
riods of stronger near-bottom flows observed in the 
bottom 1500 m have a duration of 1 to 2 month. De 
finite causes of the forcing and vertical extension of 
pulsations in the Peru Basin have not been found, 
however. Driving mechanisms like deep-reaching 
mesoscale eddies have not been observed in the 
data. Possible mechanisms are episodic spill-overs 
similar to those described by Lonsdale [1980] for 
deep water exchange in the Panama Basin by spill 
overs from the Peru Basin. 
Conclusions 
Beside the ‘long-term’ alternation between pe 
riods of weak (1-3 cm/s) and strong (>5 cm/s) near 
bottom flow with typical time scales of 3 to 
5 months, there is a strong day-to-day variability of 
daily mean values during periods of weak currents. 
Therefore, stable BBL’s can only be expected du 
ring periods of strong currents. The variation of the 
daily mean flow often is faster than the time neces 
sary to build up a BML. During times of mean values 
<1 cm/s, the BBL and its sublayers might break 
down completely, i. e. the BML is no longer main- 
tained but eroded. The tidal currents are too weak to 
produce enough turbulent energy for the permanent 
maintenance of a strong BML. The transition bet 
ween hydrodynamically smooth and rough condi 
tion is controlled primarily by the grain size, respec 
tively nodule diameter. 
The local and temporal variability makes it diffi 
cult to properly assess the impact of mining opera 
tions and to model the physical environment, espe 
cially for a larger area. The extent of a disturbance 
will depend strongly on the - unpredictable - hydro- 
dynamical situation on the sea bed. This holds par 
ticularly for biological systems which react very 
sensitively to extreme values. To obtain better esti 
mates which not only give orders of magnitude, a 
quasi-synoptic pattern of CTD stations and a moo 
ring grid of less than 10 km is required. 
Acknowledgements 
The Bundesministerium für Forschung and 
Technologie supported DISCOL under grants 
03R389, 392, 396, 411, 417, and 03F0010F. The 
Bundesministerium für Bildung, Forschung und 
Technologie funded ATESEPP under contract num 
bers 03-G-0106 A-l. 
References 
Armi, L. and R. C. Millard, 1976: The bottom boundary 
layer in the deep ocean. J. Geophys. Res., 81, 
4983-4990. 
Berner, R. A., 1976: The benthic boundary layer from 
a viewpoint of a geochemist. In; The benthic bottom 
boundary layer’, Ed.: I. N. McCave, pp. 33-56, New 
York: Plenum Press. 
Bowden, K. F., 1978: Physical problems of the benthic 
boundary layer. Geophys. Sun/., 3, 255-296. 
Gage, J. D. and P. A. Tyler, 1991: Deep-Sea Biology: A 
natural history of organismns at the deep-sea floor. 
Cambridge; Cambridge Univerity Press, 504 pp. 
Hollister, C. D. and A. R. M. Nowell, 1991: HEBBLE 
epilogue. Marine Geology, 99, 445-460,