ABSTRACT:

Canopy interception remains one of the most uncertain components of the hydrometeorological water balance, particularly in land surface models that rely on simplified representations of vegetation structure and storm dynamics. This study benchmarks event-scale interception loss estimates from three widely used operational land surface models (Mosaic, Noah, and VIC) against direct observations from 22 forested sites in the National Ecological Observatory Network. Using 1,787 storm events, we evaluate model performance in predicting interception loss as a total depth (IL) and as a percent (I%) of precipitation amount, with and without a precipitation agreement filter. All three models systematically underestimate IL magnitude and variability across sites, with mean bias values ranging from −6% to −18% and low R², even when rainfall inputs closely match observations. Precipitation amount, wind speed, and potential evaporation emerged as the strongest predictors of IL error, suggesting that model limitations—such as restricted canopy storage and simplified energy balance treatments—limit responsiveness to storm intensity. Errors in I% were more strongly influenced by energy-related variables and showed greater variability. Site-level factors like vegetation class and soil texture contributed minimally. These findings suggest that model performance may improve through a) expanding the meteorological variables used to drive wet-canopy evaporation and b) implementing multi-layer canopy storage schemes to better capture within-grid heterogeneity. Both would represent steps toward a more physically realistic framework for simulating interception in forested environments.

ABSTRACT:

Reach-Scale Floodplain Manning’s Roughness Dataset for the Conterminous United States Derived from Remote Sensing and Machine Learning
Gabriel Barinas1,2, Stephen Good1,2, Samuel Rivera1,3
1Water Resources Graduate Program, Oregon State University, Corvallis OR, USA
2Department of Biological and Ecological Engineering, Oregon State University, Corvallis OR, USA
3School of Civil and Construction Engineering, Oregon State University, Corvallis OR, USA
Correspondence to: Gabriel Barinas (barinasg@oregonstate.edu)

Floodplain roughness, quantified through Manning’s coefficient n, is a critical parameter in hydrological models for predicting flood dynamics and managing water resources. Traditional methods to determine n rely on generalized land cover types and often fail to capture the spatial and structural variability of floodplains, resulting in limited understanding of floodplain roughness variation at regional scales. This study integrates high-resolution remotely sensed canopy height and biomass data from NASA’s Global Ecosystem Dynamics Investigation with other spatially distributed data to map Manning’s roughness at reach scales across the conterminous United States. After evaluation of six machine learning models, the best performing approach (Random Forest) was trained on 4,927 roughness estimates from 804 sites and applied to estimate n at 17.8 million reaches within the National Hydrography Database (NHDPlus HR). These n estimates have an R² of 0.51, a root mean squared error of 0.084, and a mean absolute percentage error of 122, capturing spatial variability in floodplain roughness that traditional static methods fail to represent. We find the sparsely vegetated southwest US region exhibits the lowest mean roughness, while the Appalachian region and parts of the southeast US exhibit moderate to high mean values due to denser and more varied floodplain vegetation. Canopy height and biomass were identified as influential non-linear predictors of n, highlighting the importance of vegetation structure on floodplain roughness. This integration of remote sensing data with machine learning models provides spatially distributed estimates of Manning’s n that elucidate patterns in floodplain roughness variability from reach to continental scales. The dataset and companion code are openly available here.

ABSTRACT:

Publication Title: Continental Scale Assessment of Variation in Floodplain Roughness with Vegetation and Flow Characteristics

Quantifying floodplain flows is critical to multiple river management objectives, yet how vegetation within floodplains dissipates flow energy lacks comprehensive characterization. Utilizing over 3.4 million discharge measurements, in conjunction with aboveground biomass and canopy height measurements from NASA’s Global Ecosystem Dynamics Investigation (GEDI), this study characterizes the floodplain roughness coefficient Manning’s n and its determinates across the continental United States. Estimated values of n show that flow resistance in floodplains decreases as flow velocity increases but increases with the fraction of vegetation inundated. A new function (RMSE = 0.024, r2 = 0.74) is proposed for predicting n based of GEDI vegetation characteristics and flow velocity, with GEDI derived n values improving predictions of discharge relative to those based only on land cover. This analysis provides evidence of key hydraulic patterns of energy dissipation in floodplains, and integration of the proposed function into flood and habitat models may reduce uncertainty.