0% found this document useful (0 votes)

9 views

An Intercomparison of Remote Sensing River Discharge Estimation Algorithms From Measurements of River Height, Width, and Slope

Uploaded by

Abdishakur Mohamed

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

9 views

An Intercomparison of Remote Sensing River Discharge Estimation Algorithms From Measurements of River Height, Width, and Slope

Uploaded by

Abdishakur Mohamed

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 23

PUBLICATIONS

Water Resources Research

RESEARCH ARTICLE An intercomparison of remote sensing river discharge
10.1002/2015WR018434
estimation algorithms from measurements of river height,
Key Points: width, and slope
SWOT discharge algorithms were
tested on synthetic observations for M. Durand1, C. J. Gleason2, P. A. Garambois3, D. Bjerklie4, L. C. Smith5, H. Roux6,7, E. Rodriguez8,
19 rivers P. D. Bates9, T. M. Pavelsky10, J. Monnier11, X. Chen12, G. Di Baldassarre13, J.-M. Fiset14, N. Flipo15,
Algorithms accurately characterized
temporal dynamics of river discharge R. P. d. M. Frasson1, J. Fulton16, N. Goutal17, F. Hossain18, E. Humphries10, J. T. Minear19,
At least one algorithm estimated M. M. Mukolwe20, J. C. Neal9, S. Ricci21, B. F. Sanders22, G. Schumann9,23, J. E. Schubert22, and
discharge to <35% relative RMSE on L. Vilmin15
14/16 of nonbraided rivers
1
School of Earth Sciences and Byrd Polar and Climate Research Center, Ohio State University, Columbus, Ohio, USA,
2
Correspondence to: Department of Civil and Environmental Engineering, University of Massachusetts, Amherst, Massachusetts, USA, 3ICUBE,
M. Durand, Fluid Mechanics Team, Department of Civil Engineering, INSA Strasbourg, Strasbourg, France, 4Connecticut Water Science
[email protected]
Center, USGS, Hartford, Connecticut, USA, 5Department of Geography, University of California, Los Angeles, California,
USA, 6Universite de Toulouse, INPT, UPS, Institut de M ecanique des Fluides de Toulouse, Toulouse, France, 7CNRS, IMFT,
Citation: Toulouse, France, Jet Propulsion Laboratory, NASA, Pasadena, California, USA, 9School of Geographical Sciences,
8
Durand, M., et al. (2016), An
University of Bristol, Bristol, UK, 10Department of Geological Sciences, University of North Carolina, Chapel Hill, North
intercomparison of remote sensing
river discharge estimation algorithms Carolina, USA, 11INSA, Institut de Mathematiques de Toulouse, Toulouse, France, 12NOAA/NWS/Ohio River Forecast
from measurements of river height, Center, Wilmington, Ohio, USA, 13Department of Earth Sciences, Uppsala University, Uppsala, Sweden, 14Environment
width, and slope, Water Resour. Res., Canada, Qu ebec, Canada, 15Centre de G eosciences, MINES ParisTech, PSL Research University, Paris, France, 16USGS
52, 4527–4549, doi:10.1002/ Colorado Water Science Center, Lakewood, Colorado, USA, 17St Venant Lab Hydraul and EDF R&D, University of Paris-Est,
2015WR018434.
France, 18Department of Civil and Environmental Engineering, University of Washington, Seattle, Washington, USA,
19
USGS Geomorphology and Sediment Transport Lab, Golden, Colorado, USA, 20UNESCO-IHE Institute for Water
Received 28 NOV 2015 Education, Delft, Netherlands, 21CECI, CERFACS/CNRS, Toulouse, France, 22Department of Civil and Environmental
Accepted 10 MAY 2016
Engineering, University of California, Irvine, California, USA, 23Remote Sensing Solutions, Inc., Monrovia, California, USA
Accepted article online 13 MAY 2016
Published online 15 JUN 2016

Abstract 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.

1. Introduction
Rivers link atmospheric, terrestrial, and oceanic processes and route approximately two fifth of the global
total rainfall over land back into the ocean [Vo €ro
€smarty et al., 2000; Oki and Kanae, 2006]. In doing so, they
represent an important resource for agriculture and urban development as well a major hazard during flood
events. In these contexts, accurate estimation of river discharge (also ‘‘streamflow’’ or ‘‘runoff,’’ units of vol-
ume per unit time) is vital, as it quantifies the amount of water resources available for human consumption,
C 2016. American Geophysical Union.
V defines the quantity of water that must be routed during a flood event, and indicates overall watershed
All Rights Reserved. response to atmospheric forcing. Despite its importance, our knowledge of global river discharge is

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4527

19447973, 2016, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2015WR018434 by National Institutes Of Health Malaysia, Wiley Online Library on [22/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Water Resources Research 10.1002/2015WR018434

surprisingly poor. This lack of knowledge represents an acute problem, given the possible acceleration of
the water cycle due to global warming [Huntington, 2006]. Improved river discharge estimates, with greater
spatial coverage of the global system of rivers, are needed to develop process-based scientific understand-
ing of runoff at large spatial scales (i.e., how water is routed into and through rivers) and to calibrate and
constrain hydrologic models to forecast effects of future changes in the terrestrial hydrologic cycle. The
forthcoming Surface Water and Ocean Topography (SWOT) mission measurements of river water surface
elevation (WSE), top width, and free-surface slope may allow periodic estimation of river discharge for all
rivers wider than 100 m, with a goal of estimating discharge for rivers wider than 50 m [Biancamaria et al.,
2015; Pavelsky et al., 2014]. Accurate discharge estimates from these measurements would enable tremen-
dous advances in global hydrologic studies.
Given that discharge is the product of flow area and velocity, in situ measurement of river discharge
requires spatially explicit measurements of the vertical velocity profile using a current meter in a transect
orthogonal to river flow [Turnipseed and Sauer, 2010]. While in situ measurements of discharge can be
highly accurate, they are time consuming and impractical for continuous monitoring. Such monitoring is
often achieved via river gauges that leverage periodic, simultaneous measurements of river stage (height
above some arbitrary datum) and discharge to develop a ‘‘rating curve.’’ With this rating curve defined, mea-
surement of river stage is performed (usually) by pressure transducer, allowing for nearly continuous predic-
tion of river discharge via the rating curve.
SWOT-based discharge will never be a replacement for in situ discharge measurements. SWOT overpasses
have a 21 day cycle, and will sample midlatitude locations irregularly in time approximately 3 times per
cycle [Biancamaria et al., 2015], rather than nearly continuously, as in a gauge estimate. While this is
adequate for addressing scientific questions related to the global water cycle, it is inadequate for many
local-scale questions on rivers where SWOT may not fully observe temporal dynamics [Biancamaria et al.,
2010], and for which data are often required at subhourly time scales. Additionally, SWOT discharge esti-
mates are not expected to be as precise as gauged discharge. On the other hand, the spatially continuous
nature of the SWOT measurements will provide data in currently ungauged basins as well as measurements
of spatially distributed phenomena such as the propagation of floodwaves along rivers [Durand et al., 2010;
Pavelsky et al., 2014; Paiva et al., 2015]. SWOT will also complement river discharge modeling. Global water
balance models use climate forcings to determine river runoff, but rely on gauges for parameter tuning
[Hunger and Doell, 2008]. SWOT can provide discharge estimates at continental scale in order to help reduce
current model annual runoff errors that range from 10 to 80%, and are commonly 40% [Oki et al., 1999; Raw-
lins et al., 2003; Widen-Nilsson et al., 2007; Gosling and Arnell, 2011]. SWOT will complement both gauges
and water balance modeling and can form a key component in understanding the global water budget—if
discharge algorithms with sufficient accuracy can be developed.
River discharge estimation from satellite remote sensing of river hydraulic variables including width, stage,
slope, surface velocity, and channel pattern has been explored and discussed in recent decades [Smith et al.,
1996; Smith, 1997; Bjerklie et al., 2003; Kouraev et al., 2004; Bjerklie et al. 2005a; Dingman and Bjerklie, 2006; Bjer-
klie, 2007; Birkinshaw et al., 2010; Michailovsky et al., 2012]. In some of these studies, altimetry measurements
were utilized to estimate discharge with a rating curve available from an in situ discharge gage [e.g., Kouraev
et al., 2004]. Other studies envisioned estimating discharge in ungauged rivers, and utilized traditional flow
laws where only a subset of the hydraulic quantities specified by the flow laws were observed remotely, and
pointed to the need for estimating the unobserved variables, such as roughness coefficient or river bathymetry
[e.g., Bjerklie et al., 2003]. Roux and Dartus [2005, 2008] proposed methods for estimating unobserved hydraulic
parameters from water surface width observations, on a synthetic case and using observed maximum flood
extents. Roux and Dartus [2006] estimated a synthetic flood hydrograph by minimizing the distance between
flood extent observations and 1-D model outputs, assuming the channel geometry and roughness are known
with a given uncertainty. Lai and Monnier [2009] assimilated water levels into a 2-D model and demonstrated
that the inflow hydrograph could be estimated. The anticipated SWOT data have sparked development of new
efforts to develop discharge estimation methods (discussed in detail, below), which draw from these previous
studies and decades of heritage in the fields of hydraulics, remote sensing, and fluvial geomorphology.
Despite the future promise of the SWOT mission, discharge algorithms designed to utilize its data have not
been systematically tested across a range of river types. Bonnema et al. [2016] have made one such test, but
only for three rivers in the same basin. In this paper, we compare six discharge algorithms designed for

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4528

Figure 1. Map showing hydraulic model locations. The rivers shown are those estimated to be greater than 50 m in width, slightly adapted from the analysis of Pavelsky et al. [2014]. (a)
Platte. (b) Tanana. (c) Seine. (d) Severn. (e) Garonne—Upstream. (f) Garrone—Downstream. (g) Cumberland. (h) Kanawha. (i) Sacramento—Upstream. (j) Sacramento—Downstream. (k)
Connecticut. (l) Wabash. (m) Ohio. (n) Mississippi—Upstream. (o) Mississippi—Downstream. (p) Saint Lawrence—Upstream. (q) Saint Lawrence—Downstream. (r) Po. (s) Ganges.

SWOT output (averaged observations of WSE, slope and inundation area over 10 km reaches [Fjortoft
et al., 2014]) to estimate river discharge. Algorithms include the previously published at-many-stations
hydraulic geometry (AMHG) method [Gleason and Smith, 2014; Gleason et al., 2014], GaMo [Garambois and
Monnier, 2015], MetroMan [Durand et al., 2014], and the novel mean flow and geomorphology (MFG) and
the mean flow and constant roughness (MFCR) algorithms, in addition to an ensemble median product. We
use synthetic daily observations of WSE and width generated from hydrodynamic models for 19 rivers cov-
ering a wide range of hydrologic regimes as a stand-in for SWOT data. We first describe each of the algo-
rithms in detail and then compare their estimation results before concluding with a summary of algorithm
strengths and weakness and a prognosis for the future SWOT mission.

2. Experiment Design
We used hydrodynamic model output from 19 rivers to assess algorithm performance in this study (Figure
1). For each river, daily synthetic measurements of water surface slope, elevation, and top width

Table 1. Modeling Platform and Reference for Each Hydraulic Model, Along With the Dates Simulated in Each Model
River Name Model Reference Dates Simulated
Connecticut HEC-RAS 31 May 2011 to 27 Dec 2011
Cumberland HEC-RAS Adams et al. [2010] 1 Jan 2012 to 30 Jun 2012
Ganges HEC-RAS Siddique-E-Akbor et al. [2011] and Maswood 1 Jan 2002 to 31 Dec 2002
and Hossain [2015]
Garonne (DS) Mascaret Besnard and Goutal [2008] and Larnier [2010] 1 Jan 2010 to 31 Dec 2010
Garonne (US) HEC-RAS 1 Jan 2010 to 31 Dec 2010
Kanawha HEC-RAS Adams et al. [2010] 1 Jan 2012 to 30 Jun 2012
Mississippi (DS) HEC-RAS Adams et al. [2010] 1 Jan 2012 to 30 Jun 2012
Mississippi (US) HEC-RAS Adams et al. [2010] 1 Jan 2012 to 30 Jun 2012
Ohio HEC-RAS Adams et al. [2010] 1 Jan 2012 to 30 Jun 2012
Platte BreZo (2-D) Schubert et al. [2015] 13 Sep 2013 to 7 Oct 2013
Po HEC-RAS Di Baldassarre et al. [2009] 2 Feb 2002 to 3 Feb 2003
Sacramento (DS) HEC-RAS Rogers [2014] 1 Feb 2009 to 4 Jul 2009
Sacramento (US) HEC-RAS Rogers [2014] 1 Jan 2009 to 12 Sep 2009
Seine ProSe Vilmin et al. [2015] 1 Aug 2009 to 31 Jul 2009
Severn LISFLOOD-FP Neal et al. [2015] 2 Jun 2007 to 31 Aug 2007
St. Lawrence (DS) H2D2 (2-D) Heniche et al. [2000] 27 Apr 2013 to 16 Oct 2013
St. Lawrence (US) H2D2 (2-D) Heniche et al. [2000] 27 Apr 2013 to 16 Oct 2013
Tanana LISFLOOD-FP (2-D) Humphries et al. [2014] 24 May 2013 to 30 Oct 2013
Wabash HEC-RAS Adams et al. [2010] 1 Jan 2012 to 30 Jun 2012

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4529

Table 2. Summary of the Hydraulic Model Output Used for Each River
Reach Average

Kinematic Wave Max Max a-a-s Flow Simulated Simulation WBM Mean True WBM
Froude Nunber, Number, Mean Width WSE Range Range River Number of Length Annual Flow Average Drainage
River Name Mean (Min-Max) (Min-Max) (m) (m) (cm) Length (km) Stations (days) Estimate (cm) Flow (cm) Area (km2)
Connecticut 0.24 (0.11–0.45) 122.3 (119.9–124.5) 1,382 4.3 3,490 11 8 209 394.1 634.9 26,569
Cumberland 0.06 (0.06–0.07) 7.9 (7.5–8.7) 688 8.4 2,290 43 21 162 511.4 903.4 41,176
Ganges 0.14 (0.12–0.18) 8.9 (6.0–12.5) 11,813 8.7 41,208 103 246 365 12,160 10,950 948,028
Garonne (DS) 0.17 (0.12–0.22) 16.8 (3.6–30.6) 311 5.6 1,601 49 463 365 528.5 475.9 80,049
Garonne (US) 0.21 (0.14–0.25) 58.4 (28.6–91.5) 1,384 4.5 881 75 1158 365 54.7 155.8 13,001
Kanawha 0.06 (0.05–0.08) 3.8 (2.9–4.9) 334 3.5 2,280 18 66 162 305.0 629.3 29,823
Mississippi (DS) 0.15 (0.12–0.17) 14.1 (4.1–18.6) 6,297 8.9 20,772 173 39 162 8928 15,310 2,369,369
Mississippi (US) 0.14 (0.12–0.16) 8.2 (5.0–14.2) 1,868 5.6 6,906 56 74 162 3383 5412 1,841,226
Ohio 0.08 (0.06–0.11) 15.8 (11.4–21.0) 4,488 8.2 13,181 223 161 220 2460 4084 247,984
Platte 0.28 (0.22–0.31) 426.6 (311.6–696.7) 1,917 2.0 357 138 275 24 56.9 150.3 156,800
Po 0.17 (0.09–0.22) 8.5 (2.3–17.4) 5,515 9.5 8,216 98 68 367 841.8 1499 76,497
Sacramento (DS) 0.09 (0.07–0.11) 7.5 (4.5–12.3) 763 6.4 940 36 93 154 377.0 272.6 36,558
Sacramento (US) 0.13 (0.11–0.17) 94.2 (80.6–120.8) 1,957 6.8 1,441 75 194 305 377.0 205.2 36,558
Seine 0.04 (0.03–0.05) 33.9 (15.4–54.3) 510 3.2 787 119 195 365 205.5 304.1 44,201
Severn 0.12 (0.09–0.17) 31.7 (20.0–54.9) 98 10.8 460 65 843 75 104.8 112.8 7,495
St. Lawrence (DS) N/A N/A 15,673 2.3 5,404 22 27 139 9587 9607 931,534
St. Lawrence (US) N/A N/A 4,256 2.1 5,404 38 45 139 7753.7 9607 774,054
Tanana N/A N/A 785 1.5 1,716 94 316 161 344.8 1445 49,692
Wabash 0.08 (0.07–0.09) 68.7 (52.4–106.0) 11,791 6.5 2,724 42 13 162 565.3 979.0 76,133

corresponding to different flows were generated by a hydraulic model forced by in situ bathymetry, simu-
lated or gauged inflows at the top of the reach, and downstream water elevation boundary conditions. The
philosophy of the experiment design was to evaluate discharge estimation under essentially ideal condi-
tions: e.g., daily observations were utilized, although SWOT will measure less frequently—see Biancamaria
et al. [2015] for discussion of SWOT space-time coverage. Moreover, only minimal random measurement
errors were added to the height, width, and slope time series. Future studies will build upon what is shown
here to examine the effect of SWOT measurement errors, and space-time sampling.
In some cases, multiple models on the same river have been used (e.g., an upstream and downstream reach
of the Garonne River are both included); each of the 19 is hereafter referred to as ‘‘rivers’’ as each is an inde-
pendent set of data representing a different range of hydraulic conditions that can be used to evaluate
algorithm performance. Outputs from six different modeling platforms were used in this study, including
HEC-RAS (12 rivers) [Brunner, 2010], LISFLOOD-FP (2 rivers) [Bates et al., 2010], H2D2 (2 rivers) [Heniche et al.,
2000], BreZo (1 river) [Kim et al., 2014], Mascaret (1 river) [Goutal and Maurel, 2002], and ProSe (1 river) [Vil-
min et al., 2015]. Table 1 provides references for each individual model, and Table 2 summarizes simulation
time, reach lengths, and hydraulic characteristics for the 19 rivers in this study.
The models used here were developed for purposes other than those in this study, and so each uses differ-
ent procedures to generate the hydraulic variables of interest and each is run with different spatial and tem-
poral resolutions. A full discussion of the differing model solution schemes and solvers is outside the scope
of this paper, and the interested reader is referred to the citations for each model given above and in Tables
1 and 2 for further information. However, there are a number of core similarities across all of the models. All
of the models represented study reaches with discrete units: either cross sections perpendicular to flow or
2-D grid elements. At each of these units, the models solve for conservation of energy or momentum, using
variants of either the 1-D or 2-D St. Venant or shallow water equations. Note that all models were built using
field measurements of bathymetry. In all cases, flow boundary conditions from a stream gauge were
imposed at the top of the reach and a water surface elevation or ‘‘free-surface slope’’ boundary condition
imposed at the downstream of the reach, allowing the models to apply their particular solver to attain a sta-
ble solution for the hydraulics of each solution unit. Models were calibrated by adjusting the roughness
coefficient in order to match water surface elevation and discharge measurements. These numerical solu-
tions to the conservation equations, coupled with the model’s geometric representation of the river chan-
nel, yield the hydraulic parameters of interest for this study at each unit: water surface elevation, top width,
and water surface slope. In the case of multiple channels, hydraulic quantities were summed across chan-
nels to derive a single value, which matches Schubert et al.’s [2015] ‘‘integrated’’ form for multiple channels.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4530

Reaches on the order of 5–10 km were defined based on inflection points in the water surface elevation
data; methods to automatically identify optimal reach boundaries are needed, and are currently in develop-
ment. Reach-averaged hydraulic quantities were utilized in the Manning’s-based discharge algorithms,
while cross-section data were used directly in the hydraulic geometry-based approach. Time series of simu-
lated river height, width, and slope for all rivers were produced by adding Gaussian errors with standard
deviations of 5 cm, 5 m, and 0.1 cm/km, respectively; time series of height, width, and slope are shown in
Figures (2 and 3), and 4, respectively. The error standard deviations are admittedly somewhat arbitrary, but
were chosen to be small enough to resolve temporal variations visible in Figures 2–4.
Hydraulic regime of each river reach was characterized by two nondimensional numbers: the Froude
number and the kinematic wave number, as defined by Vieira [1983], and used, e.g., by Trigg et al. [2009].
These nondimensional numbers were calculated for each day for each reach, then averaged in time for
each reach. The mean Froude and kinematic wave numbers across all reaches, as well as the minimum
and maximum values across all reaches, are shown in Table 2 for each river. Based on comparison with
the regime diagram shown in Vieira [1983], the hydraulics of nearly all reaches can, perhaps unsurpris-
ingly, be considered to be diffusive. The only river that can be considered kinematic is the Platte River,
which exhibited both the highest average Froude and kinematic wave numbers (0.28 and 426.6, respec-
tively). All reaches exhibit subcritical and gradually varied flow, and no reach showed fully dynamic wave
behavior.
While each data set corresponds to a particular reach of a real-world river, our use of simulated inflows
and often sampled (rather than acquired via side-scan sonar) or simplified channel and floodplain
geometry results in model outputs that may not necessarily represent true hydrologic conditions for
each reach: these data are model-simulated representations of fluvial behavior rather than observations.
This is an important distinction for this study, as using these model outputs allowed us a large degree of
control over algorithm inputs and allowed us the ability to test algorithm performance without consid-
ering the effect of measurement error or noise on the data. Further studies utilizing field data and air-
borne swath altimetry are in process. Despite this caveat, validation and benchmark tests of hydraulic
models built, calibrated, and validated using field and remotely sensed observations do show surpris-
ingly good and consistent performance [e.g., Hunter et al., 2008], with water elevation predictions accu-
rate to <10 cm [e.g., Jung et al., 2012] and inundation extent prediction accuracies up to 90% [Bates
et al., 2006].
The experiment here is purposefully highly restrictive, as every algorithm under consideration was required
to use the exact same input data and no river-specific assumptions or ancillary data (aside from mean
annual flow, as described below) were allowed. All the discharge algorithms (reporting on a given river)
employed identical station data and reach lengths over the same period of record (although note that
AMHG operated on station data, whereas the other algorithms utilized reach-averaged data). For the future
SWOT mission, and in other river discharge studies, ancillary data will form one of the pillars of discharge
estimation: it is foolish to not leverage all available information. However, in this study, we seek to answer
the most basic of questions regarding discharge algorithm performance, and therefore to make a fair com-
parison all methods are restricted to the same input data. This enables an honest assessment of the base
principles of discharge estimation solely from remotely sensed data.
We assessed discharge estimation performance according to a suite of nine error metrics proposed by
Bjerklie et al. [2005b]. Of these, we found that the RRMSE, mean of the relative residuals (MRR), and the
standard deviation of the relative residuals (SDRR) to be of greatest value in discriminating algorithm per-
formance. These metrics allow fair comparison between rivers and allow us to assess how much total dis-
charge error was due to bias; e.g., a large MRR with low SDRR indicates that an algorithm correctly
matched river dynamics but exhibits an offset from true flow. We computed all three of these metrics by
first computing a spatial average of discharge across all reaches, for each time step, and in each algo-
rithm. We compute these three statistics as follows: relative residuals are first computed via: relative resid
ual5ðestimate2truthÞ=truth at each time step, and then take the mean, standard deviation, and RMSE of
the relative residual across the total number of times (which are given in Table 2). When defined in this
way, RRMSE2 5 MRR2 1 SDRR2, which allows discussion of the proportion of the RRMSE due to bias versus
time-varying errors.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4531

4532
Figure 2. Modeled river water surface elevation. The water elevation data used in this study are shown, for all reaches on each of the 19 rivers. The legend indicates reach numbers,

SWOT DISCHARGE ALGORITHM INTERCOMPARISON

Water Resources Research

which increase in the downstream direction.

DURAND ET AL.
19447973, 2016, 6, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2015WR018434 by National Institutes Of Health Malaysia, Wiley Online Library on [22/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10.1002/2015WR018434

4533
SWOT DISCHARGE ALGORITHM INTERCOMPARISON
Figure 3. As Figure 2, but modeled river width.
Water Resources Research

4534
SWOT DISCHARGE ALGORITHM INTERCOMPARISON
Figure 4. As Figure 2, but modeled river slope.
Water Resources Research

Table 3. Summary of Discharge Algorithms

Algorithm Theoretical Basis Applied Variables From Observation Estimated Variables Variable Estimation Method
AMHG At-many-stations Water surface width (W) a, b a and b are optimized to preserve
hydraulic geometry continuity between stations
using a genetic algorithm
GaMo Manning flow Change in cross-sectional Flow resistance (n) and A0 and n are both optimized to
resistance equation area of flow (dA) computed cross-sectional area of preserve continuity between
(equation (4)) from the mean width and channel at zero flow (A0) the observed reaches using a
change in water surface constrained, nonlinear
height (dH, stage), water steepest-descent optimization
surface slope (S) for several reaches
MetroMan Manning flow Change in cross-sectional area of flow Flow resistance (n) and A0 and n are both optimized to
resistance equation (dA) computed from the mean cross-sectional area of preserve continuity between
(equation (4)) width and change in water surface channel at zero flow (A0) the observed reaches (must be
height (dH, stage), water surface slope three or more) using the
(S) for several reaches Metropolis algorithm
MFG Manning flow Water surface width (W), water surface Flow resistance (n) and n is estimated from an empirical
resistance equation slope (S), and water surface height height (stage) of zero flow (B) relation between its mean value
(equation (5)) (H, stage) for the reach and slope, and then adjusted based
on an empirical relation between the
change in river cross section. B is
calibrated to an estimate of the mean
annual discharge for the time series
MFCR Manning flow Water surface width, water surface Cross-sectional area of Flow resistance is assumed constant
resistance equation slope, water surface height channel at zero flow (A0) at 0.03, an estimate of the mean annual
(equation (4)) (stage) for the reach discharge is used to calibrate A0 for the time series

3. Discharge Algorithms
Beginning approximately a decade ago, several approaches were developed using virtual SWOT observa-
tions to test discharge estimation schemes under the aegis of the SWOT virtual mission. Andreadis et al.
[2007] and Biancamaria et al. [2011] assimilated virtual SWOT observations into a hydraulic model, assuming
that river bathymetry and the Manning’s friction n were known. In further developments, Durand et al.
[2008] and Yoon et al. [2012] used assimilation approaches to estimate river bathymetry and discharge
simultaneously; n was assumed to be known a priori [Yoon et al., 2012] or known to vary within a relatively
small range [Durand et al., 2008]. The computational burden of these data assimilation schemes and diffi-
culty in estimating bathymetry, roughness, and discharge within the assimilation scheme led to a search for
simpler methods, which would be more amenable to global application, despite the continued importance
of assimilation in regional and local-scale applications. Durand et al. [2010] first showed that river bathyme-
try and depth could be estimated without running a hydraulic model via analysis of a virtual SWOT observa-
tion time series, if n were assumed known. This represented the first demonstration of river discharge
estimation with minimal a priori data requirements. Mersel et al. [2013] then demonstrated that SWOT
observations could be used to estimate channel bed elevation, provided observations were made at
enough different stages, yielding another means of obtaining prior information. Following these early
developments, there are now five proposed algorithms for use in the SWOT mission; the median of these
five is evaluated as a sixth algorithm herein. The basics of the algorithms are summarized in Table 3; each
algorithm is described below.

3.1. At-Many-Stations Hydraulic Geometry (AMHG)

At-a-station hydraulic geometry (AHG), equations (1)–(3), where a, b, c, f, k, and m are empirical best fit
parameters, were first described by Leopold and Maddock [1953], and are an often-used framework in river
remote sensing [e.g., Smith et al., 1996; Smith and Pavelsky, 2008; Pavelsky, 2014]. AMHG is a recently discov-
ered geomorphic phenomenon holding that the coefficients and exponents in traditional AHG are stably
and predictably related for a given river, thus linking individual cross sections to one another along a river
[Gleason and Smith, 2014]. Gleason and Smith found that the relationship between these quantities takes a
semilog form, and used in situ measurements of width (w), depth (d), velocity (v), and discharge (Q) to dem-
onstrate the phenomenon. Gleason and Wang [2015] further showed that AMHG arises because individual,
independent rating curves all pass through the same values of w and Q, given in practice by the spatial
modes of time mean w and Q at each cross section.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4535

w5aQb (1)

d5cQf (2)

v5kQm (3)

The AMHG discharge algorithm’s base assumptions are that AHG parameters are constant in time and that
mass is conserved in a reach. Discharge is calculated by estimating a and b in equation (1) and inverting to
solve for Q in a pairwise permutation assuming that Q is constant across all pairs. Thus, this algorithm only
requires inputs of remotely sensed width at-a-station, which differentiates it from the other methods in this
study. From a remotely sensed standpoint, this system is underconstrained even in a mass conserved reach:
there are four variables per cross section (i.e., w, Q, a, and b), and only one of them can be remotely sensed
(w). Thus, when assuming cross sections share a common discharge given a time series of width observa-
tions, there are always 2Nc 1 1 unknowns for Nc cross sections. Since AMHG relates a and b for a given river,
it simplifies the AHG system, and results in two unknowns per cross section: Q and the a/b tuple. While this
system is still underconstrained, it lends itself to heuristic optimization as it leaves only Nc 1 1 unknowns
per Nc cross sections. Gleason and Smith [2014] found a remotely sensed proxy that approximates the
AMHG for a river given a time series of remotely sensed river widths, thus allowing AHG simplification with-
out any a priori knowledge, and solved for discharge by minimizing flow residuals via genetic algorithm
optimization of AHG parameters. Gleason and Wang 2015] have since shown that this proxy is spurious and
give proposals for its replacement, but Gleason and Smith [2014] still reported discharge estimations to
within 20–36% relative Root-Mean-Squared Error (RRMSE) for three rivers in their initial demonstration, and
Gleason et al. [2014] performed a thorough sensitivity analysis of the AMHG method for 34 rivers, yielding
discharge accuracies between 26 and 41% for most river morphologies using the originally proposed proxy.
In this paper, we use the original proxy of Gleason and Smith [2014] to estimate discharge.
Importantly, Gleason et al. [2014] recommended ‘‘global’’ parameterizations of the AMHG method for use in
ungauged basins, and we follow these parameters here. Gleason et al. [2014] also showed that rivers in arid
regions, braided rivers, and low-b rivers (where all AHG b exponents are less than 0.1) reliably resulted in
poor discharge inversion. We expect that these same exclusions will apply here, so we also include some
‘‘blind’’ data filters intended to improve AMHG performance. Since equation (1) breaks down during over-
bank/floodplain flow, we first filter any widths that are 1–3 standard deviations above the mean, depending
on the shape of the distribution of input width data. Second, in order to increase the amount of rivers avail-
able to AMHG, we filter out cross sections that have a coefficient of variation (standard deviation divided by
mean) in observed widths less than 10%; this is a much less stringent filter than is typically used, although it
still results in exclusion of 8 of the 19 rivers by AMHG. Thus, we are able to estimate discharge for some riv-
ers whose width data would suggest that they are unsuitable for AMHG estimation (as they are low b).

3.2. The GaMo Algorithm

Both the GaMo and MetroMan algorithm described later utilize the following form of Manning’s equation:
1
Q5 ðA0 1dAÞ5=3 W 22=3 S1=2 (4)
n
where n is the Manning’s roughness coefficient, A0 is reach-averaged cross-sectional area at the time of the
lowest observed river elevation, dA is the change in cross-sectional area with respect to A0 , W is the
observed channel top width, and S is the slope of the water surface elevation. All quantities represent reach
averages, where reaches range from 5 to 10 km in length. The n and A0 are unknown, while the other three
terms are observed; n is assumed to be a constant in both space and time, while A0 varies in space (but not
in time). All of the observable quantities, as well as river discharge vary in both space and time, though sub-
scripts are suppressed here for brevity. Garambois and Monnier [2015] showed that including inertia terms
in the inverse hydraulic model would require high spatial resolution (i.e., subreach) knowledge of the river
bathymetry profile.
The GaMo algorithm described by Garambois and Monnier [2015] invokes continuity among reaches,
assumes that flow is constant in space (@Q=@x50), and shows that this is an underdetermined nonlinear
system with NR 3NP equations and NR ðNP 11Þ11 unknowns, where NR and NP are the number of reaches
and observation times, respectively. Garambois and Monnier [2015] utilized the Levenberg-Marquardt solver;

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4536

hence, the computed optimal solution satisfies the NR 3NP Manning equations in the least squares sense.
First guess, minimum, and maximum bounds for discharge were derived from the MFCR discharge esti-
mates (described below). Specifically, the mean, minimum, and maximum MFCR discharge values across
both time and space were utilized. First guess, minimum, and maximum bounds for n were chosen to be
0.03, 0.025, and 0.05, respectively. Minimum and maximum bounds for A were chosen as the smaller of the
minimum dA across the time series at each reach, and 100 m2, and the maximum dA across the time series,
respectively. The first guess A0 was calculated as half the range of dA across the time series, at each reach.
Garambois and Monnier [2015] highlighted the equifinality problem between n and A0 : multiple combina-
tions of n and A0 satisfy Manning’s equation. Garambois and Monnier [2015] also found that if a minimum of
three distinct flow regimes are observed, then the algorithm can be expected to converge to effective
parameters (n, A0).

3.3. Metropolis-Manning (MetroMan)

MetroMan begins with the same governing equation as GaMo, equation (4), and is similar to GaMo in that it
analyzes a time series of height, width, and slope in order to optimally estimate n and A0 , which are
assumed to be temporally invariant [Durand et al., 2014]. A difference between the two comes in the formu-
lation of the set of equations to be minimized, where equation (4) is substituted into the reach-averaged
continuity equation @Q=@x1@A=@t5q, where q represents lateral inflows into the reach; for all cases consid-
ered herein, q is zero. Note that @A=@t is observable, as it can be calculated directly from measurements of
river WSE and W. This formulation leads to a set of equations with more constraints than unknowns when
Manning’s equation is substituted into the mass balance equation.
To solve for the optimal n and A0 , a Metropolis algorithm [Metropolis et al., 1953; Gelman et al., 2004] is
invoked, to obtain both an optimal parameter estimate and an estimate of uncertainty. A Markov Chain
Monte Carlo (MCMC) approach is utilized. MetroMan was demonstrated on the Severn River using water
elevation measured at three gauges and cross-sectional area measured in situ [Durand et al., 2014]. Several
experiments with different assumptions regarding q led to relative RMSE ranging from 10% to 36%. A
follow-on study compared and contrasted algorithm performance on the Sacramento and Garonne Rivers
with synthetic data and demonstrated that discharge can be estimated to within 9% and 15% RMSE with
MetroMan, and also demonstrated different sensitivity to measurement errors [Yoon et al., 2016]. The Metro-
Man approach has been tested on a small range of rivers with good success in previously published litera-
ture, so success was expected here. Note that no roughness coefficient variability with stage is included
within the algorithm, so when those conditions exist (e.g., during out-of-bank flow events), the algorithm is
not expected to perform well.
We changed two relatively minor components of the algorithm for this study. First, MetroMan was adapted
to utilize a prior estimate of mean annual flow and n; previous studies utilized prior estimates of A0 and n,
directly. A prior n value of 0.03 was assumed for each reach. The prior mean and standard deviation of A0
were calculated for each reach using the same approach as the MFCR (described in section 3.6). This was
done using a Monte Carlo method, assuming that uncertainty in the mean annual flow estimate followed a
lognormal distribution with coefficient of variation equal to one. Possible mean flow values were simulated
from this distribution; for each possible mean flow value, a corresponding A0 value was calculated by solv-
ing equation (4), yielding a mean and standard deviation for A0. Values of A0 are furthermore limited to be
greater than the minimum dA values, such that cross-sectional flow area is positive. The second change
made to the MetroMan framework was to perform Markov Chain iterations on an estimate of base flow,
rather than on A0; this has the nontrivial advantage that base flow is the same for all reaches within a given
river, and therefore many fewer iterations are required to explore a wide range of flow values. The prior for
base flow is calculated from the prior on mean flow, using similar methods to those described for A0.

3.4. Mean-Annual Flow and Geomorphology (MFG)

The MFG algorithm uses the so-called wide-channel approximation [Tinkler, 1982], leading to a form of
Manning’s equation that approximates river depth as the difference between WSE (H) and the cross-
sectional average river bathymetry (B):

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4537

1
Table 4. Parameter Identification Windows (With Units of Days and Referenced From Initial Q5 ðH2BÞ5=3 WS1=2 (5)
n
Simulation Date; See Table 1) Utilized for GaMo and MetroMan, and the Number of Reaches
Used for the Four Manning’s-Based Algorithms
MFG assumes that an
River GaMo ID Window MetroMan ID Window Number of Reaches
acceptably accurate esti-
Connecticut 85–95 85–115 3 mate of mean annual flow
Cumberland 87–92 55–85 4
Ganges 200–220 181–201 6 will be available for SWOT
Garonne (DS) 10–19 180–210 9 rivers. Mean annual flow
Garonne (US) 10–19 2–32 16 values are calculated by
Kanawha 8–31 116–146 4
Mississippi (DS) 11–46 12–42 6 averaging daily streamflow
Mississippi (US) 105–119 12–42 5 predictions of the water
Ohio 10–40 115–145 5 balance model (WBM) [e.g.,
Platte 1–20 3–23 14
Po 33–55 90–108 16 Wisser et al., 2010] at a spa-
Sacramento (DS) 14–21 10–40 9 tial resolution of 6 min,
Sacramento (US) 42–52 10–40 7 from 1961 to 2010. Grid
Seine 64–75 143–165 4
Severn 14–21 43–73 8 resolution is on the order of
St. Lawrence (DS) 200–220 2–32 9 100 km2, and thus this
St. Lawrence (US) 16–45 2–32 7 model is adequate to pro-
Tanana 20–30 4–34 9
Wabash 16–45 17–47 4 vide a mean annual flow
estimate for rivers SWOT
will observe, which typically
have a drainage area on the order of 10,000 km2 or greater [Pavelsky et al., 2014]. An estimate of the rough-
ness coefficient is computed from a relation that scales a mean value of the roughness from observations of
width and stage. The relation, given by equation (6), was derived from analysis across a number of USGS
field data and gauges:

HW x
n5c0 n0 (6)
HW

where c0 and x are empirical coefficients, n0 is a static, reference value of n, and the overbar indicates time
averages. The value of x is typically less than zero, such that n increases at low flow, in agreement with
expectations. The MFG algorithm is not applicable during overbank flows, and does not allow, e.g., n to
increase at high flow to account for out-of-bank conditions. Values of x are calculated from time series
observations of w and H; B is calculated in order to match the time series of discharge estimates with and
expected value of the mean annual flow, derived in this study from WBM.
MFG relies on a series of observations of width and stage that covers a range of flow conditions such that
mean values of the flow dynamics begin to be approximated. n0 is estimated from the channel slope as

n0 50:22S0:18 (7)

Equation (7) was developed from data base of streamﬂow measurement data published in Bjerklie et al.
[2003]. With these relations, the variable Manning n and cross-section shape are estimated and the value of
B is calibrated by ﬁtting the mean of the observed time series estimates of discharge to the mean discharge
obtained from a water balance model or some other independent source.

3.5. Identification Window for MetroMan and Gamo

Both the MetroMan and GaMo algorithms perform optimization not on the entire time series, but rather on
a subset. This is done in general to reduce the computation costs of matrix inversions [see Garambois and
Monnier, 2015; Durand et al., 2014]. In this paper, the identification periods for both algorithms for each river
are chosen in order to include significant changes in stage, and to avoid extreme out of bank events. Identi-
fication periods are given for each algorithm in Table 4. For a given river, A0 and n inverted from the identifi-
cation period are then used to calculate Q over the whole time series.

3.6. The MFCR and Ensemble Median Algorithms

The ‘‘mean ﬂow with constant roughness’’ (MFCR) approach simply assumes that n is 0.03 and uses the
WBM mean annual ﬂow estimate described in the previous section. Then, the MFCR algorithm estimates A0
in equation (4) such that Manning’s equation applied to the average height, width, and slope gives

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4538

discharge mean to equal the WBM prior mean, subject to the constraint that the sum of A0 and any of the
dA measurements (i.e., the river cross-sectional area) is greater than zero. This algorithm represents the
simplest possible approach to estimating discharge, and is limited by the accuracy of the WBM estimate.
This approach is conceptually similar to that of Smith et al. [2015]. The second additional approach is an
ensemble product, calculated by taking the median of the ﬂow produced by each of the ﬁve algorithms at
each time step. We hypothesize that this ensemble product will be more stable than any single algorithm,
and discharge estimated this way should be the most consistent from river to river.

4. Results
Time series of river discharge estimated on each river are shown in Figure 5. Table 5 summarizes the per-
formance of each of the six algorithms for every river according to each of the three metrics described in
section 2. RRMSE is perhaps the most important assessment metric, as this error determines how well each
algorithm performed across the entire time series of observations. Thus, RRMSE can be considered the cen-
tral ‘‘accuracy’’ metric in this study, and is directly comparable across scales for the 19 rivers here. We use an
RRMSE of 35% as a threshold for algorithm performance; this number is admittedly arbitrary, but was cho-
sen to improve upon the accuracy values cited for global models (40%) in the introduction. RRMSE values
ranged from 5% to greater than 100% across all rivers and all algorithms. Viewed collectively, at least one
algorithm had an RRMSE less than 35% for 14/19 rivers, and the grand median RRMSE (across all six algo-
rithms and 19 rivers) was 55%; note that some of the algorithms yielded highly biased results, such that the
mean across all rivers and algorithms is 87%. Five of the six algorithms had the best estimation RRMSE for
at least one river, with MetroMan having the best performance for eight rivers, MFCR performing best for
four rivers, GaMO and MFG performing best for three rivers each, and AMHG giving the best estimate on
one river. Interestingly, the ensemble median algorithm did not give the best result on any river, despite
selecting the median flow value of the other five algorithms at every point in time. At least one algorithm
had an RRMSE greater than 100% in 9/19 rivers. Figure 6 summarizes and compares the performance of
each algorithm.
Braided rivers were particularly difficult to estimate, as none of the Ganges, Platte, or Tanana Rivers had an
algorithm register an RRMSE less than 35%. In addition, the Ganges and Platte Rivers each had three algo-
rithms record an RRMSE greater than 100%. This mirrors the findings of Gleason et al. [2014], and further
confirms the difficulty of estimating discharge via remote sensing in braided rivers. Removing these braided
rivers from consideration results in 14/16 rivers having at least one discharge RRMSE less than 35%.
The Cumberland and Kanawha Rivers are the two nonbraided rivers for which no algorithm hits the 35%
RRMSE mark or better. The case of the Cumberland River highlights the unforgiving nature of the RRMSE
metric in characterizing algorithm performance in periods of low flows. As can be seen in Figure 5, begin-
ning from approximately day 90, discharge was reduced to less than 250 m3/s, though discharge had aver-
aged 1481 m3/s from days 1 to 90, and all of the algorithms capture this rapid drop in flow. For example,
from days 1 to 90, MetroMan performance (characteristic of other algorithms for the Cumberland) was argu-
ably quite good: MRR was 221.8%, and SDRR was 2.8%, giving an RRMSE of 22.0%. From days 91 to 162,
MetroMan performance was quite poor: MRR was only 25.2%, but SDRR was 62.5%. Indeed, from days 91
to 162, low-head dams on the Cumberland constrained the slope to less than 1 cm/km, as shown in Figure
4: the average slope from days 91 to 162 was 0.4 cm/km. These low-slope conditions generally were still
handled fairly well by Manning’s equation; note that the upstream reaches (e.g., reach 1 on the Cumberland
in Figure 4) have steeper slopes than downstream, a fact that the algorithms can exploit. However, for a
period of 20 days (days 125–144), the slope dropped to below 0.1 cm/km for all four reaches, and was
sometimes negative; the average during this time was 0.07 cm/km. Recall that a slope error of 0.1 cm/km
was added to all slope observations. Manning’s equation led to very high relative errors, as both slopes and
flows approached zero, even though absolute errors were quite minor, during this time. This combination
of hydraulic conditions meant that overall RRMSE for MetroMan was 44.6%. In contrast to the Cumberland,
the SDRR for the Kanawha was generally negligible for all algorithms; the RRMSE is dominated by bias, with
MetroMan again representative, with an MRR of 259.6%. The reason for this poor performance can likely be
attributed to a poor prior estimate of flow. The MFCR in this case has an MRR of 260.2%; this is the most
negative estimate except for the Tanana River. Sensitivity tests have shown that the MetroMan estimation

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4539

Figure 5. Output hydrographs. Output hydrographs from each of the six proposed algorithms are shown for each of the 19 rivers as compared to true discharge. Note that all algorithms
match observed river dynamics quite well, and most of the error reported in Table 2 is here shown to be bias. AMHG was not run for 7/19 rivers in this study, as these violated previously
published restrictions on that methodology.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4540

Table 5. Summary of Discharge Estimation Output According to Three Metricsa

RRMSE (%) MRR (%) SDRR (%)

River Name AMHG GaMo MetroMan MFG MFCR Median AMHG GaMo MetroMan MFG MFCR Median AMHG GaMo MetroMan MFG MFCR Median

Connecticut n/a 29.2 696.5 43.6 161.3 93.9 n/a 19.7 571.5 243.0 149.7 84.7 n/a 21.6 399.1 6.9 60.1 40.6
Cumberland 175.6 47.9 44.6 50.6 45.5 47.2 246.5 22.7 214.5 242.6 227.3 218.8 169.9 48.0 42.3 27.5 36.5 43.5
Garrone (DS) 80.5 128.8 70.4 16.3 87.6 86.8 264.6 95.4 59.2 13.9 73.6 71.7 48.1 86.7 38.2 8.7 47.5 49.0
Mississippi (DS) 98.3 35.7 72.8 58.8 28.4 59.2 98.3 235.6 271.7 250.9 228.2 257.1 0.4 2.9 12.5 29.5 3.3 15.8
Sacramento (DS) 58.6 26.8 4.8 50.4 42.0 6.0 55.3 22.5 22.1 7.0 41.1 0.1 19.6 14.6 4.3 50.0 8.7 6.0
St. Lawrence (DS) n/a 77.4 61.3 63.6 23.4 61.3 n/a 71.2 260.5 25.8 217.1 260.5 n/a 30.5 9.7 63.6 16.0 9.7
Ganges 93.2 279.8 152.8 36.8 133.2 133.3 93.2 214.2 122.9 220.1 101.8 103.8 1.2 180.2 91.0 30.8 86.0 83.7
Kanawha n/a 46.7 59.7 51.5 60.4 55.5 n/a 246.4 259.6 251.3 260.2 255.4 n/a 4.9 4.0 4.5 4.1 3.7
Ohio n/a 42.7 33.5 41.4 42.8 41.8 n/a 241.5 231.5 239.3 241.9 240.6 n/a 9.9 11.5 13.2 8.8 10.0
Platte n/a 347.7 215.9 77.0 223.0 219.3 n/a 174.3 106.0 274.5 106.0 104.7 n/a 307.9 192.6 20.1 200.8 197.3
Po 66.4 29.8 13.9 73.6 43.4 28.1 66.3 229.6 3.3 266.1 35.7 228.1 3.4 3.9 13.5 32.3 24.6 1.8
Seine 33.9 22.5 9.1 45.3 42.1 22.2 227.4 221.7 22.6 241.7 241.7 221.2 19.9 5.9 8.8 17.7 5.8 6.6
Severn n/a 17.6 1028.7 22.5 89.6 48.9 n/a 24.1 799.3 217.9 75.1 37.8 n/a 17.3 651.6 13.7 49.1 31.2
Tanana n/a 83.6 54.5 76.0 73.6 73.0 n/a 283.4 254.3 276.0 273.4 272.9 n/a 5.8 4.7 1.7 5.3 4.2
Garrone (US) 86.2 28.9 153.1 63.2 10.4 29.5 280.9 22.4 145.3 262.7 1.4 24.8 29.7 18.4 48.2 7.9 10.3 15.9
Mississippi (US) 88.2 35.1 5.9 41.4 36.3 36.4 88.2 234.9 25.8 240.4 235.8 235.9 2.1 3.8 1.0 8.8 5.9 6.1
Sacramento (US) 11.0 69.1 383.7 87.4 128.7 69.7 1.2 64.7 374.7 41.6 125.0 67.1 10.9 24.5 82.9 77.0 30.4 18.8
St. Lawrence (US) n/a 63.6 36.7 39.9 30.8 37.9 n/a 263.4 236.6 222.6 230.7 237.4 n/a 4.7 3.0 32.9 2.9 6.2
Wabash 98.4 146.1 26.2 92.2 134.9 27.9 28.0 99.1 22.5 274.4 117.6 23.5 94.6 107.7 13.5 54.7 66.4 15.0
Median 86.2 46.7 59.7 50.6 45.5 48.9 28.0 22.7 22.1 241.7 1.4 218.8 19.6 17.3 13.5 20.1 16.0 15.0
Mean 80.9 82.1 164.4 54.3 75.7 62.0 19.2 22.1 98.2 235.1 24.8 4.8 36.4 47.3 85.9 26.4 35.4 29.7
Standard Deviation 40.0 87.0 261.2 20.0 54.9 46.9 63.2 77.9 228.5 31.4 69.7 57.1 49.9 75.8 162.6 20.9 46.0 44.4
a
Bold typeface indicates best performer among algorithms for each river and each metric.

ability is especially impacted by low-biased prior estimates of mean streamflow. Thus, MetroMan estimation
accuracy is expected to be better for a prior with a positive bias than for a negative bias. AMHG was not run
for the Kanawha, as only 1.5% of cross sections passed the low-b filter. This combination of a poor, negative
bias in the prior, and the inability to use AMHG, highlights a weakness in the overall set of algorithms.
In addition to overall performance, the stability of the algorithms across rivers is of critical importance to
the SWOT mission and to discharge inversion more generally. Stability is assessed by SDRR and its relation-
ship to the MRR per river, and also by summarizing RRMSE across rivers. For all 19 rivers, at least one algo-
rithm had an SDRR less than 31%. We had hypothesized that the ensemble median would be the most
stable algorithm, and indeed it had an SDRR less than 30% in 13 cases, tying for best SDRR performance
with GaMo, which also had 13 rivers less than 30% SDRR. For the ensemble median algorithm, the median
SDRR value across the 16 nonbraided rivers was 12.5%. Considering the summary metrics in Table 5, MFG
and MFCR, two approaches that require an a priori flow, performed quite well; one of these two had the
lowest SDRR for 8/19 rivers.
Overall, most algorithms had an even distribution of negative and positive residual bias across rivers, as
indicated by MRR. However, AMHG had a positive bias for 7/11 rivers (note that eight rivers were considered
unsuitable for AMHG estimation as they were either braided or did not pass the width variability filters
described in section 3.1), and MFG had a strong negative bias, overestimating flow in 16/19 cases. The dif-
ference between the WBM mean annual flow estimate and the mean true flow for each river (see Table 2)
can be compared with the MRR results in Table 5. Across all 19 rivers, the median WBM bias is 237.9%. Five
of the six algorithms (AMHG, GaMo, MetroMan, MFCR, and the median) have a median MRR that is less than
WBM.
Figure 5 shows that certain rivers were more easily estimated than others. In particular, all algorithms had
an RRMSE less than 45% for the Seine, confirming it as the most easily estimated river. In addition, we might
expect that the MetroMan and GaMo algorithms would perform similarly from river to river, as each is based
on solving for unknown parameters in Manning’s equation. This is confirmed in our results, as these two
algorithms both had an RRMSE less than 35% for the Downstream Sacramento, Seine, Po, Upstream Missis-
sippi, and Ohio Rivers (Figure 6). Also, while estimations from the Cumberland and Kanawha rivers were
poorer than other rivers, all algorithms but AMHG had very similar estimation accuracies in these rivers.
Beyond predictably poor performance in braided rivers, there are no apparent trends in algorithm perform-
ance based on river size, flow range, WSE variability, latitude, or hydraulic regime. A possible exception here

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4541

Figure 6. RRMSE for each river. Note that the y axis is truncated to 100% RRMSE for readability: RRMSE exceeds this limit in 18/114 cases.

is the Platte, for which algorithms performed quite poorly, and was the only kinematic river analyzed. Note
that kinematic rivers show no time variability in water surface slope, which for some algorithms similar to
those tested here leads to poor performance [Durand et al., 2010]. However, since the Platte is also braided,
and other kinematic rivers were not included, no ﬁrm conclusion can be drawn; further work is required.
Given that the data here are models of rivers forced with imposed ﬂows, comparison of algorithm perform-
ance against morphology is inappropriate. However, Gleason et al. [2014] found no apparent trends in esti-
mation accuracy for the AMHG algorithm based on river morphology in their study, so the lack of clear
physical controls on discharge estimation accuracy is perhaps unsurprising.

5. Discussion
This discharge algorithm intercomparison has highlighted both successes and failures of algorithms thus
far. First, we have seen that there is always an algorithm that estimates discharge to within 35% or less
RRMSE for 14/16 nonbraided rivers. Second, bias has proven itself to be a significant component of these
errors; the median algorithm had an SDRR less than 30% for 13/19 rivers, but an RRMSE less than 30% for
only 5/19 rivers. Third, we have seen that one single algorithm has not emerged as the ideal approach.
Instead, we hypothesize that the community will likely require moving forward with multiple approaches.
These results can be used to map future algorithm developments within the SWOT community.
Simply put, the addition of more a priori data and site-specific assumptions should make each of the algo-
rithms more effective. However, this study was truly blind. Thus, in all cases, there was information we knew
about certain rivers that we purposefully did not include in our algorithms: all methods received the same
data and were forced to make the same blind assumptions. Some prior estimate of roughness, AHG, dis-
charge, or some other variable is available for every measured or modeled river on the planet. Leveraging
these data should improve the performance of the methods here, as will using these methods in conjunc-
tion with hydrologic models in assimilation schemes, which is the likely way forward for using SWOT to esti-
mate global river discharge.
The variable nature of algorithm performance in this study confirms the difficulty of the problem and the
value of such a large-scale comparison. Each of the AMHG, GaMo, and MetroMan algorithms performed
worse than their previously published accuracies in some cases, although Gleason et al. [2014] also found
varied performance for the AMHG algorithm in their study of 34 rivers. These mixed results occur despite
the fact that inputs to the algorithms were ‘‘perfect,’’ i.e., not degraded to match expected temporal

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4542

sampling and observation error likely from SWOT. Yoon et al. [2016] found that MetroMan accuracy was not
significantly affected by using SWOT temporal sampling and expected measurement errors on the Sacra-
mento River as compared with daily sampling, but this result will likely vary among algorithms and rivers.
When faced with real-world data, these algorithms will behave differently, and likely worse, than they have
here if algorithms are not improved or ancillary data as described above are not included.
Our results strongly point to the need for algorithm synergy when conceiving of a SWOT discharge product
that will be viable for all of the world’s rivers. We have tested the easiest likely approach for this synergy
(the ensemble median), but each of these algorithms should be able to inform and constrain one another.
For example, we may find that some algorithms tend to perform poorly for certain kinds of rivers, and in
these cases, we ought to exclude those algorithms from the ensemble. As another example, it may be bene-
ficial to create hybrid approaches built by combining components of the various algorithms. For example,
the variable n of MFG could be combined with MetroMan or GaMo. Exploring these kinds of synergy should
be a key goal of future algorithm development, together with identification of the best ancillary data for
use with these methods.
It is critical to begin to be able to predict algorithm performance based on river characteristics. Perhaps surpris-
ingly, the hydraulic regime analysis (Froude and kinematic wave numbers shown in Figure 2) did not correlate
with algorithm performance, except that the one truly kinematic river (the Platte) performed quite poorly. We
continue to explore this by discussing several of the cases included in the test cases: floodplain flow, low-head
dams, and multiple channels. It had been hypothesized prior to this algorithm intercomparison that perform-
ance would suffer during floodplain flow. An interesting case in this regard is the Po River. Figure 3 clearly
shows out-of-bank flows: some reaches increase their width by an order of magnitude during the high flow
event following day 300 of the simulation. The MFG algorithm is formulated with the assumption of in-bank
flow, and overestimates flow by a large margin during this period; this leads to an underprediction of flow dur-
ing the low-flow periods and an overall MRR of 266.1%, and an RRMSE of 73.6%. However, the GaMo and Met-
roMan algorithms perform well, with RRMSE of 29.8% and 13.9%, respectively. Thus, even the simple Manning
formulations appear in-and-of-themselves to be adequate even in the presence of significant floodplain flow.
This is in spite of the fact that the calibrated hydraulic models used to generate the synthetic observations in
this study typically used different values of n in the floodplain versus the channel, a fact not accounted for in
the discharge algorithms. As a caveat to this point, it should be noted that the Po River model is built in 1-D;
thus, we can really only conclude that 1-D floodplain flow is resolved by these algorithms.
Performance in the presence of low-head dams and human management was an open question prior to
this study. Significant dams are present on the Ohio River, leading to quite low slopes, as shown in Figure 4.
However, the MetroMan algorithm performs adequately on the Ohio, with RRMSE 5 33.5%, and
SDRR 5 11.5%. Manning’s equation captured the friction losses, even in this case with low slopes, and signif-
icant backwater profiles. Performance in the case of multiple channels was also expected to degrade per-
formance. The Seine is an example of a multichannel river; throughout the domain utilized here, the river is
often two channels. The simplest solution of merging the top widths and averaging the water surface
heights [Schubert et al., 2015]) led to adequate performance, here: AMHG, GaMo, and MetroMan had RRMSE
of 33.9%, 22.5%, and 9.1%, respectively. Thus, the presence of multiple channels is not necessarily a cause
of a great deal of algorithm performance degradation.
Finally, one source of error not considered or evaluated here is error in the hydraulic models themselves. As
each model is comprised of approximate hydraulic solutions imposed on in situ data of varying quality at
varying spatial sampling and with simplified physics compared to real-world flows, it is quite likely, indeed
almost certain, that reported widths and WSE values corresponding to reported discharges contain some
error. How this error contributes to discharge estimation will vary by algorithm and by river, and assessing
this effect on our estimations is well outside our purposes here. However, to the extent that the flow laws
as implemented in the algorithm represent reality (such as with the empirical components of MFG), this
effect should contribute little to discharge algorithm error.

5.1. Algorithm-Specific Discussion

5.1.1. At-Many-Stations Hydraulic Geometry
AMHG performed much more poorly in this study than in previous publications of the method, and this can
be almost exclusively attributed to issues caused by the greatly increased number of observations used

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4543

here. Previously, the method had only been tested with a maximum of 20 observations, and the increase to
yearly or longer periods of observations caused significant issues with both AHG and AMHG. This increased
number of observations also rendered the previous remotely sensed proxy for AMHG [Gleason and Smith,
2014] ineffective at adequately characterizing AMHG, as it is mathematically constrained to a value of 0.5
when there are order-of-magnitude changes in observed widths, which occurred in numerous study rivers.
An incorrect AMHG proxy gives an incorrect relationship between AHG parameters, and therefore inverted
AHG curves are constrained to a hydraulic space outside observed values. In addition, the minimum and
maximum imposed flows in the AMHG algorithm were developed and tested on rivers other than those
here (Gleason and Smith [2014] give ‘‘global’’ values of Qmin 5 minimum observed width 3 0.5 m depth 3
0.1 m/s velocity, Qmax 5 maximum observed width 3 10 m depth 3 5 m/s velocity). As before, using river-
specific estimates of maximum and minimum depth and velocity will improve the AMHG method.
There are also hydrologic and geomorphological reasons why more observations lead to poorer AMHG per-
formance. For example, Bonnema et al. [2016] show that for the Ganges River, AMHG exhibits a sharp
improvement in accuracy when wet and dry seasons are estimated separately, while Gleason and Hamdan
[2015] show the same result for the Ganges (using different data from Bonnema et al. [2016]), noting a
reduction from 56% to 28% RRMSE when considering dry-season flows only. This is because these wet and
dry flows (or flood and nonflood regime for temperate rivers) exhibit different AHG behavior and have
much different time mean w and Q values, leading to a situation where discharge cannot be inverted [Glea-
son and Wang, 2015]. Additionally, when flows vary widely, AHG breaks down and must be represented
with multiple power laws: one for each distinct channel geometry. Thus, it is recommended that AMHG be
performed in a stepwise manner when many observations are available, as binning observations into those
with similar magnitude widths should assure a single AHG curve per bin. As with all other algorithms,
including these kinds of river-specific constraints will improve AMHG performance.
5.1.2. Garambois and Monier
The GaMo algorithm relied upon QMFCR (which in turn relied upon the WBM mean annual flow estimate) for
defining a first guess and minimum and maximum bounds for discharge. First guess and identification
bounds are also defined for n given usual values found in the literature [e.g., Chow, 1959], and for low flow
cross-sectional area from observed cross-section variations. Presumably, it is because of the use of these
bounds that some of the very large RRMSE values observed with MetroMan are not observed with GaMo
(compare Table 5), even though they are based on similar hydraulic flow laws. Note too, that despite relying
on QMFCR , the GaMo algorithm resulted in better or comparable performances than MFCR for 16/19 rivers
(Table 5). This suggests that the GaMo approach can take some advantage of the hydraulic information
from SWOT-like data to constrain river discharge estimation.
Future improvements to GaMo could involve adding additional constraining equations, or adding additional
prior information, for example, to reduce the optimization zone in hydraulic parameter space. If discharge
was supposed constant in space, the number of unknowns is NR 3NP 11 and there are still NR 3NP equa-
tions, making the inverse problem more tractable [Garambois and Monnier, 2015]. This approach could be
of interest for (large) rivers with small changes in flow between several reaches. Garambois and Monnier
[2015] also derived a robust and accurate inference method in the case that one in situ water depth mea-
surement is available. Indeed, in this case, an explicit expression for the channel bed elevation is available
as a function of the water surface slope and width, independent of n. Next, given this channel bed eleva-
tion, an approach similar to GaMo allows accurate computation of the inflow discharge.
5.1.3. Metropolis Manning
This study highlighted one limitation of MetroMan in its current formulation, namely its sensitivity to the
prior estimate of discharge. From a Bayesian point of view, this is not a drawback, but a philosophical deci-
sion: if prior information about the river is available, it ought to be used, as in any of the algorithms. How-
ever, this does represent a limitation in this study; if the prior estimate of discharge is poor, then the
MetroMan discharge estimate will also be poor. This places MetroMan in a middle ground of the algorithms,
as AMHG utilizes no prior information and MFG and MFCR completely rely on the prior estimate of flow.
Having these different philosophies as part of the overall algorithm suite is important at this point in dis-
charge algorithm development. Interestingly, MetroMan appears to be more sensitive to an underestima-
tion of the prior mean annual flow than to overestimation. Sensitivity tests on the Po, Sacramento
Downstream, and the Tanana indicated that MetroMan performs far better for a 50% overestimation of
mean annual flow than it performs for a 50% underestimation. The cause of this behavior will be explored

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4544

Figure 7. Importance of time-variable roughness coefficient. (left) The average roughness coefficient (red) and effective roughness coefficient (blue) are shown. (right) Relationship
between the roughness coefficient variability (expressed through j, see equation (7)), and MetroMan bias.

in future work, and could in principle be addressed by running several MCMC chains with different mean
values.
An additional drawback of MetroMan (and indeed MFCR and GaMo, as well) is that it assumes that n (or
AHG in AMHG) does not change in time, i.e., with flow depth. Thus, any changes in bed or bank material,
increases in debris, or reorganization of channel geometry will lead to suboptimal discharge estimation.
This decision was originally made in algorithm formulation in order to limit the number of unknowns, but it
agrees with the hydraulic models in this study. In most of these models, n does not vary in time at each
cross section, except for marginal changes during out of bank flow. Therefore, this assumption is secure for
these model data where debris and changes in substrate are generally not considered, but less secure for
future observational data.
An additional drawback is that MetroMan invokes Manning’s equation at a river reach, not at a cross section.
It is possible that Manning’s equation does not hold for reaches, even when using the true reach-averaged
roughness coefficient, and even if Manning’s equation holds at each cross section. To analyze this, we esti-
mated the ‘‘effective’’ reach average roughness coefficient for each river by calculating it at each time step
from reach-average hydraulic quantities:

~ 5Q21 A5=3 W22=3 S1=2

n (8)

In Figure 7 (left), the effective roughness n~ is compared to the reach average roughness coefficient n; the
former is calculated from (8), while the latter is calculated as the average value of the roughness coefficient
across all cross sections in a given reach. As flow decreases, n ~ increases dramatically, while n remains con-
stant. Analysis of this phenomenon revealed this behavior was due to the increase at low flow of within-
reach spatial variability of the cross-sectional area, width, and slope, combined with the nonlinear nature of
Manning’s equation. While it is tempting to analyze this in terms of fluvial geomorphology or the physics of
open channel flow, this finding is completely mathematical: n ~ diverges from n due to averaging of within-
reach spatial variability.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4545

Figure 7 (right) shows that the more n ~ diverges from n, the worse MetroMan’s bias. Here the divergence of
~ from n is characterized by a factor j5ðn=n
n ~ Þ 21. Therefore, j50 means that there is no within-reach
hydraulic variability as n5n ~ , and the more negative the j, the more variability there is (note that j < 0,
~ > n). Of particular interest is the situation described above at low flow, where j becomes more and
since n
more negative as n ~ increases but n stays constant. Figure 7 plots the minimum j for each river versus Met-
roMan MRR, and for minimum j values less than 20.5, MetroMan MRR often becomes very large. Thus, the
more within-reach variability of stream hydraulics, the worse MetroMan’s performance. Future work must
consider how to include time-varying n in MetroMan, and presumably other algorithms; work to incorporate
the time-varying n from MFG into MetroMan framework is ongoing.
It would be reasonable to assume that the significant increase in n ~ at low flow would impact GaMo in the
same way that it impacts MetroMan, as both use reach-averaged quantities, require n to be temporally
invariant, and solve for A0 and n by requiring continuity among reaches. However, the GaMo discharge
results do not diverge for high values of j. The reason for this is presumably because GaMo specifies rela-
tively restrictive minimum and maximum values of discharge, while MetroMan invokes prior information
utilizing a cost function. Because MetroMan assumed a relatively high uncertainty in the prior mean annual
flow estimate, and very precise observations with no allowance for error in Manning’s equation itself, unre-
alistic discharge estimates were produced in the cases with large variations in n. GaMo did not allow these
unrealistic discharge estimates because it was limited to the range of discharge estimated by MFCR.
5.1.4. Mean Flow and Geomorphology
Due to the empirical determination of some parameters, this method is not expected to work well for
braided rivers and tidal reaches, where the controls on the flow resistance (and thus Manning n) are very
different than in single or slightly anastomosed channels. Additionally, the accuracy and applicability of the
algorithm is expected to be limited by the need for the prior mean annual flow. By design, the MRR for the
MFG algorithm is highly correlated (R2 5 0.86) with the WBM bias. For example, the WBM mean annual flow
estimate for the upstream Garonne underestimates the flow during the experiment by 65%, while the MFG
MRR is 263%. One additional limitation of MFG is that it is not expected to work well for floodplains, since
out-of-bank flow was not used in deriving the relations. Indeed, MFG performs relatively poorly for the Po
(RRMSE 5 73.6%) due in part to significant overestimation of high flows (see Figure 5).
For the application of this algorithm, future efforts should focus on improving the empirical aspects of the
method with additional site-specific information, and over time with more observations. This is particularly
important in reaches that are braided or tidally influenced where unique more specific relations may be
developed. It is also possible that other geomorphologic information about the channel planform (including
meander length, sinuosity, and channel type) can inform the estimation of Manning n, as indicated by Bjer-
klie [2007]. Additionally, the Manning equation, which forms the physical basis for this method, also forms
the basis for the MetroMan, GaMo, and MFCR algorithms and as such the three independent methods
should eventually converge to similar values provided the assumptions and relations used in both corrobo-
rate each other. For example, the optimization scheme used in MetroMan or GaMo to optimize A0 and n
could also be used to optimize the value of B, and the value of n in MetroMan and GaMo could be esti-
mated a priori from various empirical relations and channel morphology.
5.1.5. Mean Flow and Constant Roughness
The philosophy of the MFCR algorithm was to preserve the mean flow estimated by WBM, while simply
assuming a default n value of 0.03. Any error in WBM flow should propagate directly in error in the MFCR
algorithm. However, there are cases where the WBM estimate is too low but the MFCR is biased too high,
e.g., in the case of the Connecticut River. The mean HEC-RAS flow for the Connecticut in this study (June–
December, 2011) was 1208 m3/s, whereas the WBM estimated flow for this time was 394.1 m3/s. Therefore,
the MFCR method should have resulted in a lower estimated discharge, but this was not the case as MRR
for the Connecticut was 150%. This resulted because a stage invariant roughness coefficient of 0.03 was
applied to all cross sections in the MFCR. However, there was dramatic variability among the cross sections
being averaged together to create river reaches, leading to issues like those described above for MetroMan.
Moreover, there were relatively few cross-sections available: for the first reach of 3.7 km in length, only three
cross sections were used to build the reach with time-averaged top width varying from 241 to 608 m, fur-
ther increasing the within-reach hydraulic variability. This resulted in maximum n ~ peaking at greater than
0.3, an order of magnitude greater than the specified roughness coefficient of 0.03. Therefore, given the

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4546

constraints of a roughness coefficient of 0.03 and a nonnegative flow area, flow is estimated as far higher
than specified mean annual flow.
Intriguingly, there is a slight dependency of the overall MFCR bias on the average Froude number of each
river; lower Froude numbers had more tendency to have a low bias, and vice versa; the mean Froude num-
ber explains approximately 53% of the variance in the bias of the median estimate. In general, larger higher
Strahler order streams have lower slope and lower Froude numbers, and these rivers generally had more
negative bias. Presumably, this has to do with the WBM simulations used to estimate the MFCR mean
annual flow.
These results highlight interesting conclusions about simple ‘‘prior’’ type flow estimations that point to the
need for the more complex algorithms discussed before. First, the low value of mean flow from WBM did
not agree with the large observed cross-sectional area changes, underscoring the dangers of relying on
prior conditions to estimate discharge. Second, discharge estimation via the prior resulted in the opposite
bias as WBM, pointing to how critical it is to develop a better understanding of how n ~ behaves. Finally, our
results suggest that using only a few cross sections cannot be used to represent a reach average.
5.1.6. Ensemble Median
Thus far, it is difficult to predict which algorithm will perform best on each river; thus, the ensemble median
is a highly attractive option. It is somewhat unexpected that the ensemble median is not the top performer
on any of the 19 rivers, on the basis of RRMSE. However, it is quite encouraging that overall, the median is
the least biased of any of the six estimates; the ensemble median algorithm has a mean MRR across the 19
rivers of just 4.8%. However, the standard deviation of the ensemble median algorithm’s MRR is 57.1%, sug-
gesting that variation in algorithm performance at each river was varied enough to render the ensemble
median unreliable. It is expected that this issue will be addressed in the future, as the described major issues
with reach averaging are explored, and prior and/or river-specific information is incorporated in discharge
estimation. As individual algorithms improve, so too will their ensemble products.

6. Conclusion
We are encouraged by the current state of SWOT discharge algorithms: 14/16 nonbraided rivers had an
algorithm estimate discharge within 35% RRMSE of true flow. These results include rivers with complex real
world hydraulics like 1-D floodplain flow, low-head dams, and simple multichannel rivers. Some of these
complex hydraulics and geomorphologies can have a strong effect on the methods discussed here: extreme
flood events, braided rivers, and two-stage AHG all lead to decreased performance. Discharge algorithms
must be improved to handle these cases, and methods to quantitatively predict algorithm performance
based on river morphology must be developed. Moreover, the experiments conducted in this study used
idealized synthetic daily observations with minimal noise; future studies designed to test SWOT’s ability to
estimate discharge need to use SWOT-like temporal sampling and error characteristics and work with
Acknowledgments
Funding for this work was provided by observed field and airborne data sets when possible. Our results from this experiment also indicate that
NASA SWOT Science Definition Team algorithm improvement is needed if a robust, global discharge product is to be delivered from SWOT, as no
grants NNX13AD96G and
NNX13AD88G, NASA Terrestrial
single algorithm or their ensemble median performed with consistently accurate results. The MetroMan
Hydrology Program grant algorithm comes closest, as it was the most accurate algorithm in 9/16 nonbraided rivers, but even this
NNX13AD05G, NASA SWOT Algorithm algorithm is subject to very large discharge errors in other cases. We conclude that our restrictive experi-
Definition Team, and CNES SWOT
Science Definition Team grant
ment design, where no ancillary data or river-specific assumptions were allowed is a likely cause of many of
(TOSCA). The authors thank Alison the poor results, substantiated by previous studies using each of the AMHG, GaMo, and MetroMan algo-
Macneil of the NOAA/National rithms. Future work seeking to estimate discharge from remotely sensed platforms should include as much
Weather Service Northeast River
Forecast Center for providing the ancillary data as are available, and also seek to develop multialgorithm synergy to improve stability and
Connecticut River HEC-RAS model, and accuracy of derived discharge retrievals.
Albert Kettner for providing WBM
discharge estimates. Mike Jasinski and
two anonymous reviewers provided References
comments that helped improve the
quality of the manuscript. If interested Adams, T., S. Chen, R. Davis, T. Schade, and D. Lee (2010), The Ohio River Community HEC-RAS Model, pp. 1512–1523, Am. Soc. of Civ. Eng.,
in gaining access to data or codes Reston, Va.
utilized in this study, contact Michael Andreadis, K. M., E. A. Clark, D. P. Lettenmaier, and D. E. Alsdorf (2007), Prospects for river discharge and depth estimation through assimila-
Durand ([email protected]). tion of swath-altimetry into a raster-based hydrodynamics model, Geophys. Res. Lett., 34(10), L10403, doi:10.1029/2007GL029721.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4547

Bates, P. D., M. D. Wilson, M. S. Horritt, D. C. Mason, N. Holden, and A. Currie (2006), Reach scale floodplain inundation dynamics observed
using airborne synthetic aperture radar imagery: Data analysis and modeling, J. Hydrol., 328(1-2), 306–318, doi:10.1016/
j.jhydrol.2005.12.028.
Bates, P. D., M. S. Horritt, and T. J. Fewtrell (2010), A simple inertial formulation of the shallow water equations for efficient two-
dimensional flood inundation modeling, J. Hydrol., 387(1-2), 33–45, doi:10.1016/j.jhydrol.2010.03.027.
Besnard, A., and N. Goutal (2008), Comparison between 1d and 2d models for hydraulic modeling on a flood plain: Case of Garonne river,
paper presented at River Flow 2008, Meeting sponsor was IAHR-FHC, UNESCO-IHP, UNESCO-IS, Izmir, Turkey, 3–5 Sep.
Biancamaria, S., K. M. Andreadis, M. Durand, E. A. Clark, E. Rodriguez, N. M. Mognard, D. E. Alsdorf, D. P. Lettenmaier, and Y. Oudin (2010),
Preliminary characterization of SWOT hydrology error budget and global capabilities, IEEE J. Selec. Top. Appl. Earth Obs. Remote Sens.,
3(1), 6–19, doi:10.1109/jstars.2009.2034614.
Biancamaria, S., M. Durand, K. M. Andreadis, P. D. Bates, A. Boone, N. M. Mognard, E. Rodriguez, D. E. Alsdorf, D. P. Lettenmaier, and
E. A. Clark (2011), Assimilation of virtual wide swath altimetry to improve arctic river modeling, Remote Sens. Environ., 115(2), 373–381,
doi:10.1016/j.rse.2010.09.008.
Biancamaria, S., D. P. Lettenmaier, and T. M. Pavelsky (2015), The SWOT mission and its capabilities for land hydrology, Surv. Geophys., 37,
307–337, doi:10.1007/s10712-015-9346-y.
Birkinshaw, S. J., G. M. O’Donnell, P. Moore, C. G. Kilsby, H. J. Fowler, and P. A. M. Berry (2010), Using satellite altimetry data to augment
flow estimation techniques on the mekong river, Hydrol. Process., 24(26), 3811–3825, doi:10.1002/hyp.7811.
Bjerklie, D. M. (2007), Estimating the bankfull velocity and discharge for rivers using remotely sensed river morphology information, J.
Hydrol., 341(3-4), 144–155, doi:10.1016/j.jhydrol.2007.04.011.
Bjerklie, D. M., S. L. Dingman, C. J. Vorosmarty, C. H. Bolster, and R. G. Congalton (2003), Evaluating the potential for measuring river dis-
charge from space, J. Hydrol., 278(1-4), 17–38, doi:10.1016/S0022-1694(03)00129-X.
Bjerklie, D. M., D. Moller, L. C. Smith, and S. L. Dingman (2005a), Estimating discharge in rivers using remotely sensed hydraulic information,
J. Hydrol., 309(1-4), 191–209, doi:10.1016/j.jhydrol.2004.11.022.
Bjerklie, D. M., S. L. Dingman, and C. H. Bolster (2005b), Comparison of constitutive flow resistance equations based on the Manning and
Chezy equations applied to natural rivers, Water Resour. Res., 41, W11502, doi:10.1029/2004WR003776.
Bonnema, M., S. Sikder, F. Hossain, M. Durand, C. Gleason, and D. M. Bjerklie (2016), Benchmarking wide swath altimetry based river dis-
charge estimation algorithms for thethe ganges river system, Water Resour. Res., 52, 2439–2461, doi:10.1002/2015WR017296.
Brunner, G. W. (2010), HEC-RAS River Analysis System: Hydraulic Reference Manual, Version 4.1, US Army Corps of Engineers, 417 pp.
Chow, V. (1959), Open-Channel Hydraulics, McGraw-Hill, N. Y.
Di Baldassarre, G., G. Schumann, and P. Bates (2009), Near real time satellite imagery to support and verify timely flood modeling, Hydrol.
Processes, 23(5), 799–803, doi:10.1002/hyp.7229.
Dingman, S. L., and D. M. Bjerklie (2006), Estimation of river discharge, in Encyclopedia of Hydrological Sciences, vol. 5, chap. 61, John Wiley
& Sons, Hoboken, N. J., doi:10.1002/0470848944.hsa069.
Durand, M., K. M. Andreadis, D. E. Alsdorf, D. P. Lettenmaier, D. Moller, and M. Wilson (2008), Estimation of bathymetric depth and
slope from data assimilation of swath altimetry into a hydrodynamic model, Geophys. Res. Lett., 35, L20401, doi:10.1029/
2008GL034150.
Durand, M., E. Rodriguez, D. E. Alsdorf, and M. Trigg (2010), Estimating river depth from remote sensing swath interferometry measure-
ments of river height, slope, and width, IEEE J. Selec. Top. Appl. Earth Obs. Remote Sens., 3(1), 20–31, doi:10.1109/JSTARS.2009.2033453.
Durand, M., J. Neal, E. Rodriguez, K. M. Andreadis, L. C. Smith, and Y. Yoon (2014), Estimating reach-averaged discharge for the river severn
from measurements of river water surface elevation and slope, J. Hydrol., 511, 92–104, doi:10.1016/j.jhydrol.2013.12.050.
Fjortoft, R., et al. (2014), Karin on swot: Characteristics of near-nadir ka-band interferometric sar imagery, IEEE Trans. Geosci. Remote Sens.,
52(4), 2172–2185, doi:10.1109/TGRS.2013.2258402.
Garambois, P.-A., and J. Monnier (2015), Inference of effective river properties from remotely sensed observations of water surface, Adv.
Water Resour., 79, 103–120, doi:10.1016/j.advwatres.2015.02.007.
Gelman, A., J. B. Carlin, H. S. Stern, and D. B. Rubon (2004), Bayesian Data Analysis, 675 pp., Chapman and Hall, Boca Raton, Fla.
Gleason, C. J., and A. N. Hamdan (2015), Crossing the (watershed) divide: Satellite data and the changing politics of international river
basins, Geogr. J., doi:10.1111/geoj.12155, in press.
Gleason, C. J., and L. C. Smith (2014), Toward global mapping of river discharge using satellite images and at-many-stations hydraulic
geometry, Proc. Natl. Acad. Sci. U. S. A., 111(13), 4788–4791, doi:10.1073/pnas.1317606111.
Gleason, C. J., and J. Wang (2015), Theoretical basis for at-many-stations hydraulic geometry, Geophys. Res. Lett., 42, 7107–7114, doi:
10.1002/2015GL064935.
Gleason, C. J., L. C. Smith, and J. Lee (2014), Retrieval of river discharge solely from satellite imagery and at-many-stations hydraulic geome-
try: Sensitivity to river form and optimization parameters, Water Resour. Res., 50, 9604–9619, doi:10.1002/2014WR016109.
Gosling, S. N., and N. W. Arnell (2011), Simulating current global river runoff with a global hydrological model: Model revisions, validation,
and sensitivity analysis, Hydrol. Processes, 25(7), 1129–1145, doi:10.1002/hyp.7727.
Goutal, N., and F. Maurel (2002), A finite volume solver for 1d shallow-water equations applied to an actual river, Int. J. Numer. Methods
Fluids, 38(1), 1–19, doi:10.1002/fld.201.
Heniche, M., Y. Secretan, P. Boudreau, and M. Leclerc (2000), A two-dimensional finite element drying-wetting shallow water model for riv-
ers and estuaries, Adv. Water Resour., 23(4), 359–372, doi:10.1016/S0309-1708(99)00031-7.
Humphries, E., T. M. Pavelsky, and P. Bates (2014), Two dimensional hydrodynamic modeling of a high latitude braided river, Abstract
H43H-1042 presented at 2014 Fall Meeting, AGU, San Francisco, Calif., 15–19 Dec.
Hunger, M., and P. Doell (2008), Value of river discharge data for global-scale hydrological modeling, Hydrol. Earth Syst. Sci., 12(3), 841–861.
Hunter, N. M., et al. (2008), Benchmarking 2d hydraulic models for urban flooding, Proc. ICE Water Manage., 161(1), 13–30, doi:10.1680/
wama.2008.161.1.13.
Huntington, T. G. (2006), Evidence for intensification of the global water cycle: Review and synthesis, J. Hydrol., 319(1-4), 83–95, doi:
10.1016/j.jhydrol.2005.07.003.
Jung, H. C., M. Jasinski, J.-W. Kim, C. K. Shum, P. Bates, J. Neal, H. Lee, and D. Alsdorf (2012), Calibration of two-dimensional floodplain model-
ing in the central atchafalaya basin floodway system using sar interferometry, Water Resour. Res., 48, W07511, doi:10.1029/2012WR011951.
Kim, B., B. F. Sanders, J. E. Schubert, and J. S. Famiglietti (2014), Mesh type tradeoffs in 2D hydrodynamic modeling of flooding with a
Godunov-based flow solver, Adv. Water Resour., 68, 42–61, doi:10.1016/j.advwatres.2014.02.013.
Kouraev, A. V., E. A. Zakharova, O. Samain, N. M. Mognard, and A. Cazenave (2004), Ob’ river discharge from topex/poseidon satellite altim-
etry (1992-2002), Remote Sens. Environ., 93(1-2), 238–245, doi:10.1016/j.rse.2004.07.007.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4548

Lai, X., and J. Monnier (2009), Assimilation of spatially distributed water levels into a shallow-water flood model. Part I: Mathematical
method and test case, J. Hydrol., 377(1-2), 1–11, doi:10.1016/j.jhydrol.2009.07.058.
Larnier, K. (2010), Mod elisation thermohydraulique d’un tronçon de Garonne en lien avec l’habitat piscicole: Approches statistique et
deterministe, PhD thesis, Univ. of Toulouse, Toulouse, France.
Leopold, L. B., and T. Maddock (1953), The hydraulic geometry of stream channels and some physiographic implications, Geol. Surv. Prof.
Pap., 252, 64 pp.
Maswood, M., and F. Hossain (2015), Advancing river modeling in ungauged river basins using remote sensing: The case of Ganges-
Brahmaputra-Meghna Basins, J. River Basin Manage., 14, 103–117, doi:10.1080/15715124.2015.1089250.
Mersel, M. K., L. C. Smith, K. M. Andreadis, and M. T. Durand (2013), Estimation of river depth from remotely sensed hydraulic relationships,
Water Resour. Res., 49, 3165–3179, doi:10.1002/wrcr.20176.
Metropolis, N., A. W. Rosenbluth, M. N. Rosenbluth, A. H. Teller, and E. Teller (1953), Equation of state calculations by fast computing
machines, J. Chem. Phys., 21(6), 1087–1092, doi:10.1063/1.1699114.
Michailovsky, C. I., S. McEnnis, P. A. M. Berry, R. Smith, and P. Bauer-Gottwein (2012), River monitoring from satellite radar altimetry in the
zambezi river basin, Hydrol. Earth Syst. Sci., 16(7), 2181–2192, doi:10.5194/hess-16-2181-2012.
Neal, J. C., N. A. Odoni, M. A. Trigg, J. E. Freer, J. Garcia-Pintado, D. C. Mason, M. Wood, and P. D. Bates (2015), Efficient incorporation of
channel cross-section geometry uncertainty into regional and global scale flood inundation models, J. Hydrol., 529(P1), 169–183,
doi:10.1016/j.jhydrol.2015.07.026.
Oki, T., and S. Kanae (2006), Global hydrological cycles and world water resources, Science, 313(5790), 1068–1072, doi:10.1126/
science.1128845.
Oki, T., T. Nishimura, and P. Dirmeyer (1999), Assessment of annual runoff from land surface models using total runoff integrating pathways
(trip), J. Meteorol. Soc. Jpn., 77(1B), 235–255.
Paiva, R. C. D., M. T. Durand, and F. Hossain (2015), Spatiotemporal interpolation of discharge across a river network by using synthetic
swot satellite data, Water Resour. Res., 51, 430–449, doi:10.1002/2014WR015618.
Pavelsky, T. M. (2014), Using width-based rating curves from spatially discontinuous satellite imagery to monitor river discharge, Hydrol.
Processes, 28(6), 3035–3040, doi:10.1002/hyp.10157.
Pavelsky, T. M., M. T. Durand, K. M. Andreadis, R. E. Beighley, R. C. D. Paiva, G. H. Allen, and Z. F. Miller (2014), Assessing the potential global
extent of swot river discharge observations, J. Hydrol., 519, 1516–1525, doi:10.1016/j.jhydrol.2014.08.044.
Rawlins, M. A., R. B. Lammers, S. Frolking, B. M. Fekete, and C. J. Vorosmarty (2003), Simulating pan-arctic runoff with a macro-scale terres-
trial water balance model, Hydrol. Processes, 17(13), 2521–2539, doi:10.1002/hyp.1271.
Roux, H., and D. Dartus (2005), Parameter identification using optimization techniques in open-channel inverse problems, J. Hydraul. Res.,
43(3), 311–320, doi:10.1080/00221680509500125.
Rogers, W. (2014), Central Valley Floodplain Evaluation and Delineation, Subtask 5, combined Sacramento River System Model technical
841memorandum, submitted to California Department of Water Resources, Sacramento, Calif.
Roux, H., and D. Dartus (2006), Use of parameter optimization to estimate a flood wave: Potential applications to remote sensing of rivers,
J. Hydrol., 328(1-2), 258–266, doi:10.1016/j.jhydrol.2005.12.025.
Roux, H., and D. Dartus (2008), Sensitivity analysis and predictive uncertainty using inundation observations for parameter estimation in
open-channel inverse problem, J. Hydraul. Eng., 134(5), 541–549, doi:10.1061/(ASCE)0733-9429(2008)134:5(541).
Schubert, J., W. Monsen, and B. Sanders (2015), Metric-resolution 2D river modeling at the macroscale: Computational methods and appli-
cations in a Braided River, Frontiers Earth Sci., 3(74), 1–22, doi:10.3389/feart.2015.00074.
Siddique-E-Akbor, A. H. M., F. Hossain, H. Lee, and C. K. Shum (2011), Inter-comparison study of water level estimates derived from hydro-
dynamic–hydrologic model and satellite altimetry for a complex deltaic environment, Remote Sens. Environ., 115(6), 1522–1531, doi:
10.1016/j.rse.2011.02.011.
Smith, A., C. Sampson, and P. Bates (2015), Regional flood frequency analysis at the global scale, Water Resour. Res., 51, 539–553, doi:
10.1002/2014WR015814.
Smith, L. C. (1997), Satellite remote sensing of river inundation area, stage, and discharge: A review, Hydrol. Processes, 11(10), 1427–1439,
doi:10.1002/(SICI)1099-1085(199708)11:10 < 1427::AID-HYP473 > 3.0.CO;2-S.
Smith, L. C., and T. M. Pavelsky (2008), Estimation of river discharge, propagation speed, and hydraulic geometry from space: Lena River,
Siberia, Water Resour. Res., 44, W03427, doi:10.1029/WR006133.
Smith, L. C., B. L. Isacks, A. L. Bloom, and A. B. Murray (1996), Estimation of discharge from three braided rivers using synthetic aperture
radar satellite imagery: Potential application to ungaged basins, Water Resour. Res., 32(7), 2021–2034, doi:10.1029/96WR00752.
Tinkler, K. J. (1982), Avoiding error when using the Manning equation, J. Geol., 90(3), 326–328.
Trigg, M. A., M. D. Wilson, P. D. Bates, M. S. Horritt, D. E. Alsdorf, B. R. Forsberg, and M. C. Vega (2009), Amazon flood wave hydraulics,
J. Hydrol., 374(1-2), 92–105, doi:10.1016/j.jhydrol.2009.06.004.
Turnipseed, D. P., and V. B. Sauer (2010), Discharge measurements at gaging stations, U.S. Geol. Surv. Tech. Methods, Book 3, Chap. A8, 87 pp.
Vieira, J. H. D. (1983), Conditions governing the use of approximations for the saint-venant equations for shallow surface-water flow, J.
Hydrol., 60(1-4), 43–58, doi:10.1016/0022-1694(83)90013-6.
Vilmin, L., N. Flipo, C. de Fouquet, and M. Poulin (2015), Pluri-annual sediment budget in a navigated river system: The seine river (france),
Sci. Total Environ., 502, 48–59, doi:10.1016/j.scitotenv.2014.08.110.
V€
or€osmarty, C. J., B. M. Fekete, M. Meybeck, and R. B. Lammers (2000), Global system of rivers: Its role in organizing continental land mass
and defining land-to-ocean linkages, Global Biogeochem. Cycles, 14(2), 599–621, doi:10.1029/1999GB900092.
Widen-Nilsson, E., S. Halldin, and C.-Y. Xu (2007), Global water-balance modeling with wasmod-m: Parameter estimation and regionalisa-
tion, J. Hydrol., 340(1-2), 105–118, doi:10.1016/j.jhydrol.2007.04.002.
Wisser, D., B. M. Fekete, C. J. Voeroesmarty, and A. H. Schumann (2010), Reconstructing 20th century global hydrography: A contribution to
the global terrestrial network- hydrology (gtn-h), Hydrol. Earth Syst. Sci., 14(1), 1–24, doi:10.5194/hess-14-1-2010.
Yoon, Y., M. Durand, C. J. Merry, E. A. Clark, K. M. Andreadis, and D. E. Alsdorf (2012), Estimating river bathymetry from data assimilation of
synthetic swot measurements, J. Hydrol., 464, 363–375, doi:10.1016/j.jhydrol.2012.07.028.
Yoon, Y., P.-A. Garambois, R. C. D. Paiva, M. Durand, H. Roux, and R. E. Beighley (2016), Improved error estimates of a discharge algorithm
for remotely sensed river measurements: Test cases on Sacramento and Garonne rivers, Water Resour. Res., 52, 278–294, doi:10.1002/
2015WR017319.

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4549

Estimating River Discharge With Swath Altimetry: A Proof of Concept Using AirSWOT Observations
No ratings yet
Estimating River Discharge With Swath Altimetry: A Proof of Concept Using AirSWOT Observations
8 pages
Sys 002
No ratings yet
Sys 002
19 pages
Constraining Remote River Discharge Estimation Using Reach Scale Geomorphology
No ratings yet
Constraining Remote River Discharge Estimation Using Reach Scale Geomorphology
24 pages
Jurnal Altimetri Helene Oktafiola
No ratings yet
Jurnal Altimetri Helene Oktafiola
23 pages
2010a Durand
No ratings yet
2010a Durand
12 pages
Estimating River Depth From Remote Sensing Swath Interferometry Measurements of River Height, Slope, and Width
No ratings yet
Estimating River Depth From Remote Sensing Swath Interferometry Measurements of River Height, Slope, and Width
12 pages
wrcr.20176 Gis Rs
No ratings yet
wrcr.20176 Gis Rs
15 pages
4
No ratings yet
4
18 pages
remotesensing-16-01808
No ratings yet
remotesensing-16-01808
25 pages
Water 15 03470 v4
No ratings yet
Water 15 03470 v4
16 pages
Water Resources Research - 2020 - Rodr Guez - Observing Rivers With Varying Spatial Scales - 2
No ratings yet
Water Resources Research - 2020 - Rodr Guez - Observing Rivers With Varying Spatial Scales - 2
42 pages
RivWidthCloud_An_Automated_Google_Earth_Engine_Algorithm_for_River_Width_Extraction_From_Remotely_Sensed_Imagery
No ratings yet
RivWidthCloud_An_Automated_Google_Earth_Engine_Algorithm_for_River_Width_Extraction_From_Remotely_Sensed_Imagery
5 pages
Water Resources Research - 2021 - Altenau - The Surface Water and Ocean Topography SWOT Mission River Database SWORD A
No ratings yet
Water Resources Research - 2021 - Altenau - The Surface Water and Ocean Topography SWOT Mission River Database SWORD A
15 pages
1-s2.0-S0034425715301747-main
No ratings yet
1-s2.0-S0034425715301747-main
14 pages
Water 15 03813
No ratings yet
Water 15 03813
12 pages
Hou - Hess 22 6435 2018
No ratings yet
Hou - Hess 22 6435 2018
14 pages
Article 1
No ratings yet
Article 1
24 pages
Estimation of River Discharge From Non Trapezoidal Open Channel Using QuickBird 2 Satellite Imagery Utilisation Des Images Satellites de Quickbird 2
No ratings yet
Estimation of River Discharge From Non Trapezoidal Open Channel Using QuickBird 2 Satellite Imagery Utilisation Des Images Satellites de Quickbird 2
15 pages
Clark OSTST Poster2010 PDF
No ratings yet
Clark OSTST Poster2010 PDF
1 page
Water Diplomacy in Action: Contingent Approaches to Managing Complex Water Problems
From Everand
Water Diplomacy in Action: Contingent Approaches to Managing Complex Water Problems
CSPtrade
No ratings yet
Remote Sensing: Hydrology: Progress in Physical Geography October 2009
No ratings yet
Remote Sensing: Hydrology: Progress in Physical Geography October 2009
21 pages
A Novel Method of Satellite Based River Discharge Estimation Using River Hydraulic Geometry Through Genetic Algorithm Technique
No ratings yet
A Novel Method of Satellite Based River Discharge Estimation Using River Hydraulic Geometry Through Genetic Algorithm Technique
14 pages
Near Field
No ratings yet
Near Field
28 pages
Feysia Et Al 2014. Automated Water Extraction Index
No ratings yet
Feysia Et Al 2014. Automated Water Extraction Index
13 pages
hydrology-12-00025
No ratings yet
hydrology-12-00025
41 pages
s41597-024-03923-8
No ratings yet
s41597-024-03923-8
7 pages
Remotesensing 16 00656 v2
No ratings yet
Remotesensing 16 00656 v2
38 pages
Schumann Et Al. - 2008 - Comparison of Remotely Sensed Water Stages From LiDAR, Topographic Contours and SRTM-annotated PDF
No ratings yet
Schumann Et Al. - 2008 - Comparison of Remotely Sensed Water Stages From LiDAR, Topographic Contours and SRTM-annotated PDF
14 pages
Remotesensing 16 00184
No ratings yet
Remotesensing 16 00184
19 pages
River Conservation and Oversight
From Everand
River Conservation and Oversight
Bhagwanti Kakkar
No ratings yet
1) Fleuves
No ratings yet
1) Fleuves
20 pages
Complexity of Transboundary Water Conflicts: Enabling Conditions for Negotiating Contingent Resolutions
From Everand
Complexity of Transboundary Water Conflicts: Enabling Conditions for Negotiating Contingent Resolutions
Lawrence Susskind
No ratings yet
Do Remote Sensing Mapping Practices Adequately Address Localized Flooding? A Critical Overview
No ratings yet
Do Remote Sensing Mapping Practices Adequately Address Localized Flooding? A Critical Overview
17 pages
Hydrology for Engineers
From Everand
Hydrology for Engineers
Akula Achari
No ratings yet
50 Year Flood Hec Ras
No ratings yet
50 Year Flood Hec Ras
9 pages
Estimation of Multiple Inflows and Effective Channel by Assimilation of
No ratings yet
Estimation of Multiple Inflows and Effective Channel by Assimilation of
24 pages
Azizian Brocca 2019
No ratings yet
Azizian Brocca 2019
3 pages
Remotesensing 15 04697
No ratings yet
Remotesensing 15 04697
20 pages
Hwre Ta1
No ratings yet
Hwre Ta1
12 pages
A Method of Watershed Delineation For Flat Terrain Using Sentinel-2A Imagery and DEM: A Case Study of The Taihu Basin
No ratings yet
A Method of Watershed Delineation For Flat Terrain Using Sentinel-2A Imagery and DEM: A Case Study of The Taihu Basin
18 pages
remotesensing-11-00943
No ratings yet
remotesensing-11-00943
4 pages
sustainability-16-11041
No ratings yet
sustainability-16-11041
18 pages
Improving The Resolution and Accuracy of Groundwater Level Anomalies Using The Machine Learning-Based Fusión Model in The North China Plain
No ratings yet
Improving The Resolution and Accuracy of Groundwater Level Anomalies Using The Machine Learning-Based Fusión Model in The North China Plain
19 pages
Establishing Stage-Discharge Rating Curves in Developing Countries: Lake Tana Basin, Ethiopia
No ratings yet
Establishing Stage-Discharge Rating Curves in Developing Countries: Lake Tana Basin, Ethiopia
27 pages
Testing UAV-Derived Topography For Hydraulic Modelling in A Tropical Environment
No ratings yet
Testing UAV-Derived Topography For Hydraulic Modelling in A Tropical Environment
25 pages
Water 12 03198 v3
No ratings yet
Water 12 03198 v3
26 pages
Flood Monitoring Throgh Remote Sensing PDF
100% (1)
Flood Monitoring Throgh Remote Sensing PDF
212 pages
Surface Water Storage in Rivers and Wetlands
No ratings yet
Surface Water Storage in Rivers and Wetlands
35 pages
On The Role of Digital Terrain Topography and Land Use Dynamics in Flood Hazard Assessment of Urban Floodplain
No ratings yet
On The Role of Digital Terrain Topography and Land Use Dynamics in Flood Hazard Assessment of Urban Floodplain
26 pages
Essoar 10502735 1
No ratings yet
Essoar 10502735 1
26 pages
(IJCST-V10I4P18) :fatima Ibrahim Abdelmoutalib, Dieter Fritsch
No ratings yet
(IJCST-V10I4P18) :fatima Ibrahim Abdelmoutalib, Dieter Fritsch
10 pages
Prediction of Discharge From Water Surface Width in A Braided River With Implications For At-A-Station Hydraulic Geometry
No ratings yet
Prediction of Discharge From Water Surface Width in A Braided River With Implications For At-A-Station Hydraulic Geometry
11 pages
Griffith_2001
No ratings yet
Griffith_2001
18 pages
Hydr-2011jhm1345 1
No ratings yet
Hydr-2011jhm1345 1
22 pages
Barriers to Adopting Satellite Remote Sensing for Water Quality Management Done
No ratings yet
Barriers to Adopting Satellite Remote Sensing for Water Quality Management Done
12 pages
Azizian Brocca 2019
No ratings yet
Azizian Brocca 2019
3 pages
University of Sao Paulo, School of Engineering, Brazil E-Mail: (Rmlucci Webpage
No ratings yet
University of Sao Paulo, School of Engineering, Brazil E-Mail: (Rmlucci Webpage
9 pages
Neural Networks and M5 Model Trees in Modelling Water Level-Discharge Relationship
No ratings yet
Neural Networks and M5 Model Trees in Modelling Water Level-Discharge Relationship
16 pages
Mancini 等 - 2024 - Heuristic estimation of river bathymetry in braided streams using digital image processing
No ratings yet
Mancini 等 - 2024 - Heuristic estimation of river bathymetry in braided streams using digital image processing
24 pages
Water Assignment
No ratings yet
Water Assignment
17 pages
Daft OTAD 13e PPT Ch05
No ratings yet
Daft OTAD 13e PPT Ch05
21 pages
Mondelēz International - Case Study
No ratings yet
Mondelēz International - Case Study
2 pages
LECTURE NOTES ON Sustainable Livelihood & Sustainable Devt
No ratings yet
LECTURE NOTES ON Sustainable Livelihood & Sustainable Devt
10 pages
April Revision Sheet Grade 5. Model. Answers.
No ratings yet
April Revision Sheet Grade 5. Model. Answers.
9 pages
The Benefits of Cool Roofs
No ratings yet
The Benefits of Cool Roofs
6 pages
DAM Smart PDF
No ratings yet
DAM Smart PDF
66 pages
Standard Plate Count Agar (Apha
No ratings yet
Standard Plate Count Agar (Apha
3 pages
Sustainable Development Goals (SDG) Nomor 6 (Air Bersih & Sanitasi)
No ratings yet
Sustainable Development Goals (SDG) Nomor 6 (Air Bersih & Sanitasi)
2 pages
Wastewater Treatment Plants For Hospitals in Kerala, India - GME
No ratings yet
Wastewater Treatment Plants For Hospitals in Kerala, India - GME
12 pages
Pollution Prevention Management Plan
No ratings yet
Pollution Prevention Management Plan
29 pages
Acid Rain and Ozone
No ratings yet
Acid Rain and Ozone
2 pages
Norlito Jr. C. Llano Bsce 3B Topic 3: Water Balance
No ratings yet
Norlito Jr. C. Llano Bsce 3B Topic 3: Water Balance
11 pages
Public Open Spaces
No ratings yet
Public Open Spaces
96 pages
Water As An Ecological Factor and A Physical Environment of Life
100% (1)
Water As An Ecological Factor and A Physical Environment of Life
2 pages
Dept.: Electr. Power & M/Cs Course:Power System Economics Code: EPE422 4 Year Electr. Power & M/Cs Sheet (1) /2 Pages Prof. Mahdi El-Arini
No ratings yet
Dept.: Electr. Power & M/Cs Course:Power System Economics Code: EPE422 4 Year Electr. Power & M/Cs Sheet (1) /2 Pages Prof. Mahdi El-Arini
2 pages
GRI 302 - Energy 2016
No ratings yet
GRI 302 - Energy 2016
18 pages
A Manuscript Utilizing Corn Cobs Ash As Suitable Admixture For Eco Friendly Concrete Hollow Blocks
No ratings yet
A Manuscript Utilizing Corn Cobs Ash As Suitable Admixture For Eco Friendly Concrete Hollow Blocks
3 pages
Keenjhar Lake
No ratings yet
Keenjhar Lake
13 pages
Sustainable Living
No ratings yet
Sustainable Living
1 page
Action Research Revision
No ratings yet
Action Research Revision
11 pages
Vent Pipe Specs p29
No ratings yet
Vent Pipe Specs p29
64 pages
Unit 1 Basic Maintenance and Repair
No ratings yet
Unit 1 Basic Maintenance and Repair
9 pages
Radioactive Pollution
No ratings yet
Radioactive Pollution
2 pages
Reflective Report - Managing Resources in an International Business Environment
No ratings yet
Reflective Report - Managing Resources in an International Business Environment
10 pages
Natural vegetation images
No ratings yet
Natural vegetation images
3 pages
Blue Accelerator
No ratings yet
Blue Accelerator
2 pages
Ehs & Esg Mis Report_lntep_mmyy
No ratings yet
Ehs & Esg Mis Report_lntep_mmyy
18 pages
The Neogene and Quaternary Periods
No ratings yet
The Neogene and Quaternary Periods
17 pages
Management' Administration' Management Management Management
No ratings yet
Management' Administration' Management Management Management
20 pages
Why Are Some Animals Endangered?
No ratings yet
Why Are Some Animals Endangered?
8 pages

An Intercomparison of Remote Sensing River Discharge Estimation Algorithms From Measurements of River Height, Width, and Slope

Uploaded by

An Intercomparison of Remote Sensing River Discharge Estimation Algorithms From Measurements of River Height, Width, and Slope

Uploaded by

PUBLICATIONS

Water Resources Research

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4527

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4528

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4529

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4530

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4531

SWOT DISCHARGE ALGORITHM INTERCOMPARISON

which increase in the downstream direction.

Table 3. Summary of Discharge Algorithms

3.1. At-Many-Stations Hydraulic Geometry (AMHG)

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4535

3.2. The GaMo Algorithm

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4536

3.3. Metropolis-Manning (MetroMan)

3.4. Mean-Annual Flow and Geomorphology (MFG)

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4537

3.5. Identification Window for MetroMan and Gamo

3.6. The MFCR and Ensemble Median Algorithms

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4538

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4539

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4540

Table 5. Summary of Discharge Estimation Output According to Three Metricsa

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4541

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4542

5.1. Algorithm-Specific Discussion

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4543

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4544

~ 5Q21 A5=3 W22=3 S1=2

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4545

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4546

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4547

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4548

DURAND ET AL. SWOT DISCHARGE ALGORITHM INTERCOMPARISON 4549

You might also like