Despite of many published tables of Manning’s roughness coefficients there is not a standard scheme for most of specific land uses.
You can use the available suggested roughness tables and then have a modification according to the categories of your land uses in the area under study. Also an expert judgment or making number of assumptions can be possible options if you have a comprehensive knowledge about the land cover of your study area.
Hope it helps.
Some of them are listed/Attached in the below:
Hunukumbura, P.B., Weerakoon, S.B., Herathm, S., 2007. Development of a cell-based model to derive direct runoff hydrographs for ungauged mountainous basins. Journal of Mountain Science, 4(4), 309–320.
Kilgore, J.L., 1997. Development and evaluation of a GIS-based spatially distributed unit hydrograph model. M.Sc Thesis, Biological Systems Engineering, Virginia Tech. 118p.
Usul, N., Yilmaz, M., 2002. Estimation of instantaneous unit hydrograph with Clark’s technique in GIS. 22nd International of ESRI User Conference, ESRI on-line, San Diego. 21p.
Your question is very interesting, if i understand the problem is related to assign values to roughness coefficients (eg. Manning) in distributed models (eg. 2D models).
I recommend you this site:
http://www.iberaula.com
You can download de model Iber and its tutorials (vídeos & pdf's) in order to release the simulations (2D or 3D), and the respective assigment of roughness coefficients (in raster format) and other aspects that may be of interest for you.
After determining/assigning the appropriate values to each land use category, you can add the Roughness Values to the attribute table of your land use map (vector format) in GIS environment. And then convert that vector to raster if it is required.
I did some MIKE flood modeling a while ago, and I recall that there were roughness maps available to download for ArcGIS based on land use data for the US territory. I am not sure though whether those maps are available for your modeling area.
Look for this NLCD data site: http://www.mrlc.gov/nlcd11_data.php
You may need to reclassify the land use type into the roughness value using textbook information. For example, the land use type is provided with its code number (e.g., 11 for open water), and you need to find the way to convert it to roughness values. I recall there was a simple formula to do this. In my memory, inverse values of the code number was roughness values, but not sure now.