subcatchment width

Subcatchment width is hard to define without a sketch. It is NOT the length of overland flow, but is the width of the area where overland flow occurs. It is easier to calculate than to describe!

I always start with the advice in the 1988 manual (page 96 - 104)

1. Find the length of the main drainage channel. The stream order or size of channel/pipe used to define the main drainage channel will vary based on the size of the subcatchment. This step calls for judgement. In general, it is shorter than the longest path through the subcatchment.
2. See if the channel is skewed to one side of the catchment. If it is in the center, the length of the channel is the same as the subcatchment width.
3. Adjust the width based on skew. Sk = (A2 - A1)/A where
Sk = skew factor
A1 = area to one side of channel
A2 = area to the other side of the channel
A = total area
The width is weighted by the skew factor:
W = (2 - Sk) * L where
W = subcatchment width
L = length of main drainage channel

4. Calibrate, if data are available. In general, I don't calibrate with subcatchment width or area, because these are more easily measurable than other parameters, such as depression storage and infiltration.

Bill Frost
KCI Technologies, Inc.

Dear Everyone

I have problem*in the Runoff Module,
What's subcatchment width and how calculate the width
Thank you all very much

[andrewzimba]

Originally Posted by Bill Frost

3. Adjust the width based on skew. Sk = (A2 - A1)/A where
Sk = skew factor
A1 = area to one side of channel
A2 = area to the other side of the channel
A = total area
The width is weighted by the skew factor:
W = (2 - Sk) * L where
W = subcatchment width
L = length of main drainage channel

Umm, doesn't this mean that if my areas are symmetrical (A1=A2), my Sk will be zero? If Sk is 0, then 2-Sk = 2, which makes W = 2L.

This makes no sense. I can construct a theoretical basin with A1=A2 that for certain has W <> 2L!!

[pinkpig]

Catchment width is an important parameter to me in SWWW's linear reservoir runoff model. Width is a much more sensitive parameter to runoff than some other ones, like slope.

I would always calibrate width if it is all possible.

Width is more like a conceptual parameter other than filed measureable one from my understanding, although the name is called "width".

The method I usually use is first to estimate an average overland flow lenth of the subcatchment (or dominant/representative overland flow length) from contour/topo map, and then divide the basin area by this average/dominant/representative length to get a first hand width estimate.

Average overland flow length is estimated along a path from rain dropping to where it enters storm sewer or channel or outlet. Use average value, not longest or shortest.

As I said, I will always calibrate width if I can.

Originally Posted by Bill Frost

Subcatchment width is hard to define without a sketch. It is NOT the length of overland flow, but is the width of the area where overland flow occurs. It is easier to calculate than to describe!

I always start with the advice in the 1988 manual (page 96 - 104)

1. Find the length of the main drainage channel. The stream order or size of channel/pipe used to define the main drainage channel will vary based on the size of the subcatchment. This step calls for judgement. In general, it is shorter than the longest path through the subcatchment.
2. See if the channel is skewed to one side of the catchment. If it is in the center, the length of the channel is the same as the subcatchment width.
3. Adjust the width based on skew. Sk = (A2 - A1)/A where
Sk = skew factor
A1 = area to one side of channel
A2 = area to the other side of the channel
A = total area
The width is weighted by the skew factor:
W = (2 - Sk) * L where
W = subcatchment width
L = length of main drainage channel

4. Calibrate, if data are available. In general, I don't calibrate with subcatchment width or area, because these are more easily measurable than other parameters, such as depression storage and infiltration.

Bill Frost
KCI Technologies, Inc.