Large wood mobility processes in low order Chilean river channels
Andrés Irouméa,*, Luca Maob, Andrea Andreolic, Héctor Ulload, María Paz Ardilese
aUniversidad Austral de Chile, Faculty of Forest Sciences and Natural Resources, Valdivia,
Chile
bPontificia Universidad Católica de Chile, Department of Ecosystems and Environments,
Santiago, Chile
cUniversidad de Concepción, Department of Forestry and Environmental Management,
Concepción, Chile
dUniversidad Austral de Chile, Graduate School, Faculty of Forest Sciences and Natural
Resources, Valdivia, Chile
eUniversidad Austral de Chile, School of Civil Engineering, Faculty of Engineering
Sciences, Valdivia, Chile
* Corresponding author: Andrés Iroumé, Universidad Austral de Chile, Faculty of Forest
Sciences and Natural Resources, Independencia 631, 5110566 Valdivia, Chile. Tel +56-63-
2293003, [email protected]
Key points
LW mobility was similar between periods when flows exceeded bankfull stage and
periods where they did not
LW mobility correlates with unit stream power
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
123
Study in Chilean rivers on LW entrainment and transport
Abstract
Large wood (LW) mobility was studied over several time periods in channel segments of
four low-order mountain streams, southern Chile. All wood pieces found within the
bankfull channels and on the streambanks extending into the channel with dimensions more
than 10 cm in diameter and 1 m in length were measured and their position was referenced.
Thirty six percent of measured wood pieces were tagged to investigate log mobility. All
segments were first surveyed in summer and then after consecutive rainy winter periods.
Annual LW mobility ranged between 0 and 28%. Eighty four percent of the moved LW had
diameters ≤ 40 cm and the 92% lengths ≤ 7 m. LW mobility was higher in periods when
maximum water level (Hmax) exceeded channel bankfull depth (HBk) than in periods with
flows less than HBk but the difference was not statistically significant. Dimensions of
moved LW showed no significant differences between periods with flows exceeding and
with flows less than bankfull stage. Statistically significant relationships were found
between annual LW mobility (%) and unit stream power (for Hmax) and Hmax/HBk. The mean
diameter of transported wood pieces per period was significantly correlated with unit
stream power for H15% and H50% (the level above which the flow remains for 15 and 50% of
the time, respectively). These results contribute to understand the complexity of LW
mobilization processes in mountain streams, and can be used to assess and prevent potential
damage due to LW mobilization during floods and to consider wood transport along with
peak discharge in LW dominated Chilean mountain channels.
Keywords: Large wood; Log mobility; Bankfull stage; Mountain channels
2
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
456
1. Introduction
The significance and effects of in-stream large wood (LW) on the morphology and ecology
of stream ecosystems are widely recognized, especially on steep and narrow mountain
streams (Bilby and Ward, 1989; Robison and Beschta, 1990; Beechie and Sibley, 1997;
May and Gresswell, 2003; Chen et al., 2008). In-stream large wood, either isolated or
jammed, has the potential to alter local flow by reducing stream power and increasing flow
resistance, and creating log-steps, trapping sediments and dissipating hydraulic potential
energy (Beschta and Platts, 1986; Bilby and Ward, 1989; Bilby and Bisson, 1998; Gurnell
et al., 2002; Faustini and Jones, 2003, Mac Farlane and Wohl, 2003, Montgomery et al.,
2003; Rosenfeld and Huato, 2003; Comiti et al., 2008). The presence of LW is also
important for the ecology of fluvial systems as it increases the heterogeneity and quality of
stream habitats, and the biological diversity of aquatic biota (Beschta and Platts, 1986;
Bisson et al., 1987; Bilby and Ward, 1989; Maser and Sedell, 1994; Diez et al., 2001; Chen
et al., 2008; Vera et al., 2012).
Avalanches, landslides and debris flows are the dominant processes which determine wood
delivery to streams in steep forested low-order catchments, whereas tree mortality and bank
erosion are relatively more important in recruiting LW in medium-sized streams (Keller
and Swanson, 1979; Robison and Beschta, 1990; Bilby and Bisson, 1998; Hairston-Strang
and Adams, 1998; Martin and Benda, 2001; May and Gresswell, 2003; Reeves et al., 2003).
Overall, land use and management history of the catchments, forest density and
composition, species and age of the riparian forests, and inclination and stability of the
streambanks also condition the supply of LW to river systems (Hairston-Strang and Adams,
1998; Gurnell et al., 2002; Hassan et al., 2005; Iroumé et al., 2011; Ulloa et al., 2011).
3
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
789
Besides the morphological and ecological benefits, LW loading and downstream
mobilization can increase the associated hazard of floods. In-stream wood becomes
potentially dangerous to human infrastructure only during high-magnitude events, during
which log recruitment and transport can be very important (Castiglioni, 1974; Ishikawa,
1990; Braudrick and Grant, 2000; Daniels and Rhoads, 2003; Andreoli, 2006). Yet, wood
entrainment and transport have been often overlooked and received relatively little research
attention, as reported by Braudrick and Grant (2000) and ratified 10 years after by Curran
(2010) and Schenk et al. (2013).
Key properties of wood dimensions and fluvial processes determine wood dynamics
(Gurnell et al., 2002; Merten et al., 2010). The ratios of piece length and diameter to
bankfull channel width and depth are undoubtedly the two dimensionless parameters most
commonly adopted to predict wood mobility (Lienkaemper and Swanson, 1987; Bilby and
Ward, 1989; Abbe et al., 1993; Nakamura and Swanson, 1994; Young., 1994; Berg et al.,
1998; Diez et al., 2001; Martin, 2001; Gurnell et al., 2002; Gurnell, 2003; Warren and
Kraft, 2008; Wohl and Goode, 2008; Cadol and Wohl, 2010; Merten et al., 2010; Wohl,
2011). From theoretical models and flume experiments Braudrick and Grant (2000) found
that piece length did not significantly affect the threshold of movement for logs shorter than
channel width, noting also that extrapolation of their results was likely limited to low-
gradient alluvial systems. Although logs shorter than bankfull width do not necessarily
always move (Warren and Kraft, 2008), wood pieces tend to travel when the ratio of piece
length to bankfull channel width is lower than 0.5 in large rivers (Abbe et al., 1993) or
lower than 1.0 in smaller rivers (Lienkaemper and Swanson, 1987). According to
Mazzorana (2009), the ratio of piece diameter to bankfull channel depth that conditions the
4
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
101112
entrainment for smooth logs with an approximate cylindrical form is between 0.8 and 1,
and decreases to 0.7 and 0.6 for LW with branches and with rootwads, respectively.
Unattached, less dense and more decayed wood pieces without rootwads and oriented at 45º
and 90º to the flow are more mobile (Abbe and Montgomery, 1996; Braudrick and Grant,
2000; Bocchiola et al., 2006; Cadol and Wohl, 2010), while logs buried, anchored and
capable to sprout are more stable (Gurnell et al., 2002; Scherer, 2004; Wohl and Goode,
2008; Merten et al., 2010).
LW dynamics is also highly dependent on flow regime and channel slope (Gurnell et al.,
2002; Van der Nat et al., 2003; Cadol and Wohl, 2010; Merten et al., 2010). LW elements
are more stable in first-order streams and are mobilized only during episodic extreme
events (Bilby and Ward, 1989; Robison and Beschta, 1990; Gurnell, 2003; Swanson, 2003),
but as stream size and depth increase hydraulic processes dominate and LW is less stable
(Keller and Swanson, 1979; Harmon et al., 1986; Abbe and Montgomery, 2003; Gurnell,
2003). The magnitude and sequence of a series of flows are key factors for LW movement
(Haga et al., 2002; Wohl and Goode, 2008), and rapid increases of the hydrostatic forces of
buoyancy and lift and the hydrodynamic force of drag facilitate transport of in-stream wood
(Wohl et al., 2012). The ratio of peak water level to log diameter exerts a great influence
over LW mobility (Wohl and Goode, 2008), temporal variation of mobility rates are
explained by variation in peak flows and peak unit stream power (Wohl and Goode, 2008;
Cadol and Wohl, 2010), a flow magnitude greater than the previous flows is necessary to
re-transport most logs (Haga et al., 2002), and MacVicar et al. (2009) report for a single
flood on the Ain River (France) that wood transport rates are one order of magnitude higher
on the rising than on the falling limb of the hydrograph.
5
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
131415
LW moves farther and more frequently in large (> fifth order) than small streams (Bilby,
1985; Lienkaemper and Swanson, 1987; Bilby and Ward, 1989, 1991; Martin and Benda,
2001), smaller pieces move farther than larger pieces (Lienkaemper and Swanson, 1987;
Young, 1994) and piece diameter strongly influences depth of flow required to entrain and
transport logs, thereby influencing distance their displacement length (Bilby and Ward,
1989; Abbe et al., 1993; Braudrick et al., 1997; Braudrick and Grant, 2000). Pieces tend to
stop when the water depth of the segment at peak flows is less than the diameter of the logs
(Haga et al., 2002), or approximately half the piece diameter for Abbe et al. (1993). The
locations of stable or recurring LW jams can reduce LW movement distances for wood of
all sizes (Braudrick and Grant, 2001; Haga et al., 2002; Warren and Kraft, 2008).
In Latin America, besides a few anecdotal references to in-stream wood (Vidal Gormaz,
1875; Garcia-Martínez and Lopez, 2005) the first reports on LW are those by Montgomery
et al. (2003) and Wright and Flecker (2004). More abundant research has developed since
2007 focusing mainly on LW morphologic and hydraulics roles (Andreoli et al., 2007;
Comiti et al., 2008; Mao et al., 2008, 2013; Cadol et al., 2009; Cadol and Wohl, 2010;
Iroumé at al., 2010, 2011, 2014; Ulloa et al., 2011; Wohl et al., 2009, 2012) and also on the
ecology of low order mountain channels (Vera et al., 2012). First analyses of LW
mobilization in Latin American streams are the reports by Andreoli et al. (2008), Mao et al.
(2008) and Iroumé et al. (2011).
Despite the general agreement on the need of a minimum relative dimension for logs to be
moved in low-order streams, this evidence still needs to be proved in a variety of mountain
environments. Furthermore, the dependency of wood transport and dynamics to channel
morphology needs to be further explored, and verified under different flow regimes and
6
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
161718
over long time periods. This paper addresses these gaps by examining LW mobility over
several years in channel segments of four low-order (Strahler’s ordering system) mountain
streams, located in southern Chile. These streams represent two contrasting mountain
environments, one from the Andes with higher slopes and rainfall, geology of pyroclastic
rocks and old growth native forests, and the other from the Coastal range with gentler
slopes, geology of metamorphic rocks and a mixture of second growth native forests and
plantations with exotic tree species. The transport from a sample of over 1000 tagged logs
was surveyed during several time periods characterized by peakflows with return periods
up to 5 years. The study is focused at verifying that wood pieces shorter and thinner than
bankfull channel width and depth, respectively, are transported more than longer and ticker
pieces, as reported by previous researches. Also, the tagged logs are expected to move
associated to higher-magnitude events, and collected evidence contribute at expanding the
actual knowledge in an area to considerable uncertainty but high importance for river
management and flood risk control purposes.
2. Materials and methods
2.1. Study sites
Large wood mobility was studied over several time periods in channel segments of the
Pichún (37º30’12’’S; 72º45’54’’W, catchment area of 431 ha), El Toro (38º09'11’’S;
71º48'12’’W, catchment area of 1,783 ha), Tres Arroyos (38º27'57’’S; 71º33'44’’W, with a
catchment area of 907 ha) and Vuelta de Zorra (39º58’12’’S; 73º34’13’’W, catchment area
of 587 ha) streams, located both in the Coastal and Andes mountain ranges in southern
7
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
192021
Chile (Fig. 1). Geographic coordinates and catchment areas correspond with the location of
the downstream end of the study segments.
The southernmost study catchment is Vuelta de Zorra located in the Valdivian Coastal
Reserve some 40 km to the south-west of the city of Valdivia, The Rivers Region. Seventy
five percent of its area is covered by a second growth evergreen native forest, the 24% with
a Eucalyptus nitens plantation established in 1999 and 1% corresponds to several sites
where native species are regenerating. Long-term annual rainfall in the area exceeds 2300
mm. An exhaustive description of this site can be found in Iroumé et al. (2010, 2011) and
Ulloa et al. (2011).
The Tres Arroyos catchment is located within the Malalcahuello-Nalcas Forest National
Reserve in the Andes mountain range, some 2 km to the north-east of the small town of
Malalcahuello, Araucania Region. A little more than 64% of the catchment area is covered
by old-growth native forests, 23% by herbs and shrubs near the tree line, 6.4% by younger
(<40 yr) conifer plantations (Pseudotsuga menziesii, Pinus radiata, P. ponderosa, P.
monticola and P. contorta) and 6.4% are unvegetated sandy volcanic ashes around the
watershed divide. Long term mean annual rainfall in the area exceeds 2500 mm. Additional
information is presented in Andreoli et al. (2007, 2008) and Comiti et al. (2008).
The El Toro catchment is located within the Malleco Forest National Reserve in the Andes
mountain range, some 80 km to the north-east of the town of Victoria, Araucania Region.
In 2002 a wild forest fire severely affected the Malleco Forest Reserve, and within the
study catchment 88% of the forests were totally burned and the remaining 12% partially
affected (Andreoli et al., 2007). The burned forests corresponded to the Coihue-Raulí-Tepa
forest type, with Nothofagus dombeyi (coihue), N. nervosa (rauli) and Laureliopsis
8
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
222324
philippiana (tepa) the most important tree species. Long term mean annual rainfall in the
area is ~ 3000 mm.
Pichún is located in the eastern aspect slopes of the Coastal mountain range, some 20 km to
the south-west of the town of Nacimiento, Biobío Region. Eighty four percent of the area is
covered by E. globulus plantations and the remaining 16% corresponds to roads and
riparian vegetation. The eucalyptus plantations were established from 2005 and 2008 in
terrains previously covered by two consecutive rotations of P. radiata plantations. Long
term mean annual rainfall in the area is ~ 1190 mm, and more information from this
catchment can be found in Iroumé et al. (2011) and Ulloa et al. (2011).
An overview of the four study channels is presented Fig. 2.
2.2. Channel surveys
Since March 2005 for the Tres Arroyos and from November 2008-March 2009 for Pichún,
El Toro and Vuelta de Zorra, a segment of the main stream of each catchment has been
surveyed for LW and channel morphological characterization.
Each surveyed segment was divided into individual reaches defined based on uniformity of
either slope, channel width or LW abundance and numbered from downstream (Reach 1) to
upstream (see Andreoli et al., 2007; Comiti et al., 2008; Iroumé et al., 2010, 2011; Ulloa et
al., 2011). Numbered wooden stakes installed on the banks indicate the limits of each of the
reaches and serve as reference points for log positioning.
Following the thalweg line the longitudinal profile of each reach was surveyed using a laser
distance meter with an inclinometer in order to calculate reach length and channel slope.
Reach mean channel bankfull width and depth were estimated from cross-sections surveyed
at ≤ 15-m intervals as in Iroumé et al. (2010). Along the study segments at Pichún, El Toro
9
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
252627
and Vuelta de Zorra, and partially at Tres Arroyos, the channels feature well-defined
sections and reach bankfull stages were defined as the height of the floodplain surfaces. In
less well-defined sections, bankfull channel was identified by the height of the lower limit
of the vegetation, changes in sediment size in the small lateral bars and flow-deposited
organic debris, and observing as many bankfull indicators found in every reach. As
described by Andreoli et al. (2007) and Iroumé et al. (2010) all wood pieces found within
the bankfull channel and lying on the streambank with at least part of the piece extending
into the channel with dimensions of greater than 10 cm in diameter and 1 m in length, were
measured for length and mid-diameter and their position was referenced to natural features
(rocks, big trees) and to the numbered wooden stakes. Each LW piece was also classified
according to its type (log, rootwad, log with attached rootwad, branch, full tree), tree
species (broadleaved, conifer), state of decay (still alive, low, medium, high, based on
visual estimation), and source (bankcutting, landslides, toppling due to windthrow, fluvial
transport, sawn or harvested residue), position in the channel (in-bankfull channel, logstep,
bankfull line, channel-spanning logs, channel margins), orientation to stream flow (parallel,
orthogonal, oblique) and aggregation (single elements or jam forming logs).
For wood mobilization studies several LW pieces were randomly selected and tagged
during the surveys taking care that the frequency distribution of tagged logs remained
similar to that of all logs (Ruz, 2013).
A 1.54 km-long segment divided in 17 individual reaches was first surveyed in the Tres
Arroyos between March–April 2005 (Andreoli et al., 2007) at the end of the drier summer
months. A total of 1210 elements were recorded within the bankfull channel of the study
segment, and 322 of them (27% of the total) were tagged. This segment was re-surveyed in
10
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
282930
December 2005 after one rainy winter period and every tagged wood piece that had moved
downstream from its initial position was re-referenced and re-classified to investigate log
mobility. Log displacements shorter than 0.5 m were discarded. The Tres Arroyos was
surveyed again in January 2009 after three rainy winter periods from the previous survey,
and on this occasion the study segment was extended upstream of the original sampled
segment to a total length of 2.07 km with the addition of five reaches. Additional LW
pieces were tagged to a total of 461 elements (22% of the total of wood elements found
within the bankfull channel of the 2.07 km-long segment). Considering the channel
bankfull area as reference, LW volume in this 2.07 km-long segment is 1057 m3/ha (Iroumé
et al., 2014). This extended segment was again surveyed after consecutive rainy winter
periods at the end of 2009, 2010, 2011 and 2012, and every tagged wood piece that had
moved from its initial position was re-referenced and re-classified as described above.
Vuelta de Zorra, El Toro and Pichún were first surveyed between November 2008 and
March 2009, after the 2008 rainy winter period. The length and number of reaches of the
study channel segments were 1.56 (16 reaches), 2.19 (17 reaches) and 1.0 km (12 reaches)
for Vuelta de Zorra, El Toro and Pichún, respectively. After recording every wood element
within the bankfull channels, a total of 395, 117 and 79 LW pieces were tagged at the
Pichún, El Toro and Vuelta de Zorra study segments, i.e. 81, 16 and 71 % of the total,
respectively. These segments were re-surveyed at the end of the 2009, 2010, 2011 and
2012 rainy seasons, and every tagged wood piece found downstream from its initial
location was re-referenced and re-classified. Considering the channel bankfull area as
reference, LW volume is 109, 202 and 56 m3/ha for Vuelta de Zorra, El Toro and Pichun
respectively (Iroumé et al., 2014).
11
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
313233
The length of surveyed periods ranged from 274 to 393 days and included one rainy winter
period (hereafter annual periods). The Tres Arroyos was surveyed for a longer period
(2006-2008, 1140 days) spanning three winter periods. During each re-survey, all new LW
recruited from bankcutting or toppling was measured, classified, positioned and tagged, and
considered within the number of tagged LW in every reach for the next study period. Then,
at the beginning of a new study period the tagged sample in each reach was the number of
tagged logs at the beginning of the previous period, minus those tagged pieces not found in
their original reach plus the new recruited tagged pieces.
Information related to study catchments and reach and segment bankfull width and length
and longitudinal slope is summarized in Table 1.
Unit stream power ω (W m-2) was calculated from a simplification of the traditional
equation (i.e. Rigon et al., 2012), as follows:
ω=γ . (√g . H ) . ( H .W Bk ) . S
W Bk
(1)
where γ is the specific weight of water (N m-3), g is the acceleration due to gravity in (m s-
2), WBk is the mean segment bankfull width (m), S is the mean segment channel slope (m m-
1), and H is (m) the water depth.
The study channels do not present artificial interventions in the streams, or obstacles or
other factors limiting LW transport.
2.3. Precipitation and discharge
Precipitation and discharge are monitored in the study catchments with continuous digital
recorders. All catchments exhibit pluvial regimes with maximum monthly discharges 12
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
343536
occurring during the winter rainy months (May to August), but snowfall can occasionally
occur in the El Toro and Tres Arroyos study basins.
Rainfall in every site is measured using digital tipping-bucket gauges with a resolution of ~
0.2 mm. Water level gauging stations operate in the Tres Arroyos since 1997 and mid-2008
controlling catchment areas of 626 and 907 ha, respectively. Additionally water level
gauging stations are operating since mid-2008 controlling 413, 1760 and 585 ha at the
Pichún, El Toro and Vuelta de Zorra sites, respectively. The gauging stations are located in
or near the downstream end of each study segment (the exception is the one operating since
1997 at Tres Arroyos), in natural sections of the El Toro, Tres Arroyos and Vuelta de Zorra
streams and in a man-made concrete rectangular channel for Pichún. Water level is
measured at 10-15 minute intervals using digital sensors with a resolution of ± 2 mm. For
each of the study periods and stream channels, Hmax (the maximum water level) was
identified. Water level duration analyses were performed for each study period and channel
to define HX% (the level above which the flow remains for X% of the time).
Bankfull stages (HBk) at the location of the water level gauging stations were defined as
described above.
With data from the Tres Arroyos water level gauging station which has operated since 1997
and has the longest record of the study segments, a Gumbel flood-frequency analysis on the
single highest peak flow in each year allowed to estimate that the return periods of the
measured peakflows in this channel during the study were ~ 5 years. This is consistent with
the general condition in the country during these years, characterized by normal flows and
the absence of major flood events.
2.4. Statistical analysis
13
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
373839
LW mobility rate (in %) in every segment was expressed as the ratio of the number of
tagged wood elements that had moved in each period to the total number of tagged LW.
Large wood mobility (in %) and dimensions (diameter, length) were compared (t-tests)
among periods where maximum water level was higher and lower than bankfull stage.
Stepwise regressions were used to examine relationships between a range of mobilized LW
features (mobility, diameter, length and travel distance) as dependent variables and stream
water level characteristics during each annual period (Hmax the maximum water level; HBk
the bankfull stage; HX%, the level above which the flow remains for X% of the time; number
of floods exceeding HBk; combinations of these; unit stream power for different H) as
independent variables. Variables were t-tested for normality or otherwise log10 transformed.
Regressions and differences were considered statistically significant if P ≤ 0.05. The SAS
(version 9.1) statistical software package was used.
3. Results
3.1 Large wood mobilization
If the one year-long period studies are considered, the number of moving logs ranged from
0 and 46 pieces/period. Instead, if all the surveyed periods including the 1140 day-long
2006-2008 period at the Tres Arroyos catchment are considered, the number of moving
logs appears to range between 0 to 54 pieces/period (Table 2). The annual LW mobility
rate, expressed as the ratio of the number of mobilized tagged wood elements to the total
number of tagged LW, ranged between 0 and 28.2% but exhibited high variability among
different years and between catchments.
14
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
404142
The percentage of LW mobility was higher (range 1.2 to 28.2%, mean 9.2%) in annual
periods when maximum water level (Hmax) exceeded channel bankfull depth (HBk) than in
periods with flows lower than bankfull stage (range 0 to 7.3%, mean 3%), Fig.3. However,
mean annual mobility is not statistically different (t-test) among periods whene Hmax>HBk
and Hmax≤ HBk.
Hmax did not exceed HBk in periods 2010 for Vuelta de Zorra and Tres Arroyos, 2011 in El
Toro and 2011 and 2012 for Pichun. HBk was exceeded in Vuelta de Zorra in six occasions
in 2012 and in four in 2009 and 2011, and four times in El Toro during 2009. In Pichun,
only one peakflow exceeded bankfull level in 2009 and 2010. Water level records at the
different gauging stations are summarized in Fig. 4, where dashed horizontal lines indicate
bankfull stage in the corresponding cross sections.
If only annual periods when Hmax>HBk are considered, a significant linear regression (R =
0.56, P ≤ 0.05) appears between LW mobility (in %) and the ratio Hmax/HBk. This linear
relationship does not improve by replacing Hmax by the mean of all H exceeding HBk in
every period. In addition, LW mobility (in %) was not related to the number of peakflows
exceeding HBk.
A statistically significant (P ≤ 0.05) multiple linear regression with unit stream power (with
H = Hmax in eq. 1) and the ratio Hmax/HBk as independent variables explains the 42.4% of the
variation on annual LW mobility (%). Considering each independent variable separately,
linear regressions indicate that LW mobility (in %) increases significantly with increasing
unit stream power and Hmax/HBk, Fig. 5.
3.2 The characteristics of transported wood
15
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
434445
Seventy two percent of the diameters and 81% of the lengths of the tagged wood pieces
were ≤ than 40 cm and 7 m, respectively (Table 3). Eighty four percent of the moved LW
had diameters ≤ than 40 cm and 91.6% had lengths ≤ than 7 m. Few logs with diameter
greater than 60 cm and length over 10 m moved.
For 49 and 78% of all moved logs in the study catchments the ratios of piece length to
bankfull channel width (Llog/WBk) were lower than 0.25 and 0.5, respectively, Fig. 6. The
ratio Llog/WBk was ≤ than 1.0 for the 96% of the transported wood pieces, and Pichun was
the only study catchment where an important number of moved pieces (23%) had a
Llog/WBk ratio higher than 1.0. In Fig. 6, the percentage line of tagged LW does not reach
100% because the 3% are elements with relative length greater than 1.5.
In all study catchments the diameter of 59 and 96% of the moved logs was lower than 0.25
and 0.5 of bankfull channel depth (Fig. 6). For only 4% of the moved logs the ratios
Dlog/HBk were higher than 0.5, but in all cases ≤ than 1.0.
At the Vuelta de Zorra, Tres Arroyos and Pichun channels, the movements of LW were
concentrated in a few reaches, whereas in El Toro wood transportation was distributed
along the entire segment (Fig.7). In Vuelta de Zorra, reaches 6 to 9 concentrated most of
the movement during the study period. In the Tres Arroyos LW movement was
concentrated mainly in reaches 8 to 10 and in reach 1 at the Pichun channel.
Considering tree species, broadleaved corresponds to more than the 91% of the moved logs
(Table 4). Conifers have an important participation in Pichun explained by remains of the
P. radiata plantations that covered this catchment from the beginning of the 1960´s until
2005 (Iroumé at al., 2011; Ulloa et al., 2011). Conifers in Vuelta de Zorra and Tres Arroyos
16
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
464748
are mainly logs of the native Saxegothaea conspicua and Araucaria araucana species,
respectively (Andreoli et al., 2007; Iroumé at al., 2010).
Ninety three percent of all moved wood pieces were logs, while branches and rootwads
were ~ 3% each (Fig 8), and only one full tree and one log with attached rootwad were
mobilized during the study period. Forty two percent of the moved logs were found in the
first surveys oriented parallel, 37% orthogonal and 21% oblique to stream flow (Fig.8).
Eighty four percent of the moved logs were classified in the first surveys as being fluvially
transported from upstream, 7% recruited from natural causes (toppling due to windthrow,
broken branches, death), 7% as sawn or harvested residues, 1.2% from bankcuttings and
0.4% from lansdslides, Fig.8. In Vuelta de Zorra and especially in Pichun, moved logs
classified in the first surveys as sawn or harvested residues were important (5.2 and 54.5%,
respectively), which reflects the forest operations that took place in these two catchments
(see Iroumé at al., 2010, 2011; Ulloa et al., 2011).
Considering all study catchments, 67.7% of the moved logs were found in the first surveys
within the bankfull channel, 17.9% in the bankfull range, 10.8 % in the channel margins,
2.4 % in log-steps and 1.2 % as channel-spanning logs, Fig.8. The amount of logs moved
from the channel margins was higher in Vuelta de Zorra and Pichun, which can be again
associated to the forest operations that took place in these two catchments. When tagged,
47.2% of the LW were lying on the channels as single elements and 52.8% in
accumulations, Fig.8. Finally, 25% of the wood pieces in accumulations were in contact
with less or equal to other 5 elements and 28% was jam forming logs. After transport, 37%
of the LW was found as single elements and 63% as jammed in accumulations. Also, 22.7
% of transported pieces were found in logjams.
17
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
495051
3.3 Relationships between mobilized wood pieces and stream flow features
Diameter and length of moved LW were not significantly different between annual periods
with flows exceeding and with flows less than bankfull stage (Fig.3). Mean and median
diameters of mobilized LW were the same under both conditions (~ 26 and ~21 cm
respectively), but some of the larger wood pieces were moved during periods were Hmax>
HBk. However, lengths of mobilized LW were relatively similar in periods when Hmax> HBk
(mean and median lengths of 3.8 and 3.2 m) than in periods when Hmax≤ HBk (mean and
median lengths of 4.1 and 3 m), but the longest wood piece (18 m) was moved at the El
Toro in 2011 when Hmax ≤ HBk.
For a range of absolute and relative piece diameters (Dlog, < 20 cm, 20 - 40 cm and > 40 cm;
Dlog/HBk, < 0.5 and 0.5 – 1.0) and lengths (Llog, < 4 m, 4 – 7 m and > 7 m; Llog/WBk, < 0.5),
LW mobility (%) was generally higher in annual periods where maximum water level
exceeded channel bankfull depth than when flows were less than bankfull stage, but again
the differences were not statistically significant. Mobility of LW pieces with Dlog/HBk > 0.5
and Llog/WBk > 1 was very low either when maximum water level exceeded channel
bankfull depth or when it did not.
The mean length of moved logs in every annual period and study catchment was related
(statistically non-significant linear correlation) only with Hmax. Mean diameter of
transported wood pieces per period was significantly correlated (P ≤ 0.05) with unit stream
power (with H = H15% in eq. 1, the level above which the flow remains for 15% of the time)
and H50% (the level above which the flow remains for 50% of the time), Fig.9.
4. Discussion
18
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
525354
The values of LW annual mobility rates in the analyzed low-order streams with runoff
dominated by winter rains range between 0 and 28.2%. This range of mobility rates is
narrower than the 8% to 59% reported by Cadol and Wohl (2010) in tropical headwater
streams. Cadol and Wohl (2010) mention that the higher mobility of wood in this study site
than in temperate rainforests of the Pacific Northwest (Hassan et al., 2005) may be
explained by the occurrence of frequent and flashy floods. Wohl and Goode (2008) found
average yearly mobility ranging from 16% to 23% in snowmelt runoff dominated streams
(Rocky Mountains, Colorado) while Schenk et al. (2013) report mobility of ~ 41% per year
from measurements in a large low gradient river. After one year of study Mao et al. (2008)
found that wood mobility after ordinary flows was 16% in the Buena Esperanza (southern
Patagonia, Argentina) basin. Beside, using a longer record Gurnell et al. (2002) report that
less than 1% of the logs in Mack Creek (third order basins, Pseudotsuga and Tsuga spp.
forests, Oregon, USA) moved in most years after normal floods, but even in an
approximately 25-year return period flood only 11% of the pieces moved more than 10 m.
However, the Mack Creek in Oregon features very large wood pieces and a high proportion
of the tagged pieces were up on the bank outside the flood zone, so the comparison with
streams in this study in Chile should be carefully approached. Using a similar methodology
to our investigation, Warren and Kraft (2008) report that 22% of tagged LW moved 5.0 m
or more during the 4 years of their study, which is relatively similar to the 17% mobility
rate observed at the Tres Arroyos in 2006-2008 period (see Table 2).
During the study period only ordinary floods occurred, thus annual LW mobility rates up to
28% reported in the study channels are associated to return periods ≤ 5 years.
19
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
555657
Mobility varied greatly among the different years and catchments (see Table 2), and on
average it was great at El Toro and Pichun (11.6 and 8.6%, respectively) and lowest at Tres
Arroyos (2.5%). El Toro is the widest study segment (see Table 1) and several of the LW
pieces in Pichun were windthrown P. radiata trees left near the channel at the end of the
last harvesting operation. The lowest mobility in the Tres Arroyos can be explained by the
large size of the LW pieces relative to bankfull channel depth and width, the high volume
of LW stored in the channel and the dimensions and density (Nº km-1) of large logjams
(Ruz, 2013). Also, in this kind of channel important mobilization should only be expected
during episodic extreme events as indicated by Bilby and Ward (1989), Robison and
Beschta (1990), Gurnell (2003) and Swanson (2003).
Logs remained completely immobile only during the 2011 period at the Pichun channel
(Table 2), when Hmax/HBk was 0.27. Fifty seven percent of the LW tagged in this stream has
a Dlog/HBk ratio ≤ 0.25 so the hydrostatic forces of buoyancy and lift generated by this year’s
flows may have not been capable of moving the wood pieces.
LW mobility is higher in annual periods when maximum water level exceeded channel
bankfull depth than in periods with flows less than bankfull stage. However, this difference
is not statistically significant which can be explained by the fact that peakflows exceeding
bankfull depth during the study are not extraordinary flows (return periods ≤ 5 years) and
that Hmax ranged between 0.81 and 0.99 of HBk in periods with flows less than bankfull stage
and with LW movement.
Only for annual periods when Hmax>HBk, the percentage of LW mobility is significantly
correlated with the ratio Hmax/HBk but not with the number of peakflows exceeding HBk. This
partially coincides with evidence provided by Haga et al. (2002) and Wohl and Goode
20
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
585960
(2008) who indicate that the magnitude and sequence of a series of flows are key factors for
LW movement. For all annual study periods (both when maximum water level exceeded
and with flows less than channel bankfull depth), LW mobility is significantly correlated
with unit stream power (with H = Hmax in eq. 1) and Hmax/HBk, confirming Wohl and Goode
(2008) and Cadol and Wohl (2010) which state that temporal variation of mobility rates are
explained by variation in peak flows and peak unit stream power. Other variables
associated to flow characteristics are also related to LW mobility (see Table 5), but they do
not provide additional information to explain temporal variations of mobility rates as they
are positively (and several of them significantly) correlated with unit stream power for Hmax
and Hmax/HBk.
The ratios of piece length and diameter to bankfull channel width and depth for the 96% of
moved LW were ≤ 1.0 for Llog/WBk and ≤ 0.5 for Dlog/HBk which confirms the importance of
these dimensionless parameters to explain wood mobility, and these values are consistent
with Lienkaemper and Swanson (1987), Nakamura and Swanson (1994), Wohl and Goode
(2008) and Cadol and Wohl (2010) in terms of LW entrainment conditions. Pichun was the
only study channel where an important number of moved pieces (23%) had a Llog/WBk ratio
higher than 1.0. In this stream 50% of the moved LW were P. radiata wood pieces
(conifers, see Table 4) which are less dense and more susceptible to decay than the native
broadleaved (Diaz-vaz et al., 2002). Because in-stream wood in this channel is evenly
distributed among conifers and broadleaved which additionally have similar dimensions, it
is not possible to confirm that higher buoyancy and decay susceptibility facilitate LW
transport as reported by Braudrick and Grant (2000), Gurnell et al. (2002) and Wohl et al.
21
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
616263
(2012). Native broadleaved comprises more than the 90% of the in-stream wood in the
other channels.
Ninety three per cent of all moved wood pieces were logs, and before transportation 57.7%
were oriented orthogonal or oblique to stream flow and 72% were single elements or in
contact with ≤ than other 5 elements which confirm Abbe and Montgomery (1996),
Braudrick and Grant (2000), Bocchiola et al. (2006) and Cadol and Wohl (2010) that
unattached wood pieces without rootwads and not oriented parallel to the flow are more
mobile. The fact that 67.7% of the moved logs were found in the first surveys within the
bankfull channel and the 84 % of the moved logs were classified in the first surveys as
being fluvially transported from upstream, may be representative of the LW mobilization in
the rainy study streams characterized by several flashy flows approaching or exceeding
bankfull level each year.
Dimensions of moved LW (absolute diameter and length and relative to bankfull channel
width and depth) were not significantly different between annual periods with flows
exceeding and with flows less than bankfull stage. This may be explained by the fact that
peakflows exceeding bankfull depth during the study have return periods lower than 5
years, as discussed above. We did not find any significant relation between mean length of
moved logs in every annual period and study catchment and stream flow features, but the
mean diameter of transported wood pieces per period was significantly correlated with unit
stream power with H = H15% in eq. 1. This may suggest that wood diameter is the most
important factor influencing the propensity of LW to move as reported by Braudrick and
Grant (2000) from modelling and flume experiments.
22
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
646566
At the Vuelta de Zorra, Tres Arroyos and Pichun channels LW movement concentrated on
a few reaches, both in terms of wood pieces leaving (“out” in Fig.8) or arriving (“in” in
Fig.8) to a given reach. In Vuelta de Zorra, reaches 6 to 9, and especially reach 8,
experienced most of the movement during the study period. These reaches feature an
average LW volume of 243 m3 per hectare of bankfull channel while the average LW
volume in the segment is 109 m3/ha (Iroumé et al., 2014), between 2 and 5 logjams while
all the other reaches have ≤ than 2 logjams, and on average stream channel in these reaches
is wider (at bankfull level) and less steep than mean segment bankfull width and
longitudinal slope (Ruz, 2013). In the Tres Arroyos, LW movement is concentrated mainly
in reaches 8 to 10 which an average stock of 1288 m3/ha of LW while the mean segment
LW volume is 1057 m3/ha (Iroumé et al., 2014), and especially in reach 10 (with 1498
m3/ha of LW volume) which has by far the highest number of logjams (23) in the study
segment. LW movement in Pichun concentrates in reach 1 (the lowermost reach), which
has one logjam as other reaches, but bankfull channel is much wider and less steep than
mean segment bankfull width and longitudinal slope. Overall LW movement concentrated
in reaches which characterized by higher LW volume and wider and less steep channels.
Overall, more mobile wood pieces are logs shorter than WBk and with diameters ≤ 0.5 HBk,
and those that before mobilization have been previously fluvially transported from
upstream and were lying within the bankfull channel oriented parallel or oblique to stream
flow as single elements or in contact with ≤ than other 5 elements. Wider and less steep
reaches accumulating high quantities of wood (wood volume and jams) both in terms of
origin and destination tend to concentrate LW mobilization at least in streams draining
catchments of less than 9 km2. Annual mobility rates within a segment depend on unit
23
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
676869
stream power (with H = Hmax in eq. 1) and Hmax/HBk and is not different among periods
when maximum water level exceeded channel bankfull depth than in periods with flows
less than bankfull stage at least for peakflows with return periods ≤ 5 years.
5. Conclusions
This study shed light on the processes of entrainment and transport of LW in high-gradient
mountain streams, which have been less studied than small streams and large rivers. Our
results confirm previous evidence, especially on the characteristics of moved LW (absolute
and relative dimensions to bankfull channel width and depth, orientation to flow, position
in the channel, and level of aggregation) and provide new data on annual transport rates and
some insights on their relations with stream flow. However, LW transport processes are
still not completely understood. Some of the difficulties in comparing results from different
evidence gathered around the world arise from the natural heterogeneity among
environments but also because of the variety of methodologies which have been used in
flume and field studies on wood entrainment. Further developments in LW mobility
understanding would perhaps require a more specific “common metrics” as Wohl et al.
(2010) recommend for studies focusing on amount and distribution of in-stream wood in
rivers. Future research should adopt more comparable methodologies, frequency of
surveys, length of explored river segments and others.
Annual LW mobility rates found in our study may be associated to relatively normal flows
(return periods ≤ 5 years) in low order channels. These values represent the lower mobility
rates that could be expected in our mountain catchments.
24
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
707172
Acknowledgments. This research was developed under Project FONDECYT 1110609. We
are grateful to Forestal Mininco S.A. (Pichún), Corporación Nacional Forestal (El Toro and
Tres Arroyos) and The Nature Conservancy (Vuelta de Zorra) for their interest and support
in these studies. We thank Javier Badilla and Daniel Antileo for the field work and we
acknowledge the valuable comments and suggestions by four anonymous reviewers.
References
Abbe, T.B., Montgomery, D.R., Featherston, K., McClure E., 1993. A process-based
classification of woody debris in a fluvial network; preliminary analysis of the Queets
River, Washington. EOS Transaction of the American Geophysical Union 74, 296.
Abbe, T.B., Montgomery, D.R., 1996. Large woody debris jams, channel hydraulics and
habitat formation in large rivers. Regulated Rivers: Research and Management 12, 201–
221.
Abbe, T.B., Montgomery, D.R., 2003. Patterns and processes of wood debris accumulation
in the Queets river basin, Washington. Geomorphology 51, 81-107.
Andreoli, A., 2006. Volume and Characteristics of Woody Debris in Mountain Rivers of
Italian Dolomites and Southern Andes. Tesis, Dottorato di Ricerca in Gestione Ambientale
dei Bacini Idrografici e Tecniche di Rappresentazione del Territorio, Università degli Studi
de Padova, Padova, Italia. 116p.
Andreoli, A., Comiti, F., Lenzi, M.A., 2007. Characteristics, distribution and geomorphic
role of large woody debris in a mountain stream of the Chilean Andes. Earth Surface
Processes and Landforms 32, 1675–1692.
25
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
737475
Andreoli, A., Comiti, F., Mao, L., Iroumé, A., Lenzi, M.A., 2008. Evaluación de los
volúmenes y de los efectos hidro-morfológicos del material leñoso en dos torrentes
Andinos (Chile). Ingeniería del Agua 15 (3), 189–204 (in Spanish).
Beechie, T.J., Sibley, T.H., 1997. Relationships between channel characteristics, woody
debris, and fish habitat in Northwestern Washington streams. Trans. Am. Fish. Soc. 126,
217–229.
Berg, N., Carlson, A., Azuma, D., 1998. Function and dynamics of woody debris in stream
reaches in the central Sierra Nevada California. Can. J. Fish. Aquat. Sci. 55:1807–1820.
Beschta, R. L., Platts, W., 1986. Morphological features of small streams: significance and
function. Water Resour. Bull. 22(3), 369–379.
Bilby, R.E., 1985. Influence on stream size on the function and characteristics of large
organic debris. West Coast Meeting of the National Council of the Paper Industry for Air
and Stream Improvement, pp. 1–14.
Bilby, R.E., Bisson, P.A., 1998. Function and distribution of large woody debris. In:
Naiman, R.J., Bilby, R.E. (Eds), River Ecology and Management: Lessons from the Pacific
Coastal Ecoregion. Springer New York, 324–346.
Bilby, R. E., Ward, J.W., 1989. Changes in characteristics and function of woody debris
with increasing size of streams in western Washington. Trans. Am. Fish. Soc. 118, 368–
378.
Bilby, R. E., Ward, J.W., 1991. Changes in characteristics and function of large woody
debris in streams draining old-growth, clear-cut, and second-growth forests in southwestern
Washington. Can. J. Fish. Aquat. Sci. 48, 2499–2508.
26
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
767778
Bisson, P.A., Bilby, R.E., Bryant, M.D., Dolloff, C.A., Grette, G.B., House, R.A., Murphy,
M.L., Koski, K.V., Sedell, J.R., 1987. Large woody debris in forested streams in the Pacific
Northwest: past, present, and future. In: Salo, E.O., Cundy, T.W. (Eds), Streamside
Management: Forestry and Fishery Implications, Institute of Forest Resources,
Contribution No. 57. University of Washington, Seattle, WA, 143–190.
Bocchiola, D., Rulli, M.C., Rosso, R., 2006. Flume experiments on wood entrainment in
rivers. Advances in Water Resources 29, 1182–1195.
Braudrick, C.A., Grant, G.E., 2000. When do logs move in rivers? Water Resour. Res. 36
(2), 571-583.
Braudrick, C.A., Grant, G.E., 2001. Transport and deposition of large woody debris in
streams: a flume experiment. Geomorphology 41, 263-283.
Braudrick, C.A., Grant, G.E., Ishikawa, Y., Ikeda, H., 1997. Dynamics of wood transport in
streams: a flume experiment. Earth Surf. Processes Landforms 22, 669–683.
Cadol, D., Wohl, E., 2010. Wood retention and transport in tropical, headwater streams, La
Selva Biological Station, Costa Rica. Geomorphology 123, 61-73.
Cadol, D., Wohl, E., Goode, J.R., Jaeger, K.L., 2009. Wood distribution in neotropical
forested headwater streams of La Selva, Costa Rica. Earth Surf. Processes Landforms 34,
1198–1215.
Castiglioni, G.B., 1974. Le calamità naturali nelle Alpi. Proc. XXI Congresso Geografico
Italiano, Istituto di Geografia, Università di Padova, pp. 335.
Chen, X., Wei, X., Scherer, R.,Hogan, D., 2008. Effects of large woody debris on surface
structure and aquatic habitat in forested streams, southern interior British Columbia,
Canada. River Res. Appl. 24, 862–875.
27
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
798081
Comiti, F., Andreoli, A., Lenzi, M.A., Mao, L., 2008. Wood storage in three mountain
streams of the Southern Andes and its hydro-morphological effects. Earth Surf. Processes
Landforms 33, 244–262.
Curran, J.C., 2010. Mobility of large woody debris (LWD) jams in a low gradient channel.
Geomorphology 116, 320-329.
Daniels, M.D., Rhoads, B.L., 2003. Influence of a large woody debris obstruction on three-
dimensional flow structure in a meander bend. Geomorphology 51, 159-173.
Diaz-vaz, J. E., Poblete, H., Juacida, R., Devlieger, F., 2002. Maderas comerciales de Chile,
91 pp., Tercera edición, Valdivia, Chile (in Spanish).
Diez, J.R., Elosegi, A., Pozo, J., 2001. Woody debris in North Iberian stream: influence of
geomorphology, vegetation, and management. Environ. Manage. 28, 687–698.
Faustini, J. M., Jones, J.A., 2003. Influence of large woody debris on channel morphology
and dynamics in steep, boulder-rich mountain streams, western Cascades, Oregon.
Geomorphology 51, 187–205.
Garcia-Martínez, López, J.L., 2005. Debris flows of December 1999 in Venezuela, in
Debris-flow Hazards and Related Phenomena, edited by M. Jakob and O. Hungr, Chapter
20, pp., 519-538, Praxis, Springer Berlin Heidelberg.
Gurnell, A. M., Piégay, H., Swanson, F.J., Gregory, S.V., 2002. Large wood and fluvial
processes. Freshwater Biol. 47, 601–619.
Gurnell, A.M., 2003. Wood storage and mobility. In: Gregory, S.V., Boyer K.L., Gurnell,
A.M. (Eds.), The Ecology and Management of Wood in World Rivers. American Fisheries
Society Symposium 37, Bethesda, Md., pp. 75-91.
28
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
828384
Haga, H., Kumagai, T., Otsuki, K., Ogawa, S., 2002. Transport and retention of coarse
woody debris in mountain streams: An in situ field experiment of log transport and a field
survey of coarse woody debris distribution. Water Resour. Res. 38 (8), 1126,
10.1029/2001WR001123.
Hairston-Strang, A.B., Adams, P.W., 1998. Potential large woody debris sources in riparian
buffers after harvesting in Oregon, U.S.A. For. Ecol. Manage. 112, 67-77.
Harmon, M.E., J.F. Franklin, F.J. Swanson, P. Sollins, S.V. Gregory, J.D. Lattin, N.H.
Anderson, S.P. Cline, N.G. Auman, J.R. Sedell, G.W. Lienkaemper, K.C. Romack, and
K.W. Cummins (1986), Ecology of coarse woody debris in temperate ecosystems. In:
MacFadyen, A., Ford, E.D. (Eds), Advances in ecological research, Academic Press Inc.,
Orlando, FL., 133–302.
Hassan, M.A., Hogan, D.L., Bird, S.A., May, C.L., Gomi, T., Campbell, D., 2005. Spatial
and temporal dynamics of wood in headwater streams of the Pacific Northwest. J. Am.
Water Resour. Assoc. 41, 899-919.
Iroumé, A.,Andreoli, A.,Comiti, F., Ulloa, H., Huber, A., 2010. Large wood abundance,
distribution and mobilization in a third order Coastal mountain range river system, southern
Chile. For. Ecol. Manage. 260, 480–490.
Iroumé, A., Ulloa, H., Lenzi, M.A., Andreoli, A., Gallo, C., 2011. In-stream large wood
mobility and recruitment in two channels in the Coastal Mountain Range, Chile. Bosque
32(3), 247-254 (in Spanish).
Iroumé, A., Mao, L., Ulloa, H., Andreoli, A., 2014. Large wood volume and longitudinal
distribution in channel segments draining catchments with different land use, Chile. Open
Journal of Modern Hydrology 4, 57-66
29
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
858687
Ishikawa, Y., 1990. Studies on disaster caused by debris flow carrying floating logs down
mountain streams. SABO Depart., Public Works Research Institute, Ministry of
Construction, Tsukuba, Japan. 121 p.
Keller, E.A., Swanson, F.J., 1979. Effects of large organic material on channel form and
fluvial processes. Earth Surf. Processes Landforms 4, 361-380.
Lienkaemper, G.W.,Swanson, F.J., 1987. Dynamics of large woody debris in old-growth
Douglas-fir forests. Can. J. For. Res. 17, 150–156.
MacFarlane, W.A., Wohl, E.E., 2003. Influence of step composition on step geometry and
flow resistance in step-pool streams of the Washington Cascades. Water Resour. Res. 39:
2001WR001238.
MacVicar, B., Piegay, H., Henderson, A., Comiti, F., Oberline, C., Pecorari, E., 2009.
Quantifying the temporal dynamics of wood in large rivers: field trials of wood surveying,
dating, tracking, and monitoring techniques. Earth Surf. Processes Landforms 34:2031–
2046.
Martin, D., 2001. The influence of geomorphic factors and geographic region on large
woody debris loading and fish habitat in Alaska coastal streams. North American J.
Fisheries Manage. 21, 429–440.
Martin, D.J.,Benda, L.E., 2001. Patterns of Instream Wood Recruitment and Transport at
the Watershed scale. Trans. Am. Fish. Soc. 130, 940-958.
Maser, C., Sedell, J.R., 1994. From the Forest to the Sea: The Ecology of Wood in Streams,
Rivers, Estuaries, and Oceans. St. Lucie Press: Delray Beach, FL.
Mao, L., Burns, S., Comiti, F., Andreoli, A., Urciuolo, A., Gaviño-Novillo, M., Iturraspe,
R., Lenzi, M.A., 2008. Acumulaciones de detritos leñosos en un cauce de montaña de
30
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
888990
Tierra del Fuego: análisis de la movilidad y de los efectos hidromorfológicos. Bosque,
29(3), 197-211 (in Spanish).
Mao, L., Andreoli, A., Iroumé, A., Comiti, F., Lenzi, M.A., 2013. Dynamics and
management alternatives of in-channel large wood in mountain basins of the southern
Andes. Bosque 34(3), 319-330, 2013
May, C.L., Gresswell, R.E., 2003. Processes and Rates of Sediment and Wood
Accumulation in Headwater streams of the Oregon Coast Range, USA. Earth Surf.
Processes Landforms 28, 409-424.
Mazzorana, B.. 2009. Woody debris recruitment prediction methods and transport analysis.
PhD Thesis, Institute of Mountain Risk Engineering, University of Natural Resources and
Applied Life Sciences, Vienna. 195 p.
Merten, E., Finlay, J., Johnson, L., Newman, R., Stefan, H., Vondracek, B., 2010. Factors
influencing wood mobilization in streams. Water Resour. Res. 46, W10514,
doi:10.1029/2009WR008772.
Montgomery, D.R., Collins, B.D., Buffington, K.M., Abbe, T.B., 2003. Geomorphic effects
of wood in rivers. In: Gregory, S.V., Boyer K.L., Gurnell, A.M. (Eds.), The Ecology and
Management of Wood in World Rivers. American Fisheries Society Symposium 37,
Bethesda, Md., pp. 21-47.
Nakamura, F., Swanson, F.J., 1994. Distribution of coarse woody debris in a mountain
stream, western Cascades Range, Oregon, Can. J. For. Res. 24, 2395–2403.
Reeves, G.H., Burnett, K.M., McGarry, E.V., 2003. Sources of large wood in the main stem
of a fourth-order watershed in coastal Oregon. Can. J. For. Res. 33, 1363–1370.
31
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
919293
Rigon, E., Comiti, F., Lenzi, M.A., 2012. Large wood storage in streams of the Eastern
Italian Alps and the relevance of hillslope processes. Water Resour. Res. 48, W01518,
doi:10.1029/2010WR009854.
Robison, E.G., and R.L. Beschta (1990), Characteristics of coarse woody debris for several
coastal streams of southeast Alaska, USA. Can. J. Fish. Aquat. Sci., 47, 1684–1693.
Rosenfeld, J.S., L. Huato, L., 2003. Relationship between large woody debris
characteristics and pool formation in small coastal British Columbia streams. North
American J. Fisheries Manage. 23 (3), 928–938.
Ruz, C., 2013. Volumen, distribución longitudinal y efecto de las acumulaciones de
material leñoso en cauces de cuencas afectadas con diferentes tipos de alteraciones. Tesis,
Ingeniería Civil en Obras Civiles, Universidad Austral de Chile. Valdivia, Chile (in
Spanish).
Scherer, R., 2004. Decomposition and longevity of in-stream woody debris: a review of
literature from North America. In: Scrimgeour, G.J., Eisler, G., McCulloch, B., Silins, U.,
Monita, M. (Eds), Forest Land–Fish Conference II – Ecosystem Stewardship through
Collaboration, edited by, , Proc. Forest-Land-Fish Conf. II, April 26-28, 2004, Edmonton,
Alberta, pp. 127-133
Schenk, E.R., Moulin, B., Hupp, C.R.,Richter, J.M., 2013. Large wood budget and
transport dynamics on a large river using radio telemetry. Earth Surf. Processes
Landforms, DOI: 10.1002/esp.3463
Swanson, F.J., 2003. Wood in rivers: a landscape perspective. American Fisheries Society
Symposium 37, 299–313.
32
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
949596
Ulloa, H., Iroumé, A., Lenzi, M.A., Andreoli, A., Alvarez, C., Barrera, V., 2011. Large
wood in two catchments from the Coastal Mountain range with different land use history.
Bosque 32(3), 235-245 (in Spanish).
Van der Nat, D., Tockner, K., 2003. Large wood dynamics of complex Alpine river
floodplains. J. North American Benthological Soc. 22(1), 35-50.
Vera, M., Jara, C., Iroumé, A., Ulloa, H., 2012. Organic matter retention and diversity of
benthic communities influenced by in-stream large wood in a third-order channel in the
Coastal Mountain Range, Chaihuín, Chile. Conference Book, IUFRO Landscape Ecology
Conference, Sustaining humans and forests in changing landscapes, Forest, Society and
Global Change, Concepcion, Chile, 5th to 12th November 2012, 111.
Vidal Gormaz, F., 1875. Reconocimiento del Río Maullín, por la Comisión Esploradora de
Chiloé i Llanquihue. Anuario Hidrográfico de la Marina de Chile. Valparaíso, Instituto
Hidrográfico de la Armada de Chile, pp. 167-262.
Warren, D.R., Kraft, C.E., 2008. Dynamics of large wood in an eastern U.S. mountain
stream. For. Ecol. Manage. 256, 808–814.
Wohl, E., 2011. Threshold-induced complex behavior of wood in mountain streams.
Geology 39 (6), 587–590.
Wohl, E., Goode, J.R., 2008. Wood dynamics in headwater streams of the Colorado Rocky
Mountains. Water Resour. Res. 44, W09429, doi:10.1029/2007WR006522.
Wohl, E.E., Ogden, F.L., Goode, J., 2009. Episodic wood loading in a mountainous
neotropical watershed. Geomorphology 111, 149–159.
33
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
979899
Wohl, E.E., Cenderelli, D.A., Dwire, K.A., Ryan-Burkett, S.E., Young, M.K., Fausch,
K.D., 2010. Large in-stream wood studies: a call for common metrics. Earth Surf.
Processes Landforms 35, 618–625.
Wohl, E., Bolton, S., Cadol, D., Comiti, F., Goode, J.R., Mao, L., 2012. A two end-member
model of wood dynamics in headwater neotropical rivers. J. Hydrol. 462–463, 67–76.
Wright, J.P., Flecker, A.S., 2004. Deforesting the riverscape: the effects of wood on fish
diversity in a Venezuelan piedmont stream. Biological Conservation 120, 439-447.
Young, M.K., 1994. Movement and characteristics of streambourne coarse woody debris in
adjacent burned and undisturbed watersheds in Wyoming. Can. J. For. Res. 24, 1933–1938.
34
746
747
748
749
750
751
752
753
754
755
100101102
Figure Captions
Figure 1. Location of the study catchments. Circles correspond to main cities and triangles
to the locations of the study catchments.
Figure 2. Overview of the study channels: Pichún (above left), El Toro (above right), Tres
Arroyos (below left) and Vuelta de Zorra (below right).
Figure 3. Box plots of LW mobility (%) for Hmax ≤ HBk and Hmax > HBk (above) and LW
mobilized dimensions (diameter, below left; length, below right) for Hmax ≤ HBk and Hmax >
HBk. The cross and line within each box indicate the mean and median values, box ends are
25th and 75th percentiles, and whiskers are the minimum and maximum values.
Figure 4. Water level at the gauging stations (dashed horizontal lines indicate bankfull
stage).
Figure 5. Relationships between LW mobility (in %) and unit stream power for Hmax
(above) and the ratio Hmax/HBk (below).
Figure 6. Frequency distribution (%) of moved and tagged LW pieces below specific
Llog/WBk (above) and Dlog/HBk (below) ratios. The percentage line of tagged LW for
LLog/WBk condition does not reach 100% because the 3% are elements with relative length
greater than 1.5.
Figure 7. Number of moved LW pieces in the different reaches of the study segments.
Figure 8. Type (above left), orientation to stream flow (above right), source (center left),
initial position (center right) and aggregation (below) of moved LW.
Figure 9. Relationships between mean length of moved LW with Hmax (above), and mean
diameter of moved LW with H50% (center) and unit stream power for H15% (below).
35
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
103104105
Figure 1. Location of the study catchment. Circles correspond to main cities and triangles to
the locations of the study catchments.
36
779
780
781
782
783
106107108
Figure 2. Overview of the study channels: Pichún (above left), El Toro (above right), Tres
Arroyos (below left) and Vuelta de Zorra (below right).
37
784
785
786
787
788
109110111
/:P
RELOL
W\
+ P D[ � + %N + P D[ ! + %N
Hmax ≤ Hbk Hmax > Hbk
LW d
iam
eter
(cm
)
Hmax ≤ Hbk Hmax > Hbk
LW le
ngth
(m)
Figure 3. Box plots of LW mobility (%) for Hmax ≤ HBk and Hmax > HBk (above) and LW
mobilized dimensions (diameter, below left; length, below right) for Hmax ≤ HBk and Hmax >
HBk. The cross and line within each box indicate the mean and median values, box ends are
25th and 75th percentiles, and whiskers are the minimum and maximum values.
38
789
790
791
792
793
794
795
796
797
112113114
39
798
799
115116117
Figure 4. Water level at the gauging stations (dashed horizontal lines indicate bankfull
stage).
y = 0.0065 x - 4.262R = 0.619
0
5
10
15
20
25
30
35
40
0 500 1000 1500 2000 2500 3000 3500
LW m
obili
ty (%
)
Unit stream power for Hmax (N/m3)
y = 19.678 x - 14,514R = 0.603
0
5
10
15
20
25
30
35
40
0.0 0.5 1.0 1.5 2.0
LW m
obili
ty (%
)
Hmax/HBk
Figure 5. Relationships between LW mobility (in %) and unit stream power for Hmax
(above) and the ratio Hmax/HBk (below).
40
800
801
802
803
804
805
806
118119120
0102030405060708090
100
0 0.5 1 1.5
Freq
uece
y di
stri
butio
n (%
) of
LW p
iece
s
Llog/WBk
Moved
Tagged
0102030405060708090
100
0.0 0.5 1.0 1.5
Freq
uenc
y di
stri
butio
n (%
) of
LW p
iece
s
Dlog/HBk
Moved
Tagged
Figure 6. Frequency distribution (%) of moved and tagged LW pieces below specific
Llog/WBk (above) and Dlog/HBk (below) ratios. The percentage line of tagged LW for
41
807
808
809
810
811
812
813
814
121122123
LLog/WBk condition does not reach 100% because the 3% are elements with relative length
greater than 1.5.
0102030
Vuelta de Zorra
OutIn
Reach
Mov
ed L
W p
iece
s (N
º)
1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
18
19
20
21
22
02040
Tres Arroyos
OutIn
Reach
Mov
ed L
W p
iece
s (N
º)
1 2 3 4 5 6 7 8 9 10
11
12
13
14
15
16
17
05
10
El Toro
OutIn
Reach
Mov
ed L
W p
iece
s (N
º)
42
815
816
817
818
819
820
821
822
124125126
1 2 3 4 5 6 7 8 9 10 11 120
10
20
Pichún
OutIn
Reach
Mov
ed L
W p
iece
s (N
º)Figure 7. Number of moved LW pieces in the different reaches of the study segments.
43
823
824
127128129
Type Orientation
Log93%
Branch 3%
Full tree 1%
Rootwad3%
Parallel42%
Oblique21%
37%Orthogonal
Source Initial position
Fluvial84%
Natural7.4%
Residue7%
Bankcutting1.2%
Landslide0.4%
Bankfull line
17.9%
Bankfull channel 67.7%
Margin10.8%
Bridge1.2% Log- step
2.4%
LW aggregation before LW aggregation after
Logjam27.7%
Accumula25.1%
Single47.2%
Accumulation
Logjam22.7%
Accumula40.4%
Single37%
Accumulation
Figure 8. Type (above left), orientation to stream flow (above right), source (center left),
initial position (center right) and aggregation (below) of moved LW.
44
825
826
827
130131
y = 1.9545 x + 1.896R = 0.478
0123456789
10
0.0 0.5 1.0 1.5 2.0
Mea
n LW
leng
ht (m
)
Hmax
y = 0.884 x + 0.043R = 0.877
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.0 0.1 0.2 0.3 0.4
Mea
n LW
dia
met
er (m
)
H50% (m)
y = 0.0003 x + 0.1605R = 0.81
0.0
0.1
0.2
0.3
0.4
0.5
0 200 400 600 800 1000
Mea
n LW
dia
met
er (m
)
Unit stream power for H15% (N/m3)
Figure 9. Relationships between mean length of moved LW with Hmax (above), and mean
diameter of moved LW with H50% (center) and unit stream power for H15% (below).
45
828
829
830831
832
833
132133
Table 1. Catchment and channel characteristics in the study sites.
Catchment Location
Catchment
area
Stream
order
Length of
study segment
Nº of
reaches
Range and
mean
bankfull
width
Range and
mean
bankfull
depth
Range and
mean slope
(ha) (m) (m) (m) (m/m)
Vuelta de
Zorra
Coastal
mountains
587 3 1,557 16 8.8 – 13.4
(10.6)
0.7 – 1.4
(1.0)
0.005 – 0.09
(0.04)
Tres
Arroyos
Andes
mountains
907 3 2,070 22 6.3 – 15.0
(9.8)
0.7 – 1.8
(1.1)
0.048 – 0.18
(0.01)
El Toro Andes
mountains
1,783 3 2,188 17 8.0 – 17.5
(12.9)
1.3 – 2.5
(1.7)
0.015 – 0.11
(0.05)
Pichún Coastal
mountains
431 3 1,004 12 4.0 – 5.7
(4.8)
0.7 – 1.0
(0.8)
0.042 – 0.25
(0.1)
46
834
835
836
134135
Table 2. LW mobility and dimensions of moved wood pieces in the different study sites and periods.
2009 2010 2011 2012 2005 2006-2008 2009 2010 2011 2012 2009 2010 2011 2012 2009 2010 2011 201214.11.2008 11.11.2009 23.11.2010 13.12.2011 3.2005 12.2005 14.01.2009 15.01.2010 05.01.2011 05.12.2011 24.03.2009 16.03.2010 15.12.2010 10.01.2012 13.12.2008 13.11.2009 07.12.2010 04.01.201211.11.2009 23.11.2010 13.12.2011 22.11.2012 12.2005 14.01.2009 15.01.2010 05.01.2011 05.12.2011 09.01.2013 16.03.2010 15.12.2010 10.01.2012 27.11.2012 13.11.2009 07.12.2010 04.01.2012 03.12.2012362 days 377 days 385 days 345 days 275 days 1140 days 366 days 355 days 334 days 401 days 357 days 274 days 391 days 322 days 335 days 389 days 393 days 334 days
Moved pieces (Nº) 46 24 44 9 15 54 11 3 19 6 11 33 9 2 5 20 0 1Mobility (%) 11.6 6.0 11.3 2.4 4.7 17.0 2.2 0.6 3.8 1.2 9.4 28.2 7.3 1.5 7.9 25.3 0.0 1.3
10 - 47 11 - 58 10 - 47 10 - 70 11 - 44 15 - 53 20 - 40 25 - 42 12 - 100 18 - 35 18 - 50 17 - 76 18 - 55 28 - 55 12 - 26 11 - 39 - 13 - 13
(20) (25) (21) (22) (29) (36) (29) (31) (33) (30) (30) (40) (31) (42) (17) (18) (13)
Moved pieces not found (Nº)
5 8 15 0 7 23 4 3 6 0 7 22 3 0 0 6 0 0
Bankfull depth in the location of water level gauging station (m)
Vuelta de Zorra El Toro PichunTres Arroyos
Mobility
1.2 0.5
Range and (mean) diameter of moved pieces (cm)
1.3 0.5
47
837
838
839
840841
842
843
844
845
846
847
136137
Table 3. Percentage of tagged and moved LW for different dimensions (diameter and
length), every study catchments.
Vuelta de Zorra Tres Arroyos El Toro Pichún All
Tagged Moved Tagged Moved Tagged Moved Tagged Moved Tagged Moved
Diameter
(cm)
10-20 46.3 61.8 9.1 17.0 4.7 14.0 58.2 69.2 26.5 40.4
20-30 29.9 21.1 23.6 29.5 24.3 36.0 26.6 23.1 26.3 26.5
30-40 14.4 10.6 23.0 30.7 26.2 14.0 11.4 7.7 19.2 17.1
40-50 5.8 4.9 18.2 18.2 21.5 22.0 1.3 12.6 11.5
50-60 2.5 0.8 8.5 1.1 6.5 4.0 0.0 5.4 1.4
60-70 0.8 0.8 7.6 1.1 4.7 0.0 1.3 4.2 0.7
70-80 0.0 5.4 1.1 9.3 10.0 0.0 3.4 2.1
80-90 0.3 3.0 0.0 0.0 0.0 1.4 0.0
90-100 1.1 1.1 1.9 1.3 0.8 0.3
100-110 0.2 0.9 0.2
110-120 0.2 0.1
Length
(m)
1-4 48.6 65.9 69.2 90.9 22.4 24.0 53.2 73.1 55.4 66.9
4-7 28.9 25.2 19.1 9.1 42.1 50.0 26.6 26.9 25.7 24.7
7-10 11.6 6.5 6.9 23.4 24.0 6.3 10.4 7.0
48
848
849
138139
10-13 3.8 0.8 2.8 6.5 0.0 7.6 3.9 0.3
13-16 5.1 1.6 1.1 3.7 0.0 1.3 2.9 0.7
16-19 0.3 0.0 1.9 2.0 1.3 0.4 0.3
19-22 1.3 0.7 1.3 0.9
22-25 0.5 0.2 2.5 0.5
49
850
140141
Table 4. Tree species of moved LW (in %).
Vuelta de
Zorra
Tres Arroyos El Toro Pichún All
% % % % %
Broadleaved 90.6 96.7 100 50 91.2
Conifer 9.4 3.3 0 50 8.8
50
851
852
853
854
855
856
857
858
859
142143
Table 5. Linear relationships between a range of moved LW features as dependent variables and stream water level characteristics.
Independent variable Dependent variable Equation R SignificanceLW mobility (%) Unit stream power for Hmax y = 0.0065 x – 4.262 0.619 *
Hmax / HBk y = 19.678 x – 14.514 0.603 *Hmax / H50% y = 1.5297 x + 0.0463 0.547 *Hmax / H75% y = 1.0717 x + 0.6093 0.515 *
(Hmax – H75%) / H25% y = 1.8605 x + 1.4426 0.509 *(Hmax – H25%) / H75% y = 1.1031 x + 2.1878 0.502 *
Hmax / H25% y = 1.8247 x + 0.6093 0.489 *Hmax – H75% y = 7.2932 x + 1.6902 0.453 n.s.
Mean moved LW length (m) Hmax y = 1.9545 x + 1.896 0.478 *H15% y = 5.2844 x + 2.044 0.410 n.s.
Mean moved LW diameter (m) H50% y = 0.884 x + 0.043 0.877 *Unit stream power for H15% y = 0.0003 x + 0.1605 0.81 *Unit stream power for H25% y = 0.0003 x + 0.1673 0.771 *Unit stream power for H50% y = 0.0003 x + 0.1901 0.665 *Unit stream power for H75% y = 0.0002 x + 0.2229 0.447 n.s.
n.s. = non-significant.* = significant P ≤ 0.05
51
860861
862863864865
866
867
868
144145