| 08 Jan 2026
Evaluating Different Roughness Approaches and Infiltration Parameters for Vegetation-Influenced Overland Flow in Hydrological Model
Abstract. Accurately simulating overland flow in vegetated landscapes remains a challenge in hydrological modeling due to the complex interactions between vegetation, surface roughness, and soil infiltration. This study evaluates multiple methods for estimating Manning's roughness coefficient and explores the influence of vegetation on infiltration processes using the OpenLISEM model. Based on 132 artificial rainfall experiments across 22 sites in southwest Germany, the model was calibrated and validated against observed runoff data, incorporating both depth-independent and depth-dependent roughness formulations. Incorporating water depth-dependent roughness into the model can improve its performance in simulating overland flow. Beyond roughness effects, vegetation was shown to significantly alter soil hydraulic properties, particularly saturated hydraulic conductivity (Ksat). Paired site comparisons revealed that increased vegetation cover corresponded with higher infiltration capacities, emphasizing vegetation's role not only in surface resistance but also in enhancing subsurface water fluxes. The findings demonstrate that models must account for both surface and subsurface impacts of vegetation to improve runoff predictions.
I would like to begin by thanking the authors for their research and for sharing their findings for discussion. Given the increasing frequency of intense rainfall events with significant damage potential, and the considerable efforts by federal states to produce high-resolution heavy rainfall hazard maps, advancing hydrological and hydraulic modeling remains crucial. In particular, there is an urgent need to better quantify and model how varying vegetation conditions affect runoff generation and overland flow dynamics. In this context, the authors’ work is highly valuable, offering insights that are both scientifically relevant and practically applicable.
At the same time, I see room for minor improvement in the presentation of the results, with the aim of making the findings clearer and more accessible to the professional community. Before providing our detailed comments and questions, I would like to share some general critical-constructive observations:
The model setup appears to differ from the study sites described by Ries et al., as the discharge in the model is measured at a different location compared to Ries, where, to my understanding, discharge was recorded 10–20 m downstream of the plot via drainage tubes. This discrepancy could lead to distortions in the subsequent evaluation and should at least be acknowledged as a source of uncertainty. It may also partly explain the differences observed between the model results, the Feldmann data (which presumably have comparable uncertainties), and the experimental measurements. Additionally, providing a rough overview of the ranges of discharge, flow velocities, and water depths across the plot would be highly useful for context and interpretation.
The study emphasizes the influence of vegetation cover, suggesting effects not only on the roughness coefficient but also on key infiltration parameters. However, this impact may be partly inherent to the modeling approach: because the model is calibrated using the original data, where these patterns are already present, the results could reflect a degree of circular reasoning, effectively producing a self-fulfilling confirmation of the initial observations. Also, some of the figures are partly difficult to read and could be improved.
L10 introduce also ksat and psi to complete the picture
L98 Original data resolution of 1min. Does this have a impact on the results? What are the uncertainties?
L105 To explore the impact of roughness on overland flow -> To explore the impact of different roughness functions on overland flow… would be more appropriate?
L113 Microrelief / Microtopography should better fit to roughness coefficients instead of depressions which are mainly connected to retention effects
L115-120 The differences between original setup and model needs to be addressed somehow. At least that this leads to uncertainties of the results. Could be also an explanation for the differences to the results of Feldmann?
L125 model data 1s. original data 1min. How does this influence the results?
L126 … kinematic wave is used together …
L128 Porosity and initial soil moisture was estimated or measurement data? Same for all model runs?
L131 What are the reasonable ranges?
L142 Surface roughness functions -> maybe “Flow resistance parameterisation” is better
L146 complicated sentence, maybe: In this study, two depth-independent and five depth-dependent roughness functions were implemented in OpenLISEM to assess the impact of the investigated approaches on vegetation modelling.
L150 Chow´s method could be changed to method with constant coefficients as it´s more like Manning´s method and chow provided reference tables / guidelines.
L153 … it does not consider the effect of water depth in presence of vegetation. -> … it does not consider the dynamic effect of water depth on the roughness coefficient in presence of vegetation.
L155 add that the same study site and experiments have been used by Feldmann
L157 They estimate surface roughness -> roughness coefficients maybe more suitable?
L169 in equation (1): 5 ∙ hveg
L185 1-2 sentences about this approach would help for understanding
L195 Only Hinsberger et al.; not in Oberle et al. 2021
L208 Please insert the unknown sensitive parameters
L211 It remains unclear on what basis the values from Chow were selected. Average? Max or Min?
L222 (NSE) instead of L224
L232 MSE -> NSE
L248 Calibration of different roughness methods -> Wouldn´t be something like “Calibration of ksat and psi for different roughness methods” be more suitable?
L265 Lag for falling hydrograph between observed data and model maybe due to differences between model and original setup?
L279 add (ksat) (psi)
L295 The font size appears to be inconsistent within the table.
L337 line break incorrect
L346 What is with loc 13, run 6. Differences look significant in contrast to statement in the text.
L351 Isn´t this already evident from the raw data?
L357-361 From my understanding, Run 5 was carried out on the same day as Runs 2, 3, and 4. Why do the initial soil moisture conditions differ, for example, from those of Run 4?
L384 emergent
L384 also this leads to constant manning values for some of the eq?
L405 Again also possible discrepancies between setup of your model, Feldmann model and original experiments could lead to differences.