Near-bankfull floods in an Alpine stream: effects on the

sediment mobility and bedload magnitude.Rainato R.1*, Mao L.2, Picco L.1

1 Department of Land, Environment, Agriculture and Forestry, University of Padova, Padova, Italy2 Department of Ecosystems and Environment, Pontificia Universidad Catòlica de Chile, Santiago, Chile

*Corresponding author:

Tel.: + 39 0498272695; fax: +39 0498272686; e-mail address: riccardo.rainato@unipd.it

Keywords: Bedload, Alpine basin, Sediment dynamics, Bedload tracing, PIT-tags.

ABSTRACT

In mountain environment, the transport of coarse material is a key factor for many application areas

(e.g. geomorphology, ecology, hazard assessment, reservoir management). Despite such role, only

few works have focused on the in-field investigation of bedload, in particular by using multiple

monitoring methods. In this sense, the attention has been frequently placed on the effects due to

high magnitude/low frequency flood, with less focus on the “ordinary” events. This study aims to

analyze the sediment dynamics triggered by three high-frequency floods (RI = 1.1 - 1.7 yr) occurred

in the Rio Cordon basin during the 2014. For this purpose, the flood events were investigated both

in terms of sediment entrainment and bedload magnitude. The Rio Cordon is an Alpine basin

located in the North-East of Italy. The catchment has a surface of 5 km2, with an altimetric range

between 1763 and 2763 m a.s.l. Here, the Rio Cordon creek flows on an armoured streambed layer,

with a stable step-pool configuration and large boulders. Since 1986, the basin is equipped with a

permanent monitoring station that records in continuous water discharge and sediment fluxes. To

investigate the sediment mobility, 250 PIT-tags were installed in the streambed in 2012. The floods

occurred in 2014 exhibited a clear difference in terms of tracers displacement. The near-bankfull

events showed equal mobility conditions, with mean travel distance one order of magnitude higher

respect to what experienced by the under-bankfull event, that produced just a local displacement in

the tracers. Notwithstanding the entrainment observed, only the near-bankfull events caused

transport of coarse material to the monitoring station. Both events peaked to 2.06 m 3 s-1 but the

bedload differs by more than one order of magnitude, proving that, under the current limited-supply

condition, in the Rio Cordon the bedload appears more related to the sediment supply than to the

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

magnitude of hydrological features. In literature, only few field dataset are available on which both

the bedload magnitude and the transport distance of tracers were investigated in mountain streams.

In this sense, the analysis performed in the Rio Cordon demonstrated that, if supplied by coarse

particles, near-bankfull events can mobilize for long distances large amount of material. Also, the

results showed how, in mountain streams, flood events characterized by apparently similar

magnitude may lead to sediment dynamics clearly different.

INTRODUCTION

Bedload transport in mountain streams strongly affects the downstream sediment delivery (Liébault

et al., 2016), channel stability (Baewert & Morche, 2014), and thus the assessment of hazard areas

along river corridors. Especially in the context of the EU water framework directive, an accurate

assessment of sediment transport is required for flood risk mapping and management. Also, from an

ecological point of view, in many mountain regions the spawning habitats of fish species, micro-

and macro-invertebrates appear strongly affected by bedload (Vazquez-Tarrio & Menendez-Duarte,

2014; Wohl, 2015). However, bedload is notoriously difficult to be measured in field. Overall, the

high-energy and impulsive nature that characterize bedload cause that its investigation and

assessment is a challenging task. Such an issue appears particularly evident in mountain streams,

where several factors make bedload processes differente than in lowland rivers. Additionally to the

high gradient, the entrainment is strongly influenced by the high heterogeneity in the streambed

material, which results in factors as grain sorting (Hammond et al., 1984), particle size interactions

and hiding-protrusion effects (Ashworth & Ferguson, 1989), low relative roughness (Bathurst et

al., 1983), presence of strong armour layer (Lenzi, 2004), embedding and exposed patches

(Bathurst, 2013), and slope (Lamb et al, 2008). In addition to the magnitude of flood event (Lenzi et

al., 2006a), the bedload transport rate is strongly related to sediment supply condition (Recking,

2012), and hillslopes-channel coupling (Cavalli et al., 2013). These complex conditions are

reflected in the poor performance of bedload predictive equations, which are usually derived from

laboratory experiments or specific field sites (Yager et al., 2015). Also, the availability of field data

appears quite scarce, with a lack of monitoring programs maintained in the same study site over

long-term periods. Currently, the employment of several direct and indirect monitoring methods

may enable to obtain precious field data about bedload (Mao et al., 2015). Collecting the sediment

transported over a certain interval, bedload traps allow the rate and grain size of coarse material

mobilized to be analyzed (Bunte et al., 2008). Such devices can be installed in permanent

monitoring stations, enabling analysis over long time scales (Rainato et al., 2016), or can be used as

moving traps, focusing on short time intervals (Mao et al., 2008).

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

The tracers method consists in individual particles that are collected, dried, painted and replaced

into the channel (Fraley, 2004). Single-grain tracers may enable to investigate sediment travel

distances (Olinde & Johnson, 2015), virtual velocity (Houbrechts et al., 2015), bedload transport

rates (Dell’Agnese et al., 2015), threshold conditions (Lenzi et al., 2006b), estimate bedload

volumes during flood events (Liébault & Laronne, 2008; Schneider et al., 2014), allowing to

integrate the information achievable by the traps (Ferguson & Wathen, 1998). Recently, more

sophisticated methods to mark the grains were developed. The application of the Radiofrequency

Identification technology (RFID) to the sediment tracing allowed to detect the tracers once buried,

increasing the recovery rates. In particular, to achieve a continuous tracing, the particles can be

embedded with a Passive Integrated Transponders (PIT) programmed with a unique identification

code. In terms of investigation, the PIT-tags are small, not too expensive and, potentially, allow

long-lasting monitoring. The information achievable by these tracers can be highly useful since in

mountain channels, during bedload events triggered by rainstorms, the transport rate seems to be

function of width and depth of bed scouring, as well as of travel distances of the sediment particles

(Schneider et al., 2014).

Here we present the results obtained during the 2014 monitoring season, performed in the Rio

Cordon, a small instrumented basin located in eastern Italian Alps (Dolomites). Three high-

frequency bedload events occurred in May, June and November, respectively. These events were

investigated both in terms of bedload magnitude (i.e. coarse material captured by the monitoring

station) and analyzing the displacements triggered on a population of 250 PITs installed along the

streambed. Despite the relative low magnitude of flood events, the complex features that occurred

in terms of sediment supply and hydraulic forcing enabled to observe and analyze clearly different

sediment dynamics.

STUDY AREA AND METHODS

Rio Cordon study site

The Rio Cordon basin (Dolomites, NE Italy) drains a surface of 5 km2, ranging among 1763 to 2763

m a.s.l. (Fig. 1). In the watershed prevail Alpine climatic conditions, exhibiting a prevalent nivo-

pluvial runoff regime. The average annual precipitation is equal to 1150 mm. Throughout the basin,

quaternary moraines and scree deposits are very common, but are mainly decoupled from the

drainage network. In terms of land use, the major part of the catchment consists in Alpine

grasslands (61%) and shrubs (18%). Barely, 7% of the area is forested, while 14% is bare land.

Talus slopes, shallow landslides, eroded stream banks and debris flow channels are the main

sediment source areas, covering the 5.2% of the basin (Lenzi et al., 2003). Due to distance and the

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

decoupling of such sources, normally the drainage network has a low/moderate sediment supply.

The Rio Cordon creek exhibits an average slope equal to 17%, showing a rough streambed with

step-pool configuration and large boulders. The grain size distribution (GSD) of the streambed

surface is characterized by D16 =29 mm, D50=114 mm, and D84=358 mm. Overall, the creek

highlights a well-developed armour layer (D50/D50ss = 3). The bankfull discharge was estimated

equal to 2.30 m3 s-1 (Lenzi et al., 2006a).

####### Figure 1 #######

Since 1986, a permanent monitoring station records in continuous water discharge, bedload and

suspended load of the Rio Cordon stream. The station mainly consists of an inlet flume, an inclined

grid, a storage area for bedload material, an outlet flume and a settling basin for the suspended load

material. The water discharge is hourly measured by two water level gauges and a sharp-crested

weir. In case of flood events, the sampling interval increases to 5 minute. The inclined grid

(longitudinal slope equal to 60%) enables the coarse sediment (> 20 mm) to be separated from

water and fine material. Once separated, the coarse material glides in the storage area. Also, in the

study basin 2 meterological stations are located at 1763 and 2130 m a.s.l., respectively, allowing to

continuously record air temperature, atmospheric pressure, relative humidity, solar radiation and rainfall

(hourly). Currently, the monitoring station is managed by ARPA Veneto, Regional Department for

Land Safety.

Methods

Data collected at the monitoring station were used to analyze the flood events in terms of bedload

magnitude.. The 5-minutes interval data of discharge were used to describe the hydrological

features of the floods occurred during the analyzed period, i.e. hydrograph, peak discharge (QPEAK)

and duration of the events. In case of bedload event, also the effective runoff (ER, 103 m3) was

estimated. Here, the effective runoff is defined as the portion of hydrograph volume that contributes

to the transport, exceeding the detected threshold discharges. Additionally, the data gathered by the

metereological stations were used to define the antecedent climatic conditions, in particular the

cumulative rainfalls occurred during the 24 (R24) and 48 hours (R48) pre-event. The amount of

bedload transported to the monitoring station (BL) was estimated by surveying the bedload storage

area after floods using a Terrestrial Laser Scanner (TLS). By scanning the coarse material deposited

in the storage area, the Digital Elevation Models (DEM) of bedload volumes were produced (cell

size = 0.02 m). Also, the grain size distribution of the coarse material was evaluated, using the grid

by number approach as sample method.

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

Between 2009 and 2012, 250 PIT-tags were seeded on the Rio Cordon streambed to investigate

their sediment mobility conditions. In terms of grain size, the particles embedded with PIT (b-axis =

40-190 mm) were selected in order to describe a representative sample of channel bed. Specifically,

the GSD of the tracers matches to D25 < D < D70 of the streambed surface. The PIT-tags were

installed along several cross sections, starting from a cascade/step-pool segment located 318 m

upstream to the monitoring station. To monitor the tracer displacements, an Aquartis Accueil®

mobile antenna in combination with a laser rangefinder were used. Due to the difficult conditions in

which the PIT surveys wereperformed (high-gradient stream), the exact positioning of the tracers

could be measured within a certain degree of uncertainty. For this reason, only displacement > 1 m

were considered in this analysis. Hereafter, the terms “PIT-tags” or “tracers” will be used to identify

the particles equipped with passive integrated transponders. Additionally, the terms “survey” and

“inventory” will be use to describe the field-phase in which the PITs were monitored.

RESULTS

In terms of climatic conditions, the studuy period (2014) was characterized by significant snowfalls

during the first months of the year that led to an extended snowmelt period (i.e. early April- late

June). Typical Alpine climatic conditions were observed in the following months, with frequent

rainstorms in summer and persistent rainfalls during the autumn (Fig. 2). In this sense, the highest

daily precipitation of the year was recorded on November, 5 with 126.2 mm d-1 (Fig. 2).

####### Figure 2 #######

Due to these climatic conditions, in the Rio Cordon three high frequency flood events were

observed during the 2014. First, an under-bankfull event occurred during the first phase of

snowmelt period, on May, 11 with a QPEAK = 1.00 m3 s-1 (recurrence interval, RI = 1.1 years). The

analysis of the antecedent rainfall showed that the runoff due to the precipitation was limited, with

R24 and R48 equal to 17.8 and 19 mm, respectively. Subsequently, two near-bankfull floods occurred

on June, 9 and November, 5, respectively (Tab. 1). In both cases, the discharge peaked at 2.06 m3 s-1

with a recurrence interval equal to 1.7 years, but in terms of runoff patterns the events appear quite

different. The event occurred on June, 9, can be defined as a mixed snowmelt-rainfall event. In fact,

due to the abundant snowfalls occurred in winter, the snowmelt period in 2014 lasted up to June.

The moderate rainfall recorded on June, 9 (R24 = 16.6 mm) were concentrated in the afternoon hours,

thus, combining with the runoff provided by the intense snowmelt. On the other hand, the flood

occurred on November, 5 was a typical autumn-event, which was triggered by heavy and persistent

rainfalls as stressed by R24 and R48 equal to 126.20 and 164.00 mm, respectively (Tab. 1). To

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

investigate the entrainment triggered by such events, post-flood PIT surveys were performed (Fig.

2).

####### Table 1 #######

Significant recovery rates (i.e. tracers detected on the total population) were achieved during the

three field surveys, in fact the percentages range from 70% in the inventory carried out in

November to 86% reached during the May 2014 survey (Tab. 2).

####### Table 2 #######

Among the events under investigation, a clear difference can be noted both in terms of tracers

mobilized (nm, displacement > 1 m) and travel distance. The number of PIT-tags entrained was 101

by June flood, 81 by November event, and 27 in May. As to the displacement, the mean travel

distance (Li) triggered by the near-bankfull events, i.e. June and November, are equal to 117.03 m

and 95.18 m, respectively. The mean travel distance decreased of more than one order of magnitude

in the under-bankfull event (i.e. May), reaching 2.48 m (Tab. 2). To investigate the effects triggered

by the magnitude of the event on the grain size/displacement relationship (Di/Li), the tracers

mobilized were grouped based on the flood experienced. Once grouped, the tracers were

reclassified according to their b-axis, using the ϕ size classes equal to 45.3 mm, 64 mm, 90.5 mm,

128 mm and 181 mm, and then the mean travel distance was estimated for each ϕ size class. The

Figure 3 shows the results of the Di/Li relationship, stressing out the different magnitude of

displacement triggered.

####### Figure 3 #######

The under-bankfull event (i.e. May 2014) showed, in all grain size classes, Li constantly lower of

one order of magnitude compared to those observed in the near-bankfull floods (i.e. June 2014 and

November 2014). Moreover, the flood occurred in May caused exclusively the entrainment of the

finer fraction of tracers, with no motion of D > 128 mm. On the other hand, it is worth noticing that

equal mobility conditions seem to have occurred during the events in June and November, with

mobilization even of the larger tracers (Fig. 3). In terms of Di/Li relationship, the near-bankfull

events showed a quite comparable behaviour. In detail, such similarity can be even better observed

comparing the travel distance experienced by the grain size classes (Fig. 4). Focusing on the two

near-bankfull events, similar ranges of travel distance have been observed in particular in the

classes 45.3 mm, 64 mm and 181 mm, while displacements partly different are observable in the

class 90.5 mm and 128 mm, respectively.

####### Figure 4 #######

Even if in all monitored events there was some grain entrainment, only the near bank-full events

occurred in June and November caused the transport of coarse material to the monitoring station,

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

while no bedload flux was observed during the May 2014 flood. During the first bedload event, the

water discharge peaked on June, 9, at 2:40 PM with 2.06 m3 s-1. The bedload lasted for 14 hours

between the 2:10 PM on June, 9, to 4:10 AM on June, 10, beginning and ending with discharge

equal to approximately 1.4 m3 s-1. In this sense, the effective runoff is 16.6 103 m3, with a bedload

yield equal to 65.6 m3 (Fig. 5A).

####### Figure 5 #######

Due to the considerable amount of transported sediments, also the bedload transport rate (BLr)

reached a significant value, equal to 4.7 m3 h-1 (Tab. 1). The largest particle detected in the storage

was characterized by b-axis = 230 mm. Mainly, the material consisted in coarse gravel (pebble)

with some small cobble. Collecting 262 particles, the GSD percentiles were estimated as: D16 equal

to 16 mm, D50 was 41 mm, D84 was equal to 64 mm while D90 is 76 mm (Fig. 6).

####### Figure 6 #######

A debris flow channel located in the median part of the basin, was identified as source area. The

field evidences (i.e. traces of moved loam) suggested the occurrence of a debris flow in this area

(Fig. 7).

####### Figure 7 #######

The second bedload peaked on November, 5, at 11:55 PM with a QPEAK = 2.06 m3 s-1. The bedload

started with 1.6 m3 s-1 (November, 5 – 7:25 PM) and ended to 1.8 m3 s-1 (November, 7 – 1:10 AM),

transporting 2.7 m3 to the monitoring station (Tab. 1). Overall, the bedload lasted for approximately

30 hours with a bedload rate equal to 0.1 m3 h-1, while ER is 33.3 m3 103 (Fig. 5B). Likewise to the

June bedload event, the material accumulated in the storage area was characterized in terms of

GSD. Overall, 174 particles were collected and measured by grid by number method. The D16, D50,

D84, and D90 percentiles were, respectively, 25, 38, 62, and 73 mm (Fig. 6). During the post-flood

survey no active source area was detected.

DISCUSSIONS

During the 2014 field campaign performed in the Rio Cordon, three flood events characterized by

different magnitude were investigated. In terms of QPEAK the events range between 1 to 2.06 m3 s-1,

corresponding to RI = 1.1 (under-bankfull) to RI = 1.7 (near-bankfull). The results achieved here by

the bedload tracing highlighted a clear difference in the entrainment behaviour among the events

monitored. First, the mean travel distance increases of more than one order of magnitude between

the under- to near-bankfull flood (Fig. 3), suggesting that the travel distance is clearly related to the

event-magnitude. Focusing on the near-bankfull events (both QPEAK=2.06 m3 s-1) a substantial

similarity in terms of displacements can be observed comparing the travel distance exhibited by the

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

grain size classes (Fig. 4). In this sense, QPEAK appears to better describe the travel distance/event-

magnitude relationship respect to other hydrological features (i.e. effective runoff). Regarding the

grain size tracked, only a fraction of tracers (D ≤ 128 mm) was mobilized by the under-bankfull

event. Moreover, the entrainment appears lower both in terms of tracer mobilized and travel

distance per grain size class (Fig. 3) respect to the near-bankfull events. During these floods, the

Di/Li relationship reveals that tracers experienced equal mobility conditions, in which the

entrainment appears unaffected by size of particles. Furthermore, the entire range of grain size

tracked was mobilized during the near-bankfull events. Analyzing a range of QPEAK between 0.85 to

10.42 m3 s-1, and using colored particle and radiotracers to trace the bedload, Mao & Lenzi (2007)

observed equal mobility conditions in the Rio Cordon only during floods with RI > 5, while size-

selective transport prevailed during lower magnitude events. Notwithstanding the lower flood-

magnitude, a similar behavior was observed even during the floods here discussed. As expected, the

increase of flood-magnitude led the change from size-selective transport to equal mobility

conditions. Interestingly, in this work the equal mobility conditions appear to be induced by

hydraulic forcing largely lower respect to those observed previously in the Rio Cordon. The finer

GSD currently tracked and the different bedload tracing method could explain such result. If

compared with the present work, Mao & Lenzi (2007) carried out their research about sediment

mobility marking a larger range of particle size. The authors used tracers characterized by a

diameter-range between 32 and 512 mm, while in this study the range of b-axis is 40-190 mm.

Nevertheless, the focus on a finer fraction enabled here to clearly observe that near-bankfull events

could trigger equal mobility up to the large cobbles. Additionally, the PIT-tags enabled to enhance

the quality and quantity of the data collected in field about the sediment mobility, in particular

compared to the colored particles and radiotracers. These tracers, on the one hand, allowed to

clearly detect even the local displacements and, on the other hand, have permitted to identify also

the buried PIT-tags, thereby increasing the recovery rates (Rr = 70-86%). Overall, further analysis

are required to better understand the entrainment behavior in the Rio Cordon creek, but despite the

short-term investigation, the results demonstrated that large part of streambed material can be

mobilized by near-bankfull events, causing also significant displacements (Li > 100 m). Further

analysis could be focused on the mobility of the coarser fraction of streambed material (i.e.

boulders) and to analyze the influence of the high degree of streambed armouring and bedforms on

the particles mobility.

In literature, only few works have analyzed flood events in steep streams relying on displacement of

tracers and bedload magnitude (e.g. Lenzi, 2004; Schneider et al., 2014). Here, despite all

monitored events triggered entrainment in the tracers, only the near-bankfull floods induced

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

bedload to the monitoring station. Such events appear similar, both in terms of QPEAK and Li, while

the bedload magnitude (BL) differs by one order of magnitude. Specifically, the June event

transported 65.6 m3, while barely 2.7 m3 were transported during the November flood. Additionally

to QPEAK, even ER appears to be a poor descriptor for bedload magnitude, exhibiting a negative

correlation. In fact, the volume of hydrograph that contributed to the transport in November (ER=

33.3 103 m3) was roughly two-fold higher respect to what observed in June (ER = 16.6 103 m3).

Indeed, the bedload magnitude appears better related to the sediment supply than to the

hydrological features (i.e. QPEAK and ER). In this sense, the June bedload event was highly supplied

by a large debris flow occurred in the middle part of the basin (Fig. 7), while none field evidence

was detected subsequently to the November event, suggesting that the coarse material may have

been provided by minor bank erosion or by loose material provided locally by hillslope collapses

and mobilized by the increased water stage. In terms of grain size characteristics, the bedload

transported to the monitoring station by the two near-bankfull floods appears clearly finer respect to

the streambed material (Fig. 6). Such result seems to be consistent with the concept of “traveling

bedload” (Yu et al., 2009), that typically occurs in the mountain paved creeks when supplied by

hillslopes processes. A local injection of sediment triggers bedload, but only the finer fraction is

transported, not interacting with the armour layer and bedforms, while large part of coarser

sediments is deposited on the channel bed (Schuerch et al., 2006). Notwithstanding this process is

particularly evident in the Rio Cordon, it is worth to noticing that also large cobbles (b-axis = 230

mm), larger of the tracers installed and in line with the D75 (256 mm) of streambed surface, were

transported up to the monitoring station by the June 2014 event. The results achieved by the

investigation of the most recent bedload events stress the hypothesis that in the Rio Cordon the

high-frequency floods may exhibit high bedload transport rate only if supplied by gravitation

processes (i.e. mud flow, debris flow), or rather if coupled with an active source area (Recking,

2012). Under the current limited-supply condition with a highly paved streambed, the traveling

bedload appears the main transport process in the Rio Cordon creek.

CONCLUSIONS

In the 2014, three high-frequency flood events (i.e. under- to near-bankfull event, RI = 1.1 - 1.7)

were investigated both in terms of sediment mobility and bedload magnitude. As regards the

bedload tracing, a clear difference was observed both in terms of number of tracers mobilized and

mean travel distance. Unlike to what observed in the higher magnitude floods, during the under-

bankfull event only the fraction D < 128 mm of tracers have experienced entrainment. In terms of

mean displacement, the two near-bankfull events showed a substantial similarity with average travel

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

distance one order of magnitude higher respect to what exhibited by the under-bankfull event.

Additionally, significant evidences of equal mobility conditions were observed as consequence of

the near-bankfull events, when the entrainment appeared unaffected by the particle size. Compared

to the other hydrological features, QPEAK proved to be the best descriptor for the mean displacement.

The results obtained using the PIT-tags highlighted how near-bankfull event may strongly influence

the sediment entrainment, mobilizing for long distances large part of streambed material.

Notwithstanding the tracers entrainment observed, only the near-bankfull events caused bedload

transport to the Rio Cordon monitoring station. In terms of QPEAK such events are fully comparable

(either QPEAK = 2.06 m3 s-1), while the bedload differs by more than one order of magnitude (i.e. 65.6

vs. 2.7 m3). Thus, the results suggest that in the Rio Cordon, under the current limited-supply

condition and strong armoured layer, the hydrological features of the event (i.e. QPEAK and ER) are

not the most relevant descriptors as regards the bedload magnitude. In the light of existing condition

and in absence of an exceptional flood able to altered the sediment dynamics, the high frequency

events appear to have significant bedload only if coupled and supplied by an active source area. In

the recent years the hillslope collapses were the main active sources, stressing out such hypothesis.

This work demonstrated also that, once supplied by coarse particles, the Rio Cordon creek can

easily mobilize large amount of material. In this sense, the mobility of large clasts (i.e. boulders) as

well as the influence of paved streambed and bedforms on the entrainment are issues to further

explore. Overall, only few field dataset are available in literature on which both the bedload

magnitude and the transport distance of tracers were investigated in steep streams. Notwithstanding

the short time scale, this work it is an attempt in this direction, focusing particularly in the

comparison of flood events characterized by apparently similar magnitude but that experienced

clearly different sediment dynamics.

ACKNOWLEDGMENTS

This research was supported by the Italian Research Project of Relevant Interest PRIN2010–2011,

prot. 20104ALME4; ITSE: National network for monitoring, modeling, and sustainable

management of erosion processes in agricultural land and hilly-mountainous area and by the

University of Padova Research Project CPDA149091- Woodalp.

REFERECENCES

Ashworth, P.J. & Ferguson, R.I. (1989). Size-selective entrainment of bed load in gravel bed

streams. Water Resources Research 25 (4), 627-634.

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

Baewert, H., Morche, D. (2014). Coarse sediment dynamics in a proglacial fluvial system (Fagge

River, Tyrol). Geomorphology 218, 88–97.

Bathurst, J.C., Graf, W.H., Cao, H.H. (1983). Initiation of sediment transport in steep channels with

coarse bed material. Mechanics of sediment transport. Proc.Euromech 156. Istanbul, July 1982,

207-213.

Bathurst, J.C. (2013). Critical conditions for particle motion in coarse bed materials of nonuniform

size distribution. Geomorphology 197, 170–184.

Bunte, K., Abt, S.R., Potyondy, J.P., Swingle, K.W. (2008). A comparison of coarse bedload

transport measured with bedload traps and Helley–Smith samplers. Geodinamica Acta 21 (1/2), 53–

66.

Cavalli, M., Trevisani, S., Comiti, F., Marchi, L. (2013). Geomorphometric assessment of spatial

sediment connectivity in small Alpine catchments. Geomorphology 188, 31-41.

Dell’Agnese, A., Brardinoni, F., Toro, M., Mao, L., Engel, M., Comiti, F. (2015). Tracing bedload

transport in a high-elevation, formerly-glaciated mountain basin. Earth Surface Dynamics 3, 527-

542.

Ferguson, R.I. & Wathen, S.J. (1998). Tracer-pebble movement along a concave river profile:

Virtual velocity in relation to grain size and shear stress, Water Resources Research 34 (8), 2031–

2038.

Fraley, L. (2004). Methods of measuring fluvial sediment. Center for Urban Environment Research

and Education. University of Maryland, Baltimore, 2004.

Hammond, F.D.C., Heathershaw, A.D., Langhorne, D.N. (1984). A comparison between Shields'

threshold criterion and the movement of loosely packed gravel in a tidal channel. Sedimentology 31

(1), 51-62.

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

Houbrechts, G., Levecq, Y., Peeters, A., Hallot, E., Van Campenhout, J., Denis, A., Petit, F. (2015).

Evaluation of long-term bedload virtual velocity in gravel-bed rivers (Ardenne, Belgium).

Geomorphology 251, 6-19.

Lamb, M.P., Dietrich, W.E., Venditti, J.G. (2008). Is the critical Shields stress for incipient

sediment motion dependent on channel-slope? Journal of Geophysical Research: Earth

Surface 113 (2). F02008.

Lenzi, M.A. (2004). Displacement and transport of marked pebbles, cobbles and boulders during

floods in a steep mountain stream. Hydrological Processes 18(10), 1899-1914.

Lenzi, M.A., Mao, L., Comiti, F. (2003). Interannual variation of suspended sediment load and

sediment yield in an alpine catchment. Hydrological Sciences Journal 48(6), 899-915.

Lenzi, M.A., Mao, L., Comiti, F. (2006a). Effective discharge for sediment transport in a mountain

river: Computational approaches and geomorphic effectiveness. Journal of Hydrology 326 (1-4),

257-276.

Lenzi, M.A., Mao, L., Comiti, F., (2006b). When does bedload transport begin in steep boulder-bed

streams? Hydrological Processes 20, 3517-3533.

Liébault, F. & Laronne, J. B. (2008). Evaluation of bedload yield in gravel-bed rivers using scour

chains and painted tracers: The case of the Esconavette Torrent (Southern French Prealps),

Geodinamica Acta 21 (1), 23–34.

Liébault, F., Jantzi, H., Klotz, S., Laronne, J.B., Recking, A. (2016). Bedload monitoring under

conditions of ultra-high suspended sediment concentrations. Journal of Hydrology 540, 947–958.

Mao, L. & Lenzi, M.A. (2007). Sediment mobility and bedload transport conditions in an alpine

stream. Hydrologicl Processes 21, 1882–1891.

Mao, L., Uyttendaele, G.P., Iroumé, A., Lenzi M.A. (2008). Field based analysis of sediment

entrainment in two high gradient streams located in Alpine and Andine environments.

Geomorphology 93, 368-383.

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

Mao, L., Carrillo, R., Escauriaza, C., Iroume A. (2015). Flume and field-based calibration of

surrogate sensors for monitoring bedload transport. Geomorphology 253, 10-21.

Olinde, L. & Johnson, J.P.L. (2015). Using RFID and accelerometer-embedded tracers to measure

probabilities of bed load transport, step lengths, and rest times in a mountain stream. Water

Resources Research, 51, 7572-7589.

Rainato, R., Mao, L., Garcia-Rama, A., Picco, L., Cesca, M., Vianello, A., Preciso, E., Scussel,

G.R., Lenzi, M.A. (2016). Three decades of monitoring in the Rio Cordon instrumented basin:

sediment budget and temporal trend of sediment yield. Geomorphology. DOI:

10.1016/j.geomorph.2016.03.012.

Recking, A. (2012). Influence of sediment supply on mountain streams bedload transport.

Geomorphology 175-176, 139-150.

Schneider, J.M., Turowski, J.M., Rickenmann, D., Hegglin, R., Arrigo, S., Mao, L., Kirchner, J.W.

(2014). Scaling relationships between bed load volumes, transport distances, and stream power in

steep mountain channels. Journal of Geophysical Research: Earth Surface 119, 533-549.

Schuerch, P., Densmore, A.L., McArdell, B.W., Molnar, P. (2006). The influence of landsliding on

sediment supply and channel change in a steep mountain catchment. Geomorphology 78, 222–235.

Vazquez-Tarrio, D. & Menendez-Duarte, R., (2014). Bedload transport rates for coarse-bed streams

in an Atlantic region (Narcea River, NW Iberian Peninsula). Geomorphology 217, 1–14.

Wohl, E. (2015). Particle dynamics: the continuum of bedrock to alluvial river segments.

Geomorphology 241, 192–208,

Yager, E.M., Kenworthy, M., Monsalve, A. (2015). Taking the river inside: Fundamental advances

from laboratory experiments in measuring and understanding bedload transport processes.

Geomorphology 244, 21-32.

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

Yu G., Wang Z., Zhang K., Chang T., Liu H. (2009). Effect of incoming sediment on the transport

rate of bed load in mountain streams. International Journal of Sediment Research 24, 260–273.

438

439

440

441

442

Figure 1: The Rio Cordon Study site.

Figure 2: Evolution of rainfall and discharge during 2014. The grey dashed lines correspond to the PIT

surveys.

443444

445

446447

448

449

Figure 3: Relationship between mean transport distances and particle size classes, for the three events

monitored.

Figure 4: Boxplot concerning the distribution of travel distances across grain size classes, for the near-

bankfull events.

450451

452

453

454455

456

457

Figure 5: Discharge time series during June 2014 (A) and November 2014 (B) bedload events. The red line

indicates the Effective Runoff (ER).

Figure 6: GSD of streambed material compared to the bedload transported by the June 2014 and November

2014 events.

458459

460

461

462463

464

Fig. 7: Hillslope collapse that supplied the June 2014 bedload event, in figure B the stream flows from top to

the bottom.

465466

467

468

469

Table 1: Main characteristics of floods monitored: R24 and R48 are the cumulative rainfall occurred during the

24 and 48 hours pre-event, respectively, QPEAK is the peak of water discharge during the flood event, RI the

recurrence interval, ER the effective runoff, BL and BLr are the amount of bedload and the transport rate,

respectively, while D50 is the 50th percentile of grain size transported.

FloodR24 R48 QPEAK RI ER BL BLr D50

(mm) (mm) (m3 s-1) (years) (103 m3) (m3) (m3 h-1) (mm)

May 2014 17.80 19.00 1.00 1.1 - - - -

June 2014 16.60 16.60 2.06 1.7 16.6 65.6 4.7 41

November 2014 126.20 164.20 2.06 1.7 33.3 2.7 0.1 38

Table 2: Main characteristics of PIT surveys: QPEAK occurred during the period monitored, Li is the mean

travel distance, n and nm are the number of tracers detected and mobilized, respectively, while Rr is the

recovery rate.

SurveyQPEAK Li n nm Rr

(m3 s-1) (m) (n) (n) (%)

May 2014 1.00 2.48 215 27 86

June 2014 2.06 117.03 192 101 77

November 2014 2.06 95.18 166 81 70

470

471

472

473

474

475

476

477

478

479