Hélène Roux

An intercomparison of remote sensing river discharge estimation algorithms from measurements of river height, width, and slope

The Surface Water and Ocean Topography (SWOT) satellite mission planned for launch in 2020 will map river elevations and inundated area globally for rivers >100 m wide. In advance of this launch, we here evaluated the possibility of estimating discharge in ungauged rivers using synthetic, daily ‘‘remote sensing’’ measurements derived from hydraulic models corrupted with minimal observational errors. Five discharge algorithms were evaluated, as well as the median of the five, for 19 rivers spanning a range of hydraulic and geomorphic conditions. Reliance upon a priori information, and thus applicability to truly ungauged reaches, varied among algorithms: one algorithm employed only global limits on velocity and depth, while the other algorithms relied on globally available prior estimates of discharge. We found at least one algorithm able to estimate instantaneous discharge to within 35% relative root-mean-squared error (RRMSE) on 14/16 nonbraided rivers despite out-of-bank flows, multichannel planforms, and backwater effects. Moreover, we found RRMSE was often dominated by bias; the median standard deviation of relative residuals across the 16 nonbraided rivers was only 12.5%. SWOT discharge algorithm progress is therefore encouraging, yet future efforts should consider incorporating ancillary data or multialgorithm synergy to improve results.

Parameter identification using optimization techniques in open-channel inverse problems

Adverse socio-economic impacts of recent floods both in Europe and other continents emphasize the need for accurate flood forecasting capabilities towards improved flood risk management services. Flood forecasting models are often data-intensive. These models are inherited with (i) conceptual parameters that often cannot be assessed by field measurements, as in conceptual models; and/or (ii) empirical parameters that their direct measurements are either difficult, for example, roughness coefficient or costly, for example, survey data. There is also a category of practical problems, where modelling is required but gauged data are not available. Models, other than purely theoretical ones, for example, Large Eddy Simulation models, need calibration and the problem is even more pronounced in the case of ungauged rivers. Optimal values of these parameters in a mathematical sense can be identified by a number of techniques as discussed and applied in this paper. New generations of satellites are now able to provide observation data that can be useful to implement these techniques. This paper presents the results of synthesized flood data emulating data obtained from remote sensing. A one-dimensional, steady-state flow in a channel of simple geometry is studied. The paper uses optimization methods and the Extended Kalman Filter to ascertain/improve the values of the parameters.

Hydraulic visibility: Using satellite altimetry to parameterize a hydraulic model of an ungauged reach of a braided river

What hydraulic information can be gained from remotely sensed observations of a river’s surface? In this study,we analyze the relationship between river bed undulations and water surfaces for an ungauged reach of the Xingu River, a first-order tributary of the Amazon river. This braided reach is crosscut more than 10 times by a ENVISAT (ENVironmental SATellite) track that extends over 100 km. Rating curves based on a modeled discharge series and altimetric measurements are used, including the zero-flow depth Z0 parameter, which describes river’s bathymetry. River widths are determined from JERS (Japanese Earth Ressources Satellite) images. Hydrodynamic laws predict that irregularities in the geometry of a river bed produce spatial and temporal variations in the water level, as well as in its slope. Observation of these changes is a goal of the Surface Water and Ocean Topography satellite mission, which has a final objective of determining river discharge. First, the concept of hydraulic visibility is introduced, and the seasonality of water surface slope is highlighted along with different flow regimes and reach behaviors. Then, we propose a new single-thread effective hydraulic approach for modeling braided rivers flows, based on the observation scales of current satellite altimetry. The effective hydraulic model is able to reproduce water surface elevations derived by satellite altimetry, and it shows that hydrodynamical signatures are more visible in areas where the river bed morphology varies significantly and for reaches with strong downstream control. The results of this study suggest that longitudinal variations of the slope might be an interesting criteria for the analysis of river segmentation into elementary reaches for the Surface Water Ocean Topography mission that will provide continuous measurements of the water surface elevations, the slopes, and the reach widths.

Improved error estimates of a discharge algorithm for remotely sensed river measurements: Test cases on Sacramento and Garonne Rivers

We present an improvement to a previously presented algorithm that used a Bayesian Markov Chain Monte Carlo method for estimating river discharge from remotely sensed observations of river height, width, and slope. We also present an error budget for discharge calculations from the algorithm. The algorithm may be utilized by the upcoming Surface Water and Ocean Topography (SWOT) mission. We present a detailed evaluation of the method using synthetic SWOT-like observations (i.e., SWOT and AirSWOT, an airborne version of SWOT). The algorithm is evaluated using simulated AirSWOT observations over the Sacramento and Garonne Rivers that have differing hydraulic characteristics. The algorithm is also explored using SWOT observations over the Sacramento River. SWOT and AirSWOT height, width, and slope observations are simulated by corrupting the “true” hydraulic modeling results with instrument error. Algorithm discharge root mean square error (RMSE) was 9% for the Sacramento River and 15% for the Garonne River for the AirSWOT case using expected observation error. The discharge uncertainty calculated from Mannings equation was 16.2% and 17.1%, respectively. For the SWOT scenario, the RMSE and uncertainty of the discharge estimate for the Sacramento River were 15% and 16.2%, respectively. A method based on the Kalman filter to correct errors of discharge estimates was shown to improve algorithm performance. From the error budget, the primary source of uncertainty was the a priori uncertainty of bathymetry and roughness parameters. Sensitivity to measurement errors was found to be a function of river characteristics. For example, Steeper Garonne River is less sensitive to slope errors than the flatter Sacramento River.

Characterization of catchment behaviour and rainfall selection for flash flood hydrological model calibration: catchments of the eastern Pyrenees

Abstract Accurate flash flood prediction depends heavily on rainfall data quality and knowledge of catchment behaviour. A methodology based on global sensitivity analysis and hydrological similarity is proposed to analyse flash storm-flood events with a mechanistic model. The behaviour of medium-sized catchments is identified in terms of rainfall–runoff conservation. On the basis of this shared behaviour, rainfall products with questionable quantitative precipitation estimation (QPE) are excluded. This facilitates selection of rainfall inputs for calibration, whereas it can be difficult to choose between two rainfall products by direct comparison. A substantial database of 43 flood events on 11 catchment areas was studied. Nash-Sutcliffe efficiencies for this dataset are around 0.9 in calibration and 0.7 in validation for flash flood simulation in 250-km2 catchments with selected QPE. The resulting calibration framework and qualification of possible losses for different bedrock types are also interesting bases for flash flood prediction at ungauged locations. Editor D. Koutsoyiannis

Modelling errors calculation adapted to rainfall Runoff model user expectations and discharge data uncertainties

A novel objective function for rainfall-runoff model calibration, named Discharge Envelop Catching (DEC), is proposed. DEC meets the objectives of: i) taking into account uncertainty of discharge observations, ii) enabling the end-user to define an acceptable uncertainty, that best fits his needs, for each part of the simulated hydrograph. A calibration methodology based on DEC is demonstrated on MARINE, an existing hydrological model dedicated to flash floods. Calibration results of state-of-the-art objective functions are benchmarked against the proposed objective function. The demonstration highlights the usefulness of the DEC objective function in identifying the strengths and weaknesses of a model in reproducing hydrological processes. These results emphasize the added value of considering uncertainty of discharge observations during calibration and of refining the measure of model error according to the objectives of the hydrological model. A novel objective function taking into account discharge observations uncertainty, model specifics and user-defined tolerance.A resulting calibration methodology is demonstrated on an existing hydrological model dedicated to flash floods.The results of state-of-the-art objective functions are benchmarked against the proposed objective function.

Evaluation of Regional-Scale River Depth Simulations Using Various Routing Schemes within a Hydrometeorological Modeling Framework for the Preparation of the SWOT Mission

The Surface Water and Ocean Topography (SWOT) mission will provide free water surface elevations, slopes, and river widths for rivers wider than 50 m. Models must be prepared to use this new finescale information by explicitly simulating the link between runoff and the river channel hydraulics. This study assesses one regional hydrometeorological model’s ability to simulate river depths. The Garonne catchment in southwestern France (56 000 km2) has been chosen for the availability of operational gauges in the river network and finescale hydraulic models over two reaches of the river. Several routing schemes, ranging from the simple Muskingum method to time-variable parameter kinematic and diffusive waves schemes, are tested. The results show that the variable flow velocity schemes are advantageous for discharge computations when compared to the original Muskingum routing method. Additionally, comparisons between river depth computations and in situ observations in the downstream Garonne River led to root-mean-square errors of 50–60 cm in the improved Muskingum method and 40–50 cm in the kinematic–diffusive wave method. The results also highlight SWOT’s potential to improve the characterization of hydrological processes for subbasins larger than 10 000 km2, the importance of an accurate digital elevation model, and the need for spatially varying hydraulic parameters.