Watershed Modeling

["Jacob J Jacobson/JAKE/CC01/INEE]

Bill,

I have been working on watershed modeling for sometime now. In the
situation that you describe, I would most definitely use two stocks.
One for the level in the reservoir and one for the storage of water
in the soil. The most compelling reason is that there will always
be a flow from the reservoir to the soil when there is water in the
reservoir but that flow will go to zero when the reservoir is dry.
Any addition of water (i.e. rainfall or as is usually the
case here in Idaho, snowfall) would start in the reservoir and infiltrate
to the soil reservoir.

By using only one stock, you would lose this interesting part of the model
(different infiltration rates, saturation of soil, lining the reservoir...)

Just my one thought for the day.

Jake Jacobson
Idaho National Engineering & Environmental Laboratory
email: jake@inel.gov
(:>)

[Jay Forrest]

Good question Bill!

I would say, no, that it is still not a desirable case for allowing a
negative stock... My logic is based on the fact that the dynamics of
depleting the water in the pond bed is very different than the dynamics of
evaporation, animal drinking, etc. in the above zero state. While the
different dynamics can clearly be handled in a single stock model, I tend
to prefer models where are transparent and the logic clear. Combining
different dynamics in a single stock IMO tends to mask the complexity of
the situation and hide logic.

Thanks for asking!
Jay Forrest
From: Jay Forrest <jay@jayforrest.com>

["Richard Dudley"]

It would seem that pond bottoms would certainly be another stock,
particularly if live water storage (water that could be drained out)
(or fish) were important in the model. Of course the amount of exposed
pond bottom would vary with fullness of the ponds and pond geometry.
Absorption of water will vary with current soil moisture, rainfall,
evapo-tranpiration, plant cover etc.

In a general model, however, I could imagine that a negative water
volume might be a useful concept: water that would need to be added
before outflow could occur. I believe that hydrologists use this idea
in cases where reservoir levels fall below the outflow point.

I am sure the hydrology, agriculture and soil science www sites will
provide more information on this than you would like. Good luck!

_______________
Richard G. Dudley
Visiting Fellow
rgd6@cornell.edu
http://www.people.cornell.edu/pages
gd6/

["Raymond T. Joseph"]

Your last scenario will probably perform the best overall. Low amounts of
rainfall onto a parched or semi-parched bed will be significantly different
from the same rainfall onto an impermeable surface. The evaporation rates
are different also between a pure water surface and mud.


Raymond T. Joseph, PE
Rtjoseph@ev1.net
Aarden Control Engineering and Science