The DWMP blog – 35. Initial loss on permeable surfaces.

This episode of the blog was prompted by the publication of a paper in The Journal of Hydrologic Engineering (https://ascelibrary.org/doi/abs/10.1061/JHYEFF.HEENG-5823).?This presented an innovative approach of using a lidar point cloud survey to measure the actual depth of depression storage on a grassy park.?(I find it interesting that it was described as a dog park.?Obviously American practice is to have so many parks that they can be single use!)

Depression storage is a significant parameter in runoff modelling as it represents the depth of rainfall that is lost on the catchment surface before any runoff forms.?The water held in the depression storage is then lost to infiltration or evaporation so that the storage is available again for the next rainfall event.?The paper suggested that the use of this survey technique was a practical approach that could be used to measure the actual depression storage of areas of a catchment and so calibrate the runoff model for an individual study.?It is a good idea, but I cannot see it coming into widespread use in the foreseeable future.

However, what was as interesting as its potential future use for calilbrating individual subcatchments was what it told us about the general behaviour of depression storage on urban grassland.

The results showed that the average depression storage in the five subcatchments that the park was divided into ranged from 1.7 to 14.3?mm with the depth increasing as the gradient reduced.?This is as expected, steeply sloping ground obviously stores less water on the surface than flat ground.

This relationship is built into the runoff model in InfoWorks ICM where:

• D is the depression storage
• k is a factor depending on the type of surface
• s is the ground slope as %

This was originally used for all surface types with a value of k of 0.71 mm for paved and roof surfaces and a value 4 times larger for permeable surfaces.?(Pitched roofs are given a fixed steep slope that is not relate to the slope of the catchment.)?The latest version of the software allows k to be automatically adjusted for surface type using the porosity of the surface, but I am not aware of anyone doing this.?Please let me know if you are.

For a 1% ground slope this equation gives about 0.7 mm of depression storage for a paved surface and 2.8 mm for permeable surfaces.

At some point the standard practice changed and it is now common to use a fixed depression storage of 2?mm for permeable surfaces irrespective of ground slope.

These newly published results are a reminder that the original calculation of D depending on ground slope was well founded on both theory and practical observation.?Perhaps we should go back to this method.

The software sets a minimum ground slope of 0.2 % but for this with a standard k factor of 2.8 this would give a depression storage of 6.3 mm.?This is still less than that reported in the new study but would be an improvement on 2?mm.