316
K. F. BOWDEN ET AL.
Clyde, where the currents were usually less than 10 cm/sec. Rhodamine B
dye solution was us4 as the tracer material in each case, released continuously at the desired depth from a container mounted on a moored buoy. In
a typical experiment the release lasted for a period of 3 to 4 hr, during a
single Rood or ebb tide, with the depth of release set at a value between 1 and
6 m. The resulting dye plume, which was normally visible at the surface, wits
traversed by a research vessel, using one or more sampling intakes at
selected depths and pumping the water through a Turner model 111
fluorometer. In the initial experiments there was only one sampling channel
but in the later series three intakes, at different depths and each connected to
its own Ruorometer, were used. The fluorometer outputs were recorded on
strip charts, the readings of which were later digitized, and in some cases
direct digital recording was utilised. A more detailed account of the experimental technique and method of analysis has been published elsewhere
(Bowden and Lewis. 1973).
3. AN ALYSS
The position of a point in the plume may be &esented by coordinates x,
measured downstream from the source, y laterally across the plume. and 2
vertically downwards from the surhce. The basic data from each traverse, at
a given depth, consisted of a series of concentration values C(y). In the later
experiments three such series, at different valucs of z, were obtained simultaneously. The downstream distance x of the traverse was converted to a
diffusion time t, knowing the velocity of the current.
For a particular traverse, the peak concentration Cp was found by examination and the integrated cross-plume concentration CA , the position j of the
centre of mass of the plume and the variance cri about the centre of mass
were computed. Most of the discussion in this paper is concerned with
relative diffision although some reference will be made later to the problem
of absolute diffusion. The rate of relative lateral diffusion may be represented
by a coefficient of eddy diffusion K,, defined by
(1)
K, = f da;/dt
Estimates of the vertical variance t~,” and the corresponding eddy
coefficient K, were also made, but the values were less reliable as mcasurements at only three depths were available.
4. LATERAL DIFFUSION
The major problem in the interpretation of diffusion experiments, familiar
to all workers in this field, is the great variability of the observed data. In the
present case, part of the variability may be attributed to the expximental
K. F. BOWDEN ET AL.
Clyde, where the currents were usually less than 10 cm/sec. Rhodamine B
dye solution was us4 as the tracer material in each case, released continuously at the desired depth from a container mounted on a moored buoy. In
a typical experiment the release lasted for a period of 3 to 4 hr, during a
single Rood or ebb tide, with the depth of release set at a value between 1 and
6 m. The resulting dye plume, which was normally visible at the surface, wits
traversed by a research vessel, using one or more sampling intakes at
selected depths and pumping the water through a Turner model 111
fluorometer. In the initial experiments there was only one sampling channel
but in the later series three intakes, at different depths and each connected to
its own Ruorometer, were used. The fluorometer outputs were recorded on
strip charts, the readings of which were later digitized, and in some cases
direct digital recording was utilised. A more detailed account of the experimental technique and method of analysis has been published elsewhere
(Bowden and Lewis. 1973).
3. AN ALYSS
The position of a point in the plume may be &esented by coordinates x,
measured downstream from the source, y laterally across the plume. and 2
vertically downwards from the surhce. The basic data from each traverse, at
a given depth, consisted of a series of concentration values C(y). In the later
experiments three such series, at different valucs of z, were obtained simultaneously. The downstream distance x of the traverse was converted to a
diffusion time t, knowing the velocity of the current.
For a particular traverse, the peak concentration Cp was found by examination and the integrated cross-plume concentration CA , the position j of the
centre of mass of the plume and the variance cri about the centre of mass
were computed. Most of the discussion in this paper is concerned with
relative diffision although some reference will be made later to the problem
of absolute diffusion. The rate of relative lateral diffusion may be represented
by a coefficient of eddy diffusion K,, defined by
(1)
K, = f da;/dt
Estimates of the vertical variance t~,” and the corresponding eddy
coefficient K, were also made, but the values were less reliable as mcasurements at only three depths were available.
4. LATERAL DIFFUSION
The major problem in the interpretation of diffusion experiments, familiar
to all workers in this field, is the great variability of the observed data. In the
present case, part of the variability may be attributed to the expximental