26-. Predicting Varied Kinematic Roughness Parameter during Storm Durations under Ungauged Catchment Situations .

Abstract

Effective overland roughness happens to be the key parameter of Kinematic Wave (KW) models, which truly govern's the overall success of model application in terms of surface runoff simulations. Most of the researchers adopted certain published unique values of Manning's roughness (N) based upon their own subjective judgments. The focus of present study is to address the temporal variability of N during storm duration, that too under less favorable situations where no sufficient hydrologic records are available. Use of unit pulse response through the Soil Conservation Service (SCS) synthetic Unit Hydrograph (UH) is demonstrated for self prediction of varied (N1) during application of KW equations at individual time steps (j) of simulation durations. A natural catchment (1400 ha) has been adopted for application of the approach, considering it as totally ungauged catchment. Synthetic SCS UH was derived utilizing realistic records of rainfall runoff depths, using SCS method. For same unit duration, responses of different rainfall excess depths (0.5-10 mm) were computed in the similar form of Direct Runoff Hydrographs (DRH), considering system to be linear. For each hydrograph ordinate the N1 values were computed with trial and error by running a simple KW model with ensured KW conditions. Time varied optimized values of N1 and overland flow depth (h1) were obtained for each unit pulse hydrograph for two specific time segments viz. rising and falling limbs of hj. Pooled set of values for these time segments were utilized for developing predictive roughness equations. Later these equations were incorporated in KW model to accommodate event based temporally varied nature of N. KW surface runoff responses were simulated for a few storms and found to match reasonably well with the available observed hydrographs on the catchment. Paper comprises a preliminary effort on successful application of KW approach for purely ungauged situations where not a single DRH record is available. Through this study yet another positive utilization SCS UH is projected to generate time varied nature of KW roughness.