Monitoring and calculation of bedload transport at the mountain torrent Urslau
Andrea KREISLER,1,* Markus MOSER,2 Michael TRITTHART,1 Johann AIGNER,1 Florian RUDOLF-MIKLAU,3 Helmut HABERSACK1
1 University of Natural Resources and Life Sciences, Christian Doppler Laboratory for Advanced Methods in River Monitoring, Modelling and Engineering, Vienna, Muthgasse 107, 1190 Vienna Austria
2 Austrian Service of Torrent and Avalanche Control, Johann Löcker Straße 3, 5580 Tamsweg, Austria
3 Ministry of Agriculture, Forestry, Environment and Watermanagement, Austrian Service for Torrent and Avalanche Control, Marxergasse 2, 1030 Vienna, Austria
*Corresponding author. E-mail: andrea.kreisler@boku.ac.at
An integrative bedload monitoring station was installed in 2010 at the downstream section of the Austrian torrent Urslau by the University of Life Sciences/Vienna, financed and supported by the Austrian Service of Torrent and Avalanche Control. At this measuring station direct (mobile bedload sampler, bedload trap) and indirect monitoring methods (geophone device) are applied. The combination of these bedload monitoring methods enables a comprehensive measurement and assessment of the bedload transport process. This paper presents the applied monitoring methods and discusses the specific restrictions and advantages of the applied devices. Exemplarily monitoring results of three years of bedload monitoring at the Urslau torrent are shown. Results concerning the temporal and spatial variability of the bedload transport process and the initiation of motion are given. Further, this paper presents outcomes of the measurements with the mobile basket sampler and the bedload trap like specific bedload rates, cross sectional bedload transport and the texture of the sediment material. The calibration of the geophone data, using direct bedload measurements is demonstrated. Measured bedload transport and the results of selected bedload transport equations for two different flow events of the Urslau are compared. For these events the equations provide a valuable reference point for practitioners.
Key words: integrative bedload monitoring system, basket sampler, geophone plates, bedload trap
1. INTRODUCTION
The knowledge of bedload transport in downstream sections of mountain torrents is essential for the examination of floods with sediment transport and for the design and implementation of protective structures. Reliable bedload data are essential for informed management decisions that affect the function of a river [Gomez, 2006]. Bedload monitoring is important to extend the understanding of the natural process and forms the prerequisite to select, apply, calibrate and validate bedload transport formulas [Habersack and Laronne, 2002] and numerical sediment transport models [Tritthart et al., 2012].
Although bedload transport is a central question in many issues it remains difficult to predict [Rickenmann et al., 2012] and to monitor. Many bedload transport formulas, where developed using
data of laboratory investigations with controlled boundary conditions, equilibrium transport, and bed level stability [Habersack and Laronne, 2002].
Comparisons with measured data showed, that measured transport rates differed from the calculated values [Bathurst, 1987; Rickenmann, 2001; Habersack and Laronne, 2002; Haddachi et al., 2013].
The measurement of bedload transport is complex as the process features a strong temporal and spatial variability [Habersack et al., 2012], varying grain sizes, high flow velocities, turbid flow [Lucia et al., 2013] and different modes of movements.
In 2010 an integrative bedload monitoring system has been installed at the downstream part of the Urslau torrent. Since 2011 bedload transport monitoring is performed.
The integrative monitoring system allows the
determination of the initiation of motion of bedload particles, bedload transport rates and total bedload yields. Additionally, bedload texture and the spatial and temporal variability of the transport process can be examined.
The integrative bedload monitoring system combines direct (mobile bedload sampler and bedload trap) and indirect (geophone device) methods. An important benefit of indirect bedload monitoring methods is the possibility to provide continuous records of the transport process over time and over the total channel width. Local field calibration using direct monitoring methods is required [Gray et al., 2010]. Gray et al. [2010]
stated that the importance of the employment of traditional basket samplers for calibration issues may will increase in future.
The application of basket samplers reveals many opportunities in bedload monitoring, but also limitations in quantifying the bedload transport process, as for example lowering the sampler on the streambed is restricted to a specific range of discharges and flow velocities [Habersack et al., 2010; Kreisler et al., 2010]. Further, using portable bedload samplers measurement duration is relatively short and the flow field may be disrupted [Gray et al., 2010]. The direct measurement of the bedload transport using bedload traps, which are installed in the stream bed, is possible at all water stages [Habersack et al., 2010] and until a filling stage of less than 80% does not disrupt the flow field [Habersack et al., 2001]. Bedload traps enable longer sampling durations and continuous recording of the bedload transport process [Reid et al., 1980;
Habersack et al., 2001; Kreisler et al., 2012].
The monitoring results of the Urslau torrent deliver a useful bedload data set, which allows a comparison with common used bedload transport equations and the measured bedload data.
The aim of this paper is to present the integrative monitoring system. Monitoring results of the direct and indirect measurement methods are illustrated and possibilities of the specific methods and the integrative monitoring system are discussed.
Further, bedload transport rates are calculated applying commonly used bedload transport formulae. The calculation results are compared to measured bedload data.
2. STUDY SITE
The Urslau torrent is located in the alpine region of the province Salzburg in Austria (Fig.1). It encompasses a drainage basin of 121.8 km² and is characterized by an average bed slope of 11.7%.
The mean flow at the gauging station Saalfelden, which is located near the junction of the Urslau and the Saalach River is 4.39 m3s-1.
The bedload monitoring station was built in 2010 and is situated near the village Maria Alm, at the downstream section of the Urslau torrent. The drainage area upstream the measurement station is 56 km2, bed width equals 8.5 m and the average bed slope near the station is 2 %.
There is a stream gauging station, 25 m upstream the bedload measuring station, which is operated by the hydrographic survey of Salzburg. Water level is measured using a hydrostatic pressure sensor and flow velocity measurements are conducted by a calibrated hydrometric impeller.
An array of geophones is embedded in the stream bed, where 7 geophones are installed at equidistant intervals of 1 m. The bedload trap is mounted in the middle of the measurement profile, directly upstream the geophone array. Measurements using the mobile basket sampler are conducted with a mobile crane from the orographic left torrent side.
Fig. 2 gives an overview of the arrangement of instruments at the measuring site.
Fig. 1 Location of catchment area upstream the measuring site in Salzburg
Fig. 2 Overview measuring site Urslau
3. BEDLOAD MONITORING METHODS 3.1 Indirect monitoring methods
Geophones are vibration sensors which originate from seismic technology. To detect bedload transport the geophone sensors are mounted on the underside of 0.36m long, 0.5m wide and 0.015m thick steel plates (Fig. 3a). These steel plates are embedded on the some level with the stream bed and are distributed over the whole cross-section. At the Urslau 7 geophones are arranged equidistantly in the measuring profile. The geophone array in the Urslau torrent is shown in Fig.3b.
Bedload particles which move over the steel plates produce vibrations which are registered by the geophone sensors. The geophone signal is sampled continuously at a rate of 10 kHz. To reduce storage space the recorded raw signal is processed and following data are stored: Sum of impulses per minute (whenever the geophone signal exceeds a given threshold an impulse is counted), the maximum amplitude per minute and the integral of the geophone signal. For special purposes it is possible to store values per second and even the raw data signal.
Bedload data originating from geophone measurements provide permanent information about the quality and distribution of bedload transport within the channel cross-section in high spatial and temporal resolution. Additionally the measurement is fully automated. The continuous recording enables the analyses of the bedload-transport events, the examination of the bedload transport – discharge relationship and the definition of the critical flow where bedload transport starts. Results concerning the transported bedload mass or bedload volume are achieved by calibrating the geophone signal with the direct bedload measurements. Good correlations are found between the geophone data and bedload mass, if bedload material larger than 10-30mm is regarded [Rickenmann and McArdell, 2007].
3.2 Direct monitoring methods
Direct bedload monitoring methods enable the determination of (specific) bedload rates and the
texture of the bedload material. In the following the basket sampler and the bedload trap, which are parts of the integrative monitoring system at the Urslau, are introduced.
Mobile basket samplers are applied in bedload monitoring for decades [Mühlhofer, 1933; Rijn, 1986]. The concept of the basket sampler, which we used at the Urslau torrent, is based on the mobile bedload trap presented in Bunte and Abt (2003). It consists of a steel frame, a sampler bag and a steel bar. The sampler intake has a width of 0.44 m and a height of 0.26 m. The mesh size of the sampler bag is 3.5 mm x 6.5 mm. Fig. 4a shows a picture of the sampler. Bedload transport measurements are done with a mobile crane and the help of two tether lines, which are fixed on the sides of the steel frame and on both riverbanks, prohibiting the sampler to drift downstream. Bedload measurements can be conducted either at several measurement verticals in the cross-section (“cross-section measurement”) or can be repeated at one single vertical (“permanent measurement”). Cross-section measurements enable, beside the identification of the specific bedload discharge at each vertical, the determination of the total bedload transport in the torrent cross section.
Permanent measurements are deployed mainly to analyze the temporal variability of the bedload transport process.
A main advantage of the basket sampler is its flexibility. Measurements can be undertaken at various verticals in the cross section. Depending on the measurement method (permanent measurement or cross-section wise measurement) specific bedload rates and bedload transport in the torrent profile can be determined. To achieve meaningful results it is required to work in a specific operational range. The range of collected bedload particles is restricted by the mesh size and the intake width. Measurements are restricted to low to medium flow stages as lowering the basket sampler to the stream bed is not possible at high flow velocities. Additionally the measuring procedure is endangered by woody debris.
The bedload trap is mounted at the same level to the stream bed directly upstream the geophone device (see Fig. 3b). It contains a sample box, which is placed on load cells. The sample box is covered by a lid with a longitudinal sampling slot.
The sampling slot is 1.6 m long and 0.5 m wide.
When starting with bedload sampling, the slot is opened hydraulically via manual control, allowing bedload material to be trapped in the sample box.
The load cells commence recording automatically the bedload mass increase in the sample box. The sample box with a bedload sample is shown in
Fig. 3 a: Mounting of the geophone sensor at the steel plate, b: Arrangement of the geophones at the measuring site
a) b) Geophone array
Fig. 4b. Maintenance work at the trap, where the measured bedload sample is removed and the empty box is reinserted, is restricted to low water conditions. For this work a dewatering construction was developed, which detains flow pressure from the trap, enabling the removal of the trap lid and the sample box.
Bedload traps enable measurements at all water stages. Bedload monitoring at high flow events can be performed. Habersack et al. (2001) presented that hydraulic and sampling efficiency are high.
Beside bedload rates, the determination of the bedload texture and the initiation of motion are possible. A disadvantage of the bedload trap is its fixed position in the stream bed and the high maintenance work.
3.3 Integrative Monitoring
The suitability of several bedload monitoring devices to measure selected parameters of interest is presented in Habersack et al. [2010]. It is not possible to monitor bedload transport process satisfactorily using a single measurement device, as each method has its specific advantages and restrictions [Habersack et al., 2010; Kreisler et al., 2010].
Geophone device, basket sampler and bedload trap form together the integrative bedload monitoring system. The combination of these methods takes advantage of the possibilities of the single monitoring methods and therefore compensates their specific restrictions in bedload monitoring. The adjustment of the single devices in the integrative monitoring system can be seen in Fig. 2.
4. BEDLOAD TRANSPORT FORMULAS For the comparison of measured bedload data and calculated bedload rates bedload transport equations were selected with an application range fitting to the Urslau torrent. Furthermore, formulae were preferred, which are often applied in practice at the Austrian Service for Torrent and Avalanche Control. In the following the applied formulae are
listed:
Formula of Rickenmann [1991]
% 20 4 , 0
1 )
( 1
.
3 1.5 1.5
2 . 0
30 90
I
s I q d q
qb d c
(1)
where qb is the volumetric transport rate per unit width, q is the specific discharge. qc is the critical flow discharge, I is the slope, s is the ratio between grain (ρs) and fluid (ρ) density, d90 and d30 are characteristic grain sizes, where 90% respectively 30% of the material by weight is finer. The critical flow discharge is calculated as following:
12 . 5 1 . 1 50 5 . 0 67 .
)1
1 ( 065 .
0
s g d I
qc (2)
where d50 is a characteristic grain size, where 50%
of the material by weight is finer and g is the acceleration due to gravity.
Additionally a flow resistance equation, which accounts for macro-roughness effects [Rickenmann and Recking, 2011] is used for the application of the equation of Rickenmann, [1991]. Nitsche et al.
[2011] combined bedload transport equations and several flow resistance equations, where macro-roughness is considered and compared the results to field measurements of 13 mountain streams. They stated that the approach of Rickenmann and Recking [2011] yields the best results.
Formula of Smart and Jäggi [1983]
I R
d I s
U d R
d g s
s m c
m s s
b
1 1 )
1 (
4 1.6
2 . 0
30
90
(3)
where gb is specific bedload rate, Rs is the hydraulic radius, Um is the average flow velocity, θc is the critical shear stress parameter after Shields and dm is the arithmetic mean diameter.
Formula of Meyer-Peter and Müller [1948]
2 / 2 3
/ 3
, 0.047( 1)
) 1 ( 8
s m
r So s st
b hI s d
Q Q k k s
g g (4)
where kst,So is the Strickler value of the stream bed, kr considers the grain roughness and is calculated as given in formula 5. Qs is the effective discharge, Q is the entire discharge in the profile and h is the water depth.
6 90
26
kr d (5)
Fig. 4 a: Mobile basket sampler b: Sample box with bedload sample
a) b)
For the application of equation 1-4, the particle sizes of the material of the stream bed were used.
5. RESULTS
Bedload transport monitoring is operated at the Urslau torrent since 2011. Several direct measurements with the mobile basket sampler and the bedload trap have been performed. Geophone data are available continuously since January 2011.
In the following some results of the bedload monitoring program are presented. Thereafter measured bedload rates at the Urslau torrent are compared to calculation results.
5.1 Monitoring results
5.1.1 Indirect monitoring methods
The application of geophones in the Urslau torrent produces high-resolution data of the bedload transport process. The temporal and spatial distribution of geophone impulses from 01.06.2012 to 31.08.2012 at the Urslau torrent is shown in Fig.
5. In the illustrated time period, various bedload peaks occur and most bedload transport is registered in the middle of the channel. We observed that in the entire measuring period at geophone #4 and #5 most bedload has been transported. The graph in Fig. 5 highlights the continuance of geophone records.
To determine the critical flow for the initiation of motion geophone data of the years 2011, 2012 and 2013 are compared to water discharge. Therefore, water discharge is divided into categories of 0.1 m3s-1 intervals and the average number of geophone impulses is calculated for the corresponding discharge categories. The diagram, which shows the discharge categories at the x-axes and the average geophone impulses at the y-axes, is given in Fig. 6. It shows that at the measuring site sediment material starts to move between discharges of 3-4 m3s-1.
As often cited in literature [Hubbell, 1964;
Habersack et al., 2001; Kreisler et al., 2012]
bedload transport reveals a strong temporal variability. This behavior also has been observed at the Urslau torrent. Fig. 7 shows the course over two days of registered geophone impulses and measured discharge. Although hydraulic conditions are relative stable, bedload transport intensity differs from zero to 3025 geophone impulses per minute.
5.1.2 Direct monitoring methods
An example of a cross-section wise measurement with the mobile basket sampler is given in Fig. 8.
The measured bedload rate and the portion of the fractions of material larger than 63 mm, from 22.5 to 63 mm and from 8 to 22.4 mm are illustrated. The highest specific bedload transport rate (2.4 kg m-1s-1) was measured in the middle of the torrent cross section. This result is consistent to the result in Fig. 5. The measured mean arithmetic diameter dm varies between 24.2 mm and 40.2 mm at this measurement.
0 500 1000 1500 2000 2500 3000
0 1 2 3 4 5 6 7 8 9 10
Geophone Impulses [15min average]
Discharge Categories [m3s-1] Initiation of Motion - Urslau
0 1000 2000 3000 4000
0 1 2 3 4
0:00:00 6:00:00 12:00:00 18:00:00 0:00:00 6:00:00 12:00:00 18:00:00 0:00:00 Sum of Geophonimpulses [Sum Imp min-1]
discharge [m3s-1]
Discharge and geophon impulses
Discharge Geophon impulses
Fig. 5 Spatial and temporal distribution of geophone impulses (01.06.2012-31.08.212)
Fig. 6 Initiation of Motion - Urslau
Fig. 7 Temporal variability of bedload transport 04.07.2012 05.07.2012
One selected example of a bedload trap measurement is given in Fig. 9. The course of mass increase of the sampled bedload material in the trap is illustrated. The cumulative curve of the registered geophone impulses is pictured too. Apparently these two curves show a similar course, indicating a good correlation of these two parameters. The average measured specific bedload discharge of this trap measurement is equal to 4.5 kg m-1s-1.
5.1.3 Calibration of geophone data
In a study calibration relations between geophone and measured bedload mass at various field sites have been compared [Rickenmann et al., 2013]. It revealed that at all field sites the sum of recorded geophone impulses increases approximately linear with bedload mass. The authors summarized that, the parameter “geophone impulses” was found to be a robust summary parameter.
The required calibration of the geophone signal at the Urslau torrent is performed by a correlation with the direct bedload transport measurements of the
mobile basket sampler and the bedload trap, which are undertaken directly downstream the geophone device.
Fig. 10 presents exemplarily the linear relation between the bedload mass, which in this case was measured with the bedload trap and the sum of geophone impulses in the corresponding time period. Particle sizes of the material, which was measured with the bedload trap, differed as well as water stages of the flow events. The correlation coefficient in Fig. 10 is equal to 0.87 and the kb factor, defined as the quotient of the sum of geophone impulses and the direct measured bedload mass [Rickenmann et al., 2013] is equal to 3.85.
Comparable values for kb factors were found for the Rofenache (kb = 3.87) and the Erlenbach (kb = 5.45) [Rickenmann et al., 2013].
5.2 Comparison of measured and calculated bedload rates
Bedload monitoring over a period of three years at the Urslau torrent showed that different types of flood events exist. Transported bedload mass is not depended on hydraulic conditions alone.
Availability of sediment material and the course of preceedings flood events affect the bedload transport process.
Fig. 11 and Fig. 12 give examples of different flood events at the Urslau torrent. In both diagrams the measured bedload rate is depicted in grey and measured discharge is illustrated with a blue area.
Additionally, the calculated bedload rates using the formulas of Rickenmann, [1991] (equation 1), Rickenmann, [1991] taking macro-roughness into account [Rickenmann & Recking, 2011], Smart and
0.1 2.4
0.8 24.2
40.2
27.4
0 10 20 30 40 50
0 1 2 3 4 5
1.5 3.5 5.5 Mean diameter d[mm] m
Bedload discharge [kgm-1s-1]
Distance to orographic left river bank [m]
Cross-section measurement (Basket Sampler - 09.08.2011)
> 63mm 22.5mm- 63mm 8mm- 22.4mm Bedload rate dm
0 10 20 30 40 50 60 70 80 90 100
22:48:00 22:49:00 22:50:00 22:51:00 22:52:00 22:53:00 22:54:00 22:55:00
Percentage of registered impulses Percentage of mass increase [%]
Zeit [hh:mm:ss]
Bedload Trap measurement - 01.06.2013
Cumulative curve of Impulses Increase of bedload mass
y = 3.85x R² = 0.87
0 10 20 30 40
0 5 10
Sum of geophone impulses [Impm-1s-1]
bedload mass [kgm-1s-1] Linear relation between
bedload mass and geophon impulses
Bedload trap measurements Linear (Bedload trap measurements) Fig. 8 Cross-section measurement with the basket sampler
(Urslau – 09.08.2011)
Fig. 9 Bedload trap measurement (Urslau – 01.06.2013)
Fig. 10 Linear relation between bedload mass (bedload trap measurement raw-data) and geophone impulses
Jäggi, [1963] (equation 3) and Meyer-Peter and Müller, [1984] (equation 4) are to be found in the various colors.
Peak values of water discharge are similar for both events. A maximum discharge of 27.2 m3s-1 has been measured at the June event and of 31.5 m3s-1 at the August event. The highest values of the measured bedload rates are as well comparable. At the June event the highest specific bedload transport rate is 20.9 kg m-1s-1. The maximum value of the August event is equal to 19.2 kg m-1s-1. The courses of discharge and bedload transport of the chosen events are different.
The event in August 2012 (Fig. 11) is characterized by a steady rising and falling limb.
Bedload transport starts when discharge begins to increase. The highest measured specific bedload transport rate amounts 19.2 kg m-1s-1. The calculated values for the highest transport rate vary between 17.1 kg m-1s-1 (formula of Meyer-Peter and Müller, 1984) and 45.5 kg m-1s-1 (formula of Rickenmann, 1991). The total calculated bedload yield, using the formula of Meyer-Peter and Müller [1984] is 1.1 times smaller than the measured bedload yield. The other equations, which have been applied, result in higher bedload yields than the measured data. Smart and Jäggi [1983] overestimate the measured value by a factor of 1.2. The result of Rickenmann [1991]
is 2.4 times higher. The result of the Rickenmann [1991] equation including the approach of Rickenmann and Recking [2011] exceeds the measured bedload yield with a factor of 1.9 at this flow event.
During the June 2012 event (Fig. 12) discharge and bedload transport rise abruptly. The average measured bedload transport rate at the rising limb is equal to 1.4 kg m-1s-1. After the flood peak is reached, discharge and bedload transport decrease, but bedload remains intense. In this period discharge is relative stable and measured specific bedload rates differ between 0.1 kg m-1s-1 and 14.3 kg m-1s-1. The highest measured bedload rate at this flow event amounts 20.9 kg m-1s-1. The results of the bedload transport formulas for the highest bedload transport rate are as following: Rickenmann [1991]
38.5 kg m-1s-1, Rickenmann [1991]/ Rickenmann and Recking [2011] 31.5 kg m-1s-1, Meyer-Peter and Müller [1984] 14.1 kg m-1s-1and Smart and Jäggi [1963] 19.5 kg m-1s-1.
Concerning the entire bedload yield of the event the outcomes of the equations of Meyer-Peter and Müller, Smart and Jäggi [1963] and Rickenmann [1991] / Rickenmann and Recking [2011] are below the measured data by factors of 7.8 (Meyer-Peter and Müller), 2.1 (Smart and Jäggi, 1983), and 1.5
(Rickenmann, 1991 / Rickenamann and Recking, 2011). The result of the Rickenmann [1991] formula exceeds the measured bedload yield with a factor of 1.8.
0 8 16 24 32 40
0 10 20 30 40 50
20:00:00 20:20:00 20:40:00 21:00:00 21:20:00 21:40:00 22:00:00 22:20:00 22:40:00 23:00:00 23:20:00 23:40:00 water dischsarge [m3s-1]
specific bedload rate [kgm-1s-1]
Calculated and measured bedload rate Event August
Discharge Measured Bedload Smart and Jäggi (1963) Meyer-Peter and Müller (1984) Rickenmann (1991)
Rickenmann (1991), Rickenmann and Recking (2011)
0 8 16 24 32 40
0 10 20 30 40 50
15:00:00 21:40:00 04:20:00 11:00:00 17:40:00 00:20:00 07:00:00 13:40:00 20:20:00 03:00:00 09:40:00 16:20:00 23:00:00 water dischsarge [m3s-1]
specific bedload rate [kgm-1s-1]
Calculated and measured bedload rate Event June
Discharge Measured Bedload Smart and Jäggi (1963) Meyer-Peter and Müller (1984) Rickenmann (1991)
Rickenmann (1991), Rickenmann and Recking (2011) Fig. 11 Calculated and measured bedload rate
(Event August 2012)
Fig. 12 Calculated and measured bedload rate (Event June 2012)
23.08.2012
22.06 23.06. 24.06 25.06 23.08.2012
6. DISCUSSION
The application of geophones at the Urslau produces high resolution data of the bedload transport process. The presented results showed that the bedload transport process reveals a strong temporal and spatial variability. Consideration on a small scale exhibited that bedload transport differs strongly, although hydraulic conditions are stable (Fig. 7). The analysis presented in Fig. 6 displayed, that a dependency of geophone data to water discharge exists, if mean monitoring data over a large time scale of three years are considered. This relationship allows the determination of the initiation of bedload transport. At the measuring site bedload particles start to move at a discharge of 3-4 m3s-1.
Direct bedload measurements with the mobile basket sampler have been undertaken at the Urslau torrent at various water stages and different measurement verticals in the cross-section. The highest measured transport rate was equal to 2.4 kgm-1s-1 at the presented measurement (Fig. 8).
At high flow stages and bedload transport rates lowering the basket sampler to the stream bed is getting difficult. Therefore bedload transport measurements are conducted with the bedload trap at high flow conditions. An example of a bedload trap measurement is given in Fig. 9. At this measurement the specific bedload rate was equal to 4.5 kgm-1s-1. Fig. 10 shows that measured bedload rates with the bedload trap vary between 0.8 and 8.6 kgm-1s-1.
Fig. 9 and Fig 10 reveal that measured bedload mass correlates with registered geophone impulses.
This reinforces the possibility to calibrate geophone data with direct measuring results and therefore the calculation of bedload rates in various time periods.
Referring to Fig. 10 a strong linear relation exists between the measured bedload mass and recorded geophone data. Studies show that the kb factor is influenced by the different grain size distributions [Rickenmann et al., 2013]. Rickenmann et al. [2013]
state that the kb factor presents a mean coefficient, which implies variable relative contributions form sediment particles of different grain sizes and further factors such as mode of particle movement, impact velocity and the location, where the sediment particle hits on the steel plate.
Bedload monitoring at Urslau torrent over a time period of three years showed that different types of flow events exist. Two events with comparable peak discharges and maximum bedload transport rates, but different courses and total bedload yields are presented. These two examples illustrate, that
prevalent bedload transport at a flow event in the Urslau torrent depends not only on the hydraulic conditions. Catchment related processes, like the availability of sediment material, and the course of preceeding flow events determine the actual bedload transport.
The measured bedload data of the selected flow events (Fig. 11 and Fig. 12) have been compared with calculation results of common bedload transport formulas. It has to be pointed out that in this paper the comparison of measured bedload data and calculation results focuses only on the downstream part of the Urslau torrent and the two selected flow events. Deviations between measured and calculated bedload rates may differ significantly for other flow events at the Urslau and for different stream types.
Considering the total bedload yield, the deviation between measured and calculated result of all applied bedload transport formulas differs for the selected events. The event in August is overestimated by all calculation results, except the equation of Meyer-Peter and Müller [1984]. This statement coincides with literature where calculated values of bedload transport formulas are higher than the measured bedload data [Bathurst, 1978; Nitsche et al., 2011; Rickenmann and Recking, 2011].
At the June event measured bedload transport remains intense after the event peak, although discharge decreases. Contrary to the August event, the total bedload yield of this event is underestimated by all calculations results, except the one of Rickenmann [1991]. The calculation results mainly underestimate bedload rates on the decreasing discharge limb, where high specifc bedload rates have been measured. This analyses highlights the challenge of applying and calibrating bedload transport formulas, as bedload transport process features a strong temporal and spatial variability and does not only depend on hydraulic conditions. Additionally, supply-limited conditions respectively transport-limited conditions effects the actual bedload transport rates.
The spectrum of calculated results varies between an underestimation of the measured data by a factor of 7.7 to an overrating of 3.4. Therefore, we want to conclude that results of the selected bedload transport equations provide a reliable range of the magnitude of bedload transport in the Urslau torrent and form in practice an important guidance value for engineers. Nevertheless, the comparison of measured and calculated results is important as it can give reliability to the engineers, and facilitates possibilities and necessities to optimize the application of the formulas.
7. CONCLUSIONS
Our experiences on the Urslau show that the application of several bedload monitoring methods is necessary, to get comprehensive results. The integrative monitoring system combines direct (mobile basket sampler and bedload trap) and indirect methods (geophones). It compensates individual shortcomings of the applied measuring devices by complementary devices.
Results of the spatial and temporal variability and initiations of motion are provided. The application of direct measurement methods enables the determination of the (specific) bedload transport at various water stages. Further, direct measurements allow the calibration of the geophone data.
The comparison of calculated and measured bedload data in this paper refers to the two selected flow events at the Urslau torrent. For these events the equations provide a valuable reference point for practitioners. The analysis of calculated and measured bedload data establishes possibilities for an improvement of bedload transport formulas for the application.
Further work will include a detailed analysis of the bedload transport formulas and input parameters. The predicted results and measured data of different flow events of the Urslau torrent will be examined.
ACKNOWLEDGMENTS: The financing of the project “Monitoring and calculation of bedload transport at downstream sections of mountain torrents at the case study Urslau” and the active support by the Austrian Service of Torrent and Avalanche Control at the Austrian Federal Ministry of Agriculture, Forestry, Environment and Water Management and the Department of the Torrent and Avalanche Control in Salzburg is gratefully acknowledged.
The financial support by the Austrian Federal Ministry of Economy, Family and Youth and the National Foundation for Research, Technology and Development is gratefully acknowledged.
The authors thank Hugo Seitz for his strong support at the development and installation of the monitoring station.
REFERENCES
Bathurst, J.C. (1978): Flow resistance of Large-Scale Roughness, Journal of Hydraulics Division, Vol. 104, No.
12, pp. 1587-1603.
Bunte, K., Abt, S.R. (2003): Sampler size and sampling time affected bed load transport rates and particle size measured with bed load traps in gravel-bed streams, Erosion and
Sediment Transport Measurements, Bogen J., Fergus T. and Walling D.E. (eds.), IAHS-Publication No. 283, pp 126-133.
Gomez, B.(2006): The potential rate of bed-load transport:
Proceedings of the National Academy of Sciences, v. 103, no. 46, p. 17170–17173.
Gray, J.R., Laronne, J.B., Marr, J.D.G. (eds). (2010):
Bedload-surrogate Monitoring Technologies, US Geological Survey Scientific Investigations Report 2010–5091. US Geological Survey: Reston, VA. http://pubs.usgs.
gov/sir/2010/5091/
Habersack, H.M., Nachtnebel, H.-P., Laronne, J.B. (2001): The continuous measurement of bedload discharge in a large alpine gravel bed river. Journal of Hydraulic Research, Vol.
39, No. 2, pp. 125-133.
Habersack, H.M., Laronne, J.B. (2002): Evaluation and improvement of bed load discharge formulas based on Helley-Smith sampling in an alpine gravel bed river. Journal of Hydraulic Engineering, Vol. 128, No. 5, pp. 484-499.
Habersack, H., Seitz, H., Liedermann, M. (2010): Integrative Automatic Bedload Transport Monitoring. International Bedload Surrogates Workshop, Minneapolis.
11.-14.04.2007. In: Gray, J.R., Laronne, J.B., and Marr, J.D.G. (Eds.), Bedload-surrogate monitoring technologies, U.S. Geological Survey SIR 2010-5091, 218-235.
Habersack, H., Kreisler, A., Aigner, J., Liedermann, M., Seitz, H. (2012): Spatio-temporal variability of bedload transport.
In: Murillo Munoz, R.E. (Ed.), River Flow 2012 Vol. 1, 423-430; Taylor & Francis, London; ISBN 978-0-415-62129-8
Haddadchi, A., Omid, M.H., Dehghani, A.A. (2013): Bedload equation analysis using bed load-material grain size. Journal of Hydrology and Hydromechanics, Vol. 61, No. 3. pp.
241-249. ISSN 0042-790X
Hubell, D.W. (1964): Apparatus and Techniques for Measuring Bedload, U.S. Geological Survey Water Supply Paper 1748, p.74.
Kreisler, A., Seitz, H., Habersack, H. (2010): Possibilities and restrictions for the deployment of basket samplers used for bedload transport measurements. Proceedings of 34th IAHR Congress 2010 Brisbane, pp. 3566-3573.
Kreisler, K., Moser, M., Seitz, H., Rudolf-Miklau, F., Habersack, H. (2012): Bedload trap measurements as part of an integrative measurement system. 12th Congress INTERPRAEVENT 2012 Grenoble / France – Extended Abstracts.
Lucia, A., Recking, A., Martin-Duque, JF., Storz-Peretz, Y., Laronne, JB. (2013): Continuous monitoring of bedload discharge in a small, steep sandy channel. Journal of Hydrology, Vol. 497. 37-50
Meyer-Peter, E., Müller, R. (1984): Formulas for bedload transport, Proceedings for 2nd meeting Int. Assoc. Hydraulic Structures Res., Stockholm, 39-64.
Mühlhofer, L. (1933): Untersuchungen über die Schwebstoff- und Geschiebeführung des Inn nächst Kirchbichl (Tirol) Die Wasserwirtschaft, 1-6:23.
Nitsche, M., Rickenmann, D., Turowski, J.M., Badoux, A., Kirchner, J.W. (2011): Evaluation of bedload transport
predictions using flow resistance equations to account for macro-roughness in steep mountain streams, Water Resour.
Res., 47, W08513, doi:10.1029/2011WR010645.
Reid, I., J. T. Layman, J.T., Frostick, L.E. (1980): The Continuous Measurement of Bedload Discharge, Journal of Hydraulic Research, Vol. 18, No. 3, pp. 243-249.
Rickenmann, D. (1991): Hyperconcentrated flow and sediment transport at steep flow. Journal of Hydraulic Engineering, Vol. 117, No. 11, pp. 1419-1439.
Rickenmann, D. (2001): Comparison of bed load transport in torrents and gravel bed streams, Water Resour. Res., Vol.
37, No. 12, pp. 3295–3305, doi:10.1029/2001WR000319.
Rickenmann, D. and McArdell, B.W. (2007): Continuous measurement of sediment transport in the Erlenbach stream using piezoelectric bedload impact sensors, Earth Surf.
Process. Landforms, Vol. 32, pp. 1362-1378.
doi: 10.1002/esp.1478.
Rickenmann, D. and Recking, A. (2011): Evaluation of flow resistance in gravel-bed streams through a large field data set, Water Resourc. Res., Vol. 47, W07538, doi:10.1029/2010WR009793.
Rickenmann, D., Turowski, J. M., Fritschi, B., Klaiber, A. and Ludwig, A. (2012): Bedload transport measurements at the Erlenbach stream with geophones and automated basket samplers. Earth Surf. Process. Landforms, Vol. 37, pp.
1000–1011. doi: 10.1002/esp.3225
Rickenmann, D., Turowski, J. M., Fritschi, B., Wyss, C., Laronne, J., Barzilai, R., Reid, I., Kreisler, A., Aigner, J., Seitz, H. and Habersack, H. (2013): Bedload transport measurements with impact plate geophones: comparison of sensor calibration in different gravel-bed streams. Earth Surf. Process. Landforms, Vol. 39, pp. 928–942.
doi: 10.1002/esp.3499
Rijn, L. C. van (1986): Manual sediment transport measurements. Delft, The Netherlands: Delft Hydraulics Laboratory-
Smart, G.M. and Jäggi, M.N.R. (1983): Sediment transport on steep slopes, pp. 89-191. Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie, Eidgenösische Technische Hochschule Zürich.
Tritthart, M., Liedermann, M., Klösch, M., Habersack, H.
(2012): Innovationen in der Modellierung von Sedimenttransport und Morphodynamik basierend auf dem Simulationsmodell iSed. Österreichische Wasser- und Abfallwirtschaft, 64, 544-552; ISSN 0945-358X
