The following appear in order; discussion points may directly
refer to one or more comments preceeding it.
Steve Koch on October
22, 1997 at 17:03:22:
I generally like the nature of this hypothesis, and it looks
like it is refutable with the measurement systems suggested. However,
I would add an additional consideration which follows from my
research with remote sensing systems studying low-level convergence
phenomena over the last few years: the DEPTH of the convergence
is probably an additional important parameter to consider (Koch
et al. 1991; Koch et al. 1997, etc. ... I can provide the references
later). Also - could you clarify what you mean when you say that
"the flow above the convergence line" is an important variable
in this problem?
Andrew Crook on October
24, 1997 at 12:01:38:
Yes, I agree that the depth of lifting will depend on the depth
of convergence. In the analytical model, the lifting depth scales
with the convergence depth, in other words they are linearly proportional.
A conference paper (in postscript format) describing the model
is online at:
In the hypothesis I tried to limit the number of control variables
to three (by the phrase ``among other things''). The hypothesis
could then be tested by finding a convergence line for which two
of the variables were relatively constant while the third varied.
The convergence depth could be added as a fourth control variable
but the other 3 variables would need to be held fixed to test
the hypothesis. Of course, this is assuming that we only verify
by examining trends in the lifting depth. The model also gives
a quantitative estimate of this depth. Hence, by using the observed
lifting depth and not just the trends, the hypothesis could be
tested in cases in which all four parameters varied, (however,
I am less confident in this verification procedure).
``The flow above the convergence line'' is depicted in Fig. 4
of the conference paper. In the model, the flow above the boundary
layer is assumed to be constant with height upstream of the convergence
line. In a real situation, the flow is likely to be sheared. It
probably wouldn't be too difficult to include shear into the analytical
model. Alternatively, a mean flow could be calculated from the
observations by averaging the velocity above the boundary layer
over a depth with scale similar to the boundary layer depth.
Click here to comment on this hypothesis.
Please reference: CROOK.