ProcED: a MATLAB package for processing ADCP estuarine data

ProcED: a MATLAB package for processing ADCP estuarine data

Fernando Genz^I; Mauro Cirano^II; Guilherme Camargo Lessa^II

^IDepartamento de Engenharia Sanitária e Ambiental, Escola Politécnica, Universidade Federal da Bahia, Rua Prof. Aristides Novis, 2, Federação, 40210-630 Salvador, BA, Brazil. Phone: [phone] - E-mail: [e-mail]

^IICentro de Pesquisa em Geofísica e Geologia, Instituto de Física, Universidade Federal da Bahia, Campus Ondina, 40170-280 Salvador, BA. Brazil - E-mails: [e-mail] ; [e-mail]

ABSTRACT

ProcED is a MATLAB based computational package designed to facilitate the handling of a large amount of data derived from mount vessel ADCPs that monitor tidal flows and discharge in estuarine cross-sections. The package routines were written to process water current information obtained with an RD Instruments Rio Grande ADCP and its WinRiver software - version 1.03. They are capable to eliminate bad ensembles, to extrapolate current velocities for the surface and bottom blanked areas and to smooth noisy cross-profiles, enhancing the visualization of the velocity field. By performing space and time correlation between cell velocities, ProcED allows for the computation of the residual velocity field, both in the entire cross-section and in the discrete vertical profiles. Results are provided in graphical and table formats, the latter also including a table with the main flow characteristics such flooding and ebbing time, tidal elevation and current asymmetry, tidal prism and computations errors. A comparison with computations performed by WinRiver shows that ProcED discharge calculations are in average 5.5% smaller. The apparent underestimation of ProcED computations is ascribed to interpolation of the cross-profiles and to its different bottom extrapolation method. In the bottom area there is a discharge overestimation by WinRiver.

Keywords: acoustic current meters, ADCP, current profiles, data processing, MATLAB, estuaries.

RESUMO

ProcED é um conjunto de rotinas computacionais para ambiente MATLAB. elaborado para facilitar a manipulação da grande Quantidade de dados de correntes medidas em seção transversal com ADCP embarcado durante o ciclo da maré em região estuarina. As rotinas foram desenvolvidas para processar os dados de corrente obtidos com o ADCP Rio Grande da RD Instruments e o seu aplicativo WinRiver reversão 1.03. As rotinas tem a capacidade de eliminar dados inválidos, extrapolar as correntes para as áreas nao medidas (superfície e fundo), bem como suavizar os campos de velocidade, destacando sua visualização. Estabelecendo a correspondência espacial e temporal das células de velocidade, o ProcED) permite o cálculo das correntes residuais para toda a seção, assim como em posições específicas. As saídas do ProcED) são geradas na forma gráfica e em arquivo texto, incluindo uma tabela que apresenta as principais características do fluxo, tais como o tempo de duração da maré enchente e vazante, as velocidades média e máxima da corrente, o prisma da maré e os erros das medidas do ADCP A comparação com os cálculos efetuados pelo WinRiver mostra que as vazões determinadas pelo ProcED são 5,5% menores em média. A aparente subestimativa nas vazões obtidas com o ProcED é atribuída à diferença dos métodos de interpolação na seção transversal e na extrapolação das correntes para a região do fundo, na qual o WinRiver deve estar superestimando as vazões.

Palavras-chave: correntómetros acústicos, ADCP, perfil de correntes, processamento de dados, MATLAB, estuários.

INTRODUCTION

Acoustic Doppler Current Profilers (ADCP's) operated from moving boats have been largely used in the characterization of horizontal and vertical flow structures (Lane et al., 1997; Peters, 1997; Sylaios & Boxall, 1998; Rippeth et al., 2002; Reed et al., 2004; Sepúlveda et al., 2004; Piedracoba et al., 2005; Schettini et al., 2006; Stanev et al., 2007), in quantifying water exchange (Chadwick & Largier, 1999) and in the estimate of suspended sediment transport (Gartner, 2004; Hoitink & Hoekstra, 2005; Madron et al., 2005; Merckelbach, 2006; Zaleski & Schettini, 2006; Schettini & Zaleski, 2006; Schettini et al., 2009) in estuaries and coastal bays.

The monitoring of water fluxes in estuarine channels is generally done with multiple passages of the instrument along a preestablished cross-section. Due to problems associated with flow turbulence, navigation and the occurrence of bad ensembles, the position of the velocity cells in successive cross-profiles¹ are rarely coincident, thus causing computing problems. Computation becomes even more complex due to the amount of data gathered along the monitoring of a full tidal cycle. Several tens of ensembles in each cross-profile are summed up, resulting in many thousand velocity measurements seldom correlated in space. If the transport of solutes or suspended elements is also considered, it is still necessary to synchronize the time of current and concentration data acquisition (Chadwick & Largier, 1999).

In order to facilitate the data treatment and analysis, a MAT-LAB freeware package named ProcED (available at <http://www.mcirano.ufba.br/ftp/pub/matlab/proced>) was written to process water current information obtained with an RD Instruments Rio Grande ADCP and its WinRiver software - version 1.03 - through a tidal cycle. This article aims to explain the routine, to present its validation and to show the type of output generated.

DATA PROCESSING

Processing water current data from a complete tidal cycle involves the following steps, modified from Chadwick & Largier (1999): i) preliminary data inspection of each cross-profile; ii) spatial data interpolation of each cross-profile; iii) spatial velocity interpolation at even steps of time between the beginning and the end of the tidal cycle; iv) measurement error estimates; and v) determination of the main characteristics of the tidal cycle, i.e., flooding and ebbing time, average and maximum flow magnitude and velocities, tidal asymmetry and tidal prism.

ProcED requires the definition, while in the field, of well established limits for the cross-section (fixed buoys), in order to maintain the cross-section width and orientation unaltered. Navigation between the limits must be kept as straight as possible. The left margin of the channel (seaward facing) is taken as a reference for cross-section orientation, whereas positive (negative) velocity values indicate seaward (landward) directed flows. ProcED data preparation involves the following:

a) Data inspection and selection of fully measured crossprofiles with WinRiver in playback mode, taking note of the profile orientation (left margin to right margin or viceversa) and eliminating useless segments of cross-profiles (e.g. turns preceding the definite route towards the opposite margin).

b) Data exportation in P-file format using WinRiver version 1.03².

c) Creation of a four-column input file to the routine ADCP_ProcED, containing the P-file names (*p.000), the time of the cross-profile (hours and fraction), distance in relation to the cross-section reference buoy (in meters), and a flag indicating the need to invert the cross-profile orientation (0 = no, 1 = yes).

d) Definition of the angle between the channel axis (ebb orientated) and the true north. This can be done either through a map/satellite image or through the dominant ebb-flow direction measured during the monitoring.

The ProcED package is form by routines ADCP_ProcED, rdpadcp.m (Pawlowicz, 2004), rot.m (Beardsley, 2004), smartmean.m, isoline.m, ADCP_ProcED_error and tidal _cycle_analysis, allowing for:

(1) The computation of the along-channel component of the current velocity (u, henceforth referred to as current velocity).

(2) The elimination of bad ensembles through interpolation.

(3) The extrapolation of the alongshore current component to the upper and lower blanked areas (the same WinRiver layers).

(4) The smoothing of the raw data by averaging three adjacent horizontal cells (optional, if contrasting values between adjacent ensembles occur).

(5) The establishment of a regular grid (matrix i, j) with spatial correspondence between all cross-profiles in the tidal cycle.

(6) The computation of the average current value for each cell, in the case of more than one cross-profile exists for a given time.

(7) The computation of the total cross-section area and water discharge.

(8) The determination of the average cross-section velocity (U).

(9) The representation of the velocity field in non-dimensional depths.

(10) The computation of the residual velocity field.

(11) The extraction of velocity profiles, in non-dimensional depths, at the location of hydrographic monitoring stations.

(12) Smoothing of the output for visualization purposes.

The spatial interpolation provides the means to calculate the velocities for a same location u(i, j) in every cross-profile, considering the need to invert the orientation of the crossprofile (Fig. 1a). The depths (i) for each column (j) are turned into non-dimensional depths, which is the base for time (t) integrations.

It is suggested that when running ProcED the user chooses the average ensembles number for the horizontal resolution and the same ADCP vertical resolution.

ProcED outputs include:

• The velocity fields for every time interval, in a graphic format and in a MATLAB file (.mat) (tidal_cycle_matrix).

• A text file with the depth-averaged velocity for each ensemble (dacv_res).

• A text file with the cross-sectional average velocity, area and total discharge for the tidal cycle (disch_res).

• A residual velocity field for the tidal cycle, in a graphic format and in a MATLAB file (residual_field).

• A MATLAB file with the velocity profiles along the water column (non-dimensional depths) for the positions occupied by the hydrographic station(s) (inf2ctd).

• An estimation of the velocity and discharge error for every time interval (error_res).

• A text file with characteristic values for the tidal cycle (tidal_synthesis).

Time interpolation is another important feature of the package. The velocity field is interpolated at even time intervals between the beginning and the end of the tidal cycle, thus allowing for the computational of the residual velocity according to (Kjerfve, 1979):

where: j is the column position; i is the line number (i = 1,2,...,11 for Z = 0 to Z = 1,0.1 spaced); k = 2,...,n - 1; t_n- t₁ = T ; T is the length of the tidal cycle; n = T/Δt is the number of time intervals, and Δt is the interpolation time interval. T is determined by the routine through the interpolation of the computed discharges, defining the time when the discharge close to the end of the tidal cycle equals that at t₁. By default n = 25 in order to avoid underestimation of the velocity values close to the flood and ebb maxima.

Depth averaged velocity (), as well as the depth averaged of any water property (e.g. salinity, temperature or suspended sediment concentration), is calculated through the equation (Kjerfve, 1979):

where: is the mean value for the property in the water column; P is the value for the water property at the position Z_iin time t.

The random error of the Rio Grande ADCP measurements depends on the ADCP frequency, the height of the cell, the number of averaged pings (WP) and the geometry of the sound beams (RD Instruments, 1996). The velocity standard deviation is proportional to WP^-0.5. The random error for each cross-profile estimated by the routine is based on the standard deviation of the "error velocity" given for each cell by WinRiver (RD Instruments, 1996). In accordance RD Instruments (2002), the standard deviation of the error associated to a single ping in measurement mode WM1 is 0.181 m/s for 0.5 m cells and 0.066 m/s for 1 m cells.

To compute the mean random error for each cross-profile, the package uses the same computation method applied to the velocity, including time and space interpolations.

As a final product, the package synthesizes the tidal cycle by providing the flooding and ebbing time, the average velocities, the tidal prism and an asymmetry index (AI_DV)(see Table 1). This asymmetry index is defined by Mantovanelli et al. (2004):

where: t_eis the ebbing time; tf is the flooding time; _eis the average ebbing time; _f is the average flooding time. The AI_DVindex considers the combined effect of the tidal asymmetries in duration (A_D) and in mean velocity ( A_V) between the ebb and flood periods, since both asymmetries are important for determining net transports.

VALIDATION AND DISCUSSION

The field data used here was obtained in flow measurements at Rio Paraguaçu estuary (northeast coast of Brazil - Fig. 2) on October 26 and November 02 2003, during semi-diurnal spring and neap tides, respectively. The cross-section was approximately 1200 m wide and 32 m deep. Detailed hydraulic and hydrographic information on the estuary was published by Genz et al. (2006) and Genz (2006). The flow was measured with an ADCP Rio Grande 600 kHz with the following configuration: operation mode WM = 1; sampling frequency TP = 0.2 Hz; cell height WS = 1 m; pings per ensemble WP = 20; instrument depth ∼1 m; vertical integration time TE = 5 to 7 s. Boat speed varied between 2.5 and 3 m/s, resulting in cross-profiles 8 to 12 minutes long and ensembles 8 to 15 m wide.

The validation was performed by comparing WinRiver and ProcED results based on the measured and the extrapolated surface and bottom discharges. It is reminded that ProcED does not extrapolate the discharge between the limits of the charted cross-section and the margins of the channel, such as WinRiver. Nevertheless, the uncharted area is close to the boundary and flow velocities are normally small. Extrapolated discharges computed by WinRiver for the two tidal cycles under consideration averaged 1.1% and 0.7% of the instantaneous spring and neap discharges, respectively.

The total spring-tide discharges measured by WinRiver varied between -16,000 m³/s and 23,500 m³/s with an average velocity magnitude of 0.45 m/s ± 0.26 m/s. The velocity error varied between 0.014 m/s and 0.029 m/s, averaging 0.019 m/s (Table 1).

ProcED discharges were smaller than WinRiver's with minimum and maximum differences in the spring cycle of 0.8% and 11.5% (mean of 5.5%). A larger difference of 51.7% (318 m³/s) was observed during low-water slack, when discharge values are small enough to make differences negligible. By segmenting the cross-section into measured and extrapolated regions, the source of differences between the ProcED and WinRiver computations is identified. Smaller differences, between 0.1% and 6.5% (mean of 2.2%), occur when considering measured and extrapolated top discharges. Larger differences come about when comparing the extrapolations for the bottom blanked area. In this case estimated ProcED discharges were, in average, 45.6% lower than WinRiver's.

The total neap-tide discharges measured by WinRiver was around 7,200 m³/s in both flood and ebb directions with an average velocity magnitude of 0.15 m/s and a standard deviation of 0.10 m/s. The velocity error varied between 0.013 m/s and 0.018 m/s, averaging 0.015 m/s (Table 1). Although nominal errors at neap tide were smaller than at spring tide, relative errors become higher due to the smaller velocity magnitude. Figure 3 shows that the error values increase exponentially with velocity magnitudes smaller than 0.15 m/s.

ProcED neap discharges were smaller than WinRiver's with minimum and maximum differences of 1.7% and 8.4% (mean of 4.7%). Larger differences, as much as 48.5% (210 m³/s), were observed close to slack water (discharges <960 m³/s). If only the measured and extrapolated top discharges are considered, minimum and maximum differences are 0.1% to 2.1% (mean of 1.0%). In this case, slack water differences fall to a maximum of 3.7%. However, once again poor agreement exists between ProcED's and WinRiver's extrapolated bottom discharges. ProcED's discharge were in average 50.7% smaller than WinRiver's , with slack water differences reaching up to 1427%.

The discharge underestimation with ProcED is ascribed to: i) the data interpolation performed by ProcED to spatially adjust cross-profiles and ii) the WinRiver's extrapolation method to fill the bottom blanked area. Whereas the former causes the small differences observed in the measured and extrapolated top discharges, the latter causes major differences in the extrapolated bottom discharges. While ProcED interpolates a linear profile to a zero velocity at the bottom, WinRiver extrapolates a logarithmic profile based U* and z0. For reasons yet unknown, the averaged extrapolated WinRiver bottom discharges are similar to those on the surface, with average velocities only 20% smaller than those at the surface (both for spring and neap data sets). ProcED results calculate bottom discharges half of that on the surface and average bottom velocities 60% smaller than those at the surface. Another important aspect about WinRiver calculations is that when depth velocity profiles are not unidirectional, as it was the case for stratified flows in many cross-profiles at the neap tide (also at slack water spring), it fails to fit a correct velocity profile through the data to properly calculate the bottom discharge (Fig. 4).

Figure 5 shows a cross-profile produced by WinRiver and its counterpart computed by ProcED. Contrasting current directions and magnitudes are observed in neighboring ensembles in the WinRiver plots. This could be ascribed to excessive boat roll, water wave interferences or even the effect of eddies that were not eliminated with the short integration time (TE). These differences were smoothed after data filtering and interpolation in ProcED (Fig. 5c), allowing for an easier interpretation of the data.

 

The velocity profiles extracted for the position of the hydrographic station (Fig. 5c) are presented in Figure 6a in nondimensional depths. The capability of ProcED to establish correlation between velocity cells both in space and time permits the computation of the residual velocity, as it is shown in Figure 6b for a single vertical profile and in Figure 7 for the entire crosssection in both tidal cycles.

Different patterns of residual circulation between the spring and neap cycles are clearly shown in Figure 6. During the spring tide (Fig. 7a), lateral flow shear is caused by the channel geometry that steers the main flow to different sides of the channel in the ebbing and flooding tide. During neap tides (Fig. 7b) density gradients along the estuary are strong enough to cause gravitational circulation.

CONCLUSIONS

The ProcED package is capable of summarizing complex estuarine flow characteristics, which involve a significant amount of computation, in a few minutes. It allows the computation of the residual velocity field and discharge based on cross-profiles that are spatially non-corresponding. The package, and especially the ADCP-ProcED routine, performed well in discharge computations, presenting differences of 5.5% (in average) for two differenttidal cycles, in relation to discharge values measured by WinRiver.

ProcED discharges were generally smaller than WinRiver's . This relative underestimation is ascribed to ProcED's interpolation of the cross-profiles and to its different bottom extrapolation method. The extrapolation method used by WinRiver relies on a power law fit of the velocity profile, which was not suitable for several neap cross-profiles where stratified flow existed. Also, there is an apparent overestimation by WinRiver of the extrapolated bottom discharge.

Careful use of WinRiver for discharge estimates in estuarine environments is required, as partially mixed and stratified conditions leads to erroneous velocity profile extrapolations both in the bottom and the surface blanked areas with the power law method.

ACKNOWLEDGMENTS

Fernando Genz thanks CAPES and CNPq/CT-HIDRO for providing his scholarship. This work was part of a CNPq funded project (grant number 550205/2002-0). Mauro Cirano was supported by a CNPq research grant.

Received on April 16, 2009 / Accepted on April 16, 2010

Recebido em 16 abril, 2009 / Aceito em 16 abril, 2010

NOTES ABOUT THE AUTHORS

Fernando Genz is a Civil Engineer (UFRGS/1991) with a MSc in Water Resources (Hydrology, IPH/1994) and a PhD in Coastal and Sedimentary Geology (UFBA/2006). He has been an Associate Researcher at the Department of Environmental and Sanitary Engineering at UFBA since 2007. His research interest is on hydrology and surface water resources, especially associated to the large catchment areas of the semi-arid NE Brazilian region, as well as monitoring fluvial discharges and their impact on the hydrodynamics of estuaries.

Mauro Cirano is an oceanographer (FURG/1991) with a MSc in Physical Oceanography at University of São Paulo (IO-USP/1995) and a PhD in Physical Oceanography at the University of New South Wales (UNSW), Sydney, Australia (2000). Since 2004, he has been working as an Associate Professor at the Federal University of Bahia (UFBA). His research interest is the study of the oceanic circulation, based on data analysis and numerical modeling, area where he has conducting research projects over the last 15 years, focusing on the meso and large-scale aspects of the circulation.

Guilherme Camargo Lessa has a Geography degree from the Federal University of Minas Gerais (UFMG/1985) and a MSc in Geography from the Federal University of Rio de Janeiro (UFRJ/1990). He obtained his PhD in Marine Sciences at the University of Sydney (Australia) in 1994. He has worked as a Post-Doc at the following universities: i) Federal University of Paraná (UFPR), ii) Federal University of Bahia (UFBA) and iii) University of California - Berkeley. He joined UFBA as an Associate Professor in 1997. His research interests cover marine sedimentology, coastal and estuarine circulation and coastal (barrier-estuarine) geological evolution.

• BEARDSLEY B. 2004. BobStuff. Available at: <http://woodshole.er.usgs.gov/operations/sea-mat/>. Access on: Aug. 18, 2004.
• CHADWICK DB & LARGIER JL. 1999. The influence of tidal range on the exchange between San Diego Bay and the ocean. J. Geophys. Res., 104 (C12): 29.885-29.899.
• GARTNER JW. 2004. Estimating suspended solids concentrations from backscatter intensity measured by Acoustic Doppler Current Profiler in San Francisco Bay, California. Mar. Geol., 211(3-4): 169-187.
• GENZ F. 2006. Avaliação dos efeitos da Barragem de Pedra do Cavalo sobre a circulação Estuarina do Rio Paraguaçu e Baía de Iguape. PhD Thesis, Universidade Federal da Bahia, 245 pp. (in Portuguese, with English Abstr.). Available at: <http://www.goat.fis.ufba.br/uploads/userfiles/31.pdf>. Access on: Oct. 15, 2009.
• GENZ F, LESSA GC & CIRANO M. 2006. The Impact of an Extreme Flood upon the Mixing Zone of the Todos os Santos Bay, Northeastern Brazil. J. Coastal. Res., SI 39: 707-712.
• HOITINK AJK & HOEKSTRA P. 2005. Observations of suspended sediment from ADCP and OBS measurements in a mud-dominated environment. Coast. Eng., 52(2): 103-118.
• KJERFVE B. 1979. Measurement and analysis of water current, temperature, salinity and density. In: DYER KR. Estuarine Hydrography and Sedimentation. Cambridge, Cambridge University Press, pp. 186-226.
• LANE A, PRANDLE D, HARRISON AJ, JONES PD & JARVIS CJ. 1997. Measuring Fluxes in Tidal Estuaries: Sensitivity to Instrumentation and Associated Data Analyses. Estuar. Coast. Shelf Sci., 45: 433-451.
• MADRON XD, FERRE B, LE CORRE G, GRENZ C, CONAN P, PUJO-PAY M, BUSCAIL R & BODIOT O. 2005. Trawling-Induced resuspension and dispersal of muddy sediments and dissolved elements in the Gulf of Lion (NW Mediterranean). Cont. Shelf Res., 25: 2387-2409.
• MANTOVANELLI A, MARONE E, SILVA ET, LAUTERT LF, KLINGENFUSS MS, PRATA VP, NOERNBERG MA, KNOPPERS BA & ÂNGULO RJ. 2004. Combined tidal velocity and duration asymmetries as a determinant of water transport and residual flow in Paranaguá Bay estuary. Estuar. Coast. Shelf Sci., 59(4): 523-537.
• MERCKELBACH LM. 2006. A model for high-frequency acoustic Doppler current profiler backscatter from suspended sediment in strong currents. Cont. Shelf Res., 26: 1316-1335.
• PAWLOWICZ R. 2004. rdpadcp.m. Available at: <http://www2.ocgy.ubc.ca/~rich>. Access on: Aug. 27, 2004.
• PETERS H. 1997. Observations of Stratified Turbulent Mixing in an Estuary: Neap-to-spring Variations During High River Flow. Estuar. Coast Shelf Sci., 45: 69-88.
• PIEDRACOBA S, SOUTO C, GILCOTO M & PARDO PC. 2005. Hydrography and dynamics of the Ría de Ribadeo (NW Spain), a wave driven estuary. Estuar. Coast. Shelf Sci., 65: 726-738.
• RD INSTRUMENTS. 1996. Acoustic Doppler Current Profiler: Principles of Operation - A Practical Primer. P/N 951-6069-00. pp. 20-21.
• RD INSTRUMENTS. 2002. WorkHorse Installation Guide. P/N 957-6152-00. pp. 33-36.
• REED RE, GLASGOW HB, BURKHOLDER JM & BROWNIE C. 2004. Seasonal physical-chemical structure and acoustic Doppler current profiler flow patterns over multiple years in a shallow, stratified estuary, with implications for lateral variability. Estuar. Coast. Shelf Sci., 60: 549-566.
• RIPPETH TP, WILLIAMS E & SIMPSON JH. 2002. Reynolds Stress and Turbulent Energy Production in a Tidal Channel. J. Phys. Oceanogr., 32: 1242-1251.
• SCHETTINI CAF & ZALESKI AR. 2006. A Utilização de Perfiladores Acústicos de Corrente por efeito Doppler na Determinação do Material Particulado em Suspensão na Água: Aplicações. Revista Brasileira de Recursos Hídricos, 11(3): 201-210.
• SCHETTINI CAF, RICKLEFS K, TRUCCOLO EC & GOLBIG V. 2006. Synoptic hydrography of a highly stratified estuary. Ocean Dynamics, 56: 308-319.
• SCHETTINI CAF, ALMEIDA DC, SIEGLE E & ALENCAR ACB. 2009. A snapshot of suspended sediments and fluid mud distributions in the Tijucas Bay (Brazil): a mixed-energy environment. Geo-Mar Lett., 29: 1-2.
• SEPÚLVEDA HH, VALLE-LEVINSON A & FRAMIÑAN MB. 2004. Observations of subtidal and tidal flow in the Río de la Plata Estuary. Cont. Shelf Res., 24: 509-525.
• STANEV EV, FLEMMING BW, BARTHOLOMÄ A, STANEVA JV & WOLFF J-O. 2007. Vertical circulation in shallow tidal inlets and back-barrier basins. Cont. Shelf Res., 27: 798-831.
• SYLAIOS G & BOXALL SR. 1998. Residual Currents and Flux Estimates in a Partially-mixed Estuary. Estuar. Coast. Shelf Sci., 46: 671-682.
• ZALESKI AR & SCHETTINI CAF. 2006. Procedimentos para Calibração de Perfiladores Acústicos de Corrente por efeito Doppler para a Determinação da Concentração de Material Particulado em Suspensão na Água. Revista Brasileira de Recursos Hídricos, 11(3): 191-200.

1

A cross-profile refers to the velocity profiles (ensembles) acquired along a cross-section of the channel at a given time.

2

Version 1.06 alters data location in the P-files generating unrealistic data in

ProcED results. An attempt is being made to fix this compatibility problem, enabling the routine to read files in ASCII format, including the

WinRiver version II.

Publication Dates

• Publication in this collection
  03 Sept 2010
• Date of issue
  June 2010

History

• Accepted
  16 Apr 2010
• Received
  16 Apr 2009