- Email: [email protected]
Revisiting morphological relationships of modern source-to-sink segments as a first-order approach to scale ancient sedimentary systems
Accepted Manuscript Revisiting morphological relationships of modern source-to-sink segments as a first-order approach to scale ancient sedimentary systems
Björn Nyberg, William Helland-Hansen, Rob L. Gawthorpe, Pål Sandbakken, Christian Haug Eide, Tor Sømme, Frode HadlerJacobsen, Sture Leiknes PII: DOI: Reference:
S0037-0738(18)30158-1 doi:10.1016/j.sedgeo.2018.06.007 SEDGEO 5361
To appear in:
Sedimentary Geology
Received date: Revised date: Accepted date:
31 March 2018 7 June 2018 12 June 2018
Please cite this article as: Björn Nyberg, William Helland-Hansen, Rob L. Gawthorpe, Pål Sandbakken, Christian Haug Eide, Tor Sømme, Frode Hadler-Jacobsen, Sture Leiknes , Revisiting morphological relationships of modern source-to-sink segments as a firstorder approach to scale ancient sedimentary systems. Sedgeo (2018), doi:10.1016/ j.sedgeo.2018.06.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Revisiting Morphological Relationships of Modern Source-to-Sink
2
Segments as a First-Order Approach to Scale Ancient Sedimentary
3
Systems
4
Björn Nyberg1, William Helland-Hansen1, Rob L. Gawthorpe1, Pål Sandbakken2, Christian Haug Eide1, Tor
5
Sømme3, Frode Hadler-Jacobsen4, Sture Leiknes2
IP
1
6
T
1
CR
Department of Earth Sciences, University of Bergen, P.O. Box 7803, 5020 Bergen, Norway. 2
7
US
Statoil ASA, Sandsliveien 90, 5254 Sandsli, Norway
3
8
Statoil ASA, Martin Linges vei 33, 1364 Fornebu, Norway
4
9
AN
Statoil ASA, Arkitekt Ebbells veg 10, 7053 Ranheim, Norway
Abstract
11
Catchments provide water and sediment to downstream sedimentary systems, and these form individual source-to-
12
sink systems. Source-to-sink systems comprise adjacent linked segments, commonly hinterland catchments, alluvial-
13
and coastal plains, the continental shelf, continental slope and submarine fan. The dimensions of the catchment and
14
how it scales to downstream segments provides insight into the sedimentary and tectonic controls that influence the
15
morphology and sedimentation patterns in a basins evolution. In ancient sedimentary successions, where the
16
sedimentary routing system is buried and inaccessible for study, or fragmented due to uplift and erosion, using
17
scaling relationships can provide a powerful tool to understand the complete sedimentary system.
18
Observational data from modern sedimentary systems provide an opportunity to create morphological and
19
sedimentological scaling relationships of segments on the entire source-to-sink system. However, previous studies
20
on global modern source-to-sink systems have typically been based on a limited number of examples restricted by
21
the data available at the time and the methodology used to analyze large datasets. In the last decade, the volume and
22
quality of remotely sensed information has significantly improved so that it is now timely to revisit scaling
23
relationships of modern source-to-sink systems’ segment morphologies, and discuss the implications of those results
24
for sedimentological parameters and applicability to ancient source-to-sink systems.
AC
CE
PT
ED
M
10
1
ACCEPTED MANUSCRIPT The results of this reanalysis show that dimensions of the catchment and submarine fan segments scale internally in
26
terms of fan width, length and area. In addition, fan area scales to its largest hinterland catchment area in agreement
27
with previous studies, however, it is important to consider all catchments that contribute sediment to a basin floor
28
region. In paleogeographic settings, where individual submarine fans are difficult to tie to a single catchment, and
29
where basin floor systems are amalgamated, the contributing sediment discharge of all catchments may be
30
significant and likely influence the scale of its submarine fan.
31
Accommodation versus sediment supply in relation to relative sea level change are important controls on the
32
position of the shoreline which vary considerably from system to system over time and space, thus influencing
33
morphological relationships between source-to-sink segments. The continental shelf should therefore be viewed as a
34
transient geomorphic feature rather than a segment of a source-to-sink system. Furthermore, the continental slope
35
length should not be used to scale other segments of the source-to-sink system, which contradicts previous research.
36
The underlying tectonic and sedimentological control on the continental shelf and slope segments, in addition to the
37
subjective interpretation of their basinward boundaries, may render those segment unsuitable for scaling the
38
morphology of other segments. The study highlights both the temporal variability and complexity of controls that
39
influence the morphology and scaling relationships of internal and adjacent linked source-to-sink segments, and the
40
need to place this in a framework of both tectonic and sedimentological history.
ED
M
AN
US
CR
IP
T
25
Keywords: source-to-sink; morphology; catchment; continental shelf; continental slope; submarine fan
43
45
1.
Introduction
AC
44
CE
42
PT
41
46
Source-to-sink (S2S) systems consist of linked segments of sediment routing pathways from areas of erosion in the
47
hinterland to areas of deposition in the basin. A segment of a S2S system is a morphologically distinct component,
48
typically in the form of the catchment, continental shelf, continental slope and submarine fan (Somme et al., 2009)
49
(Fig. 1). The morphology of a segment relates to the propagation and response to updip and internal sedimentary
50
signals occurring over different temporal and spatial scales (Moore, 1969; Helland-Hansen et al., 2016).
2
ACCEPTED MANUSCRIPT 51
Understanding morphological and sedimentological characteristics of a segment and how those characteristics scale
52
to adjacent segments, can improve our knowledge of tectonic and climatic forcing on the entire S2S system.
53 In ancient sedimentary systems, preserved stratigraphic successions often represent an incomplete or partial record
55
of segments along the sediment routing pathway (Miall, 2014). A complete perspective of the entire S2S system can
56
constrain information on system scale, morphological characteristics, and tectonic and climatic controls that are
57
important to determine dimensions, volumes and timing of preserved sediments (Allen, 2008; Allen et al., 2013;
58
Helland-Hansen et al., 2016). Morphological and sedimentological characteristics of a segment and its relationship
59
to dimensions of other segments can thus provide a useful method to predict dimensions of a segment not preserved
60
in the stratigraphic record (Sømme et al., 2009). Empirical observations of modern S2S systems offers one method
61
to constrain morphological characteristics of segments, and relationships between segments, along the complete
62
sediment routing system. A significant body of research have previously characterized the morphology,
63
sedimentation characteristics and/or scaling relationships within modern catchments (e.g., Strahler, 1952; Hack,
64
1957; Milliman and Meade, 1983; Hovius, 1996; Blum and Tornqvist, 2000; Syvitski and Milliman, 2007; Romans
65
et al., 2016; Nyberg et al., 2018), continental shelves (e.g., Inman and Nordstrom, 1971; Harris and Macmillan-
66
Lawler, 2016; Nyberg and Howell, 2016), continental slopes (e.g., Harris and Whiteway, 2011; Prather et al., 2017)
67
and submarine fans (e.g., Normark, 1970; Barnes and Normark, 1985; Wetzel, 1993; Covault and Romans, 2009;
68
Covault and Graham, 2010; Shanmugam, 2016; Harris and Macmillan-Lawler, 2018). However, a limited number of
69
studies (e.g., Wetzel, 1993; Sømme et al., 2009) have explored the scaling relationships on the morphology between
70
those segments on a global and quantitative perspective.
IP
CR
US
AN
M
ED
PT
CE
AC
71
T
54
72
One of the first quantitative approaches to correlate parameters of the proximal to distal segments of modern S2S
73
systems was Normark (1970) who discussed river load controls on submarine fan morphology offshore California.
74
Wetzel (1993) expanded this research to a global perspective of 22 modern, 2 lacustrine and 5 ancient submarine fan
75
systems and showed relationships of river load and submarine fan depositional rate, length and width. Subsequent
76
work by Somme et al. (2009) based on 29 modern S2S systems, tied morphological and sedimentological
77
relationships to the entire sediment routing system from its catchment, continental shelf, continental slope and
78
submarine fan. The authors show that the continental slope length may be a particularly useful parameter to scale
3
ACCEPTED MANUSCRIPT segments that are not preserved or inaccessible in an ancient S2S system. The results have been instrumental in a
80
number of studies to provide a first-order estimate of segment scale, where limited subsurface information is
81
available, and in further developing models of S2S systems (see Helland-Hansen et al., 2016; Allen, 2017; Snedden
82
et al., 2018). However, the scaling relationships observed are based on a few modern systems and are often
83
simplistic by mainly considering S2S segment morphology to sedimentological processes as opposed to an inter-
84
disciplinary correlation between tectonics and sedimentation.
T
79
IP
85
Nearly a decade since the completion of the original work of Somme et al. (2009), new global and quantitative
87
research and datasets have improved our understanding of the physical world. Global delineations of river drainage
88
patterns have improved through high resolution digital elevation models (DEM) such as Hydrosheds (Lehner et al.,
89
2008). Milliman and Farnsworth (2011) have continued to compile observations of catchment characteristics and
90
river load attributes around the world. Nyberg and Howell (2015) delineated a global database of modern terrestrial
91
sedimentary basins which are important sites of sediment storage and transfer to the oceans (Romans et al., 2016).
92
Harris and Whiteway (2011) have mapped and analyzed morphological characteristics of 5849 large submarine
93
canyons around the world. Global bathymetry data has improved with the SRTM 30 Plus product at a 30-arc second
94
resolution (Becker et al., 2009) which led to the development of the first global geomorphic seafloor map by Harris
95
et al. (2014). Furthermore, morphological characteristics of global continental shelves and shelf breaks have been
96
analyzed by Harris and Macmillan-Lawler (2016).
PT
ED
M
AN
US
CR
86
CE
97
The volume of new data delineating and characterizing the extent of S2S segments provides an opportunity to revisit
99
morphological scaling relationships on a truly global scale and assess the implications for sediment routing systems.
AC
98
100
The aim of this paper are threefold: (1) to introduce a quantitative database describing the morphology and
101
relationships between modern S2S segments of the catchment, continental shelf, continental slope and submarine
102
fan; (2) to discuss the temporal variability of those relationships; and (3) to revisit and revise previous models of
103
S2S systems based on the findings of the present study.
104 105
2. Overview of Database, Methodology and Sources
4
ACCEPTED MANUSCRIPT A digital database of modern S2S systems is created documenting the morphology of catchments, continental
107
shelves, continental slopes and submarine fans (Fig. 1) to analyze the morphological character and scaling
108
relationships of S2S segments. The S2S database consist of 69,586 systems of which 56,196 catchments describe
109
exorheic (i.e., that drain to the ocean, or a body of water connected to an ocean) and 13,390 catchments describe
110
endorheic systems (e.g., that drain to lakes with no outflow). In addition, the morphological character of 52
111
submarine fans are described that relate to 4,843 up dip catchments that drain 23% of the non-glaciated terrestrial
112
area. The S2S database and a data statement describing the attributes of the dataset are provided as supplementary
113
information to this article. The methodology and sources of geographical data used to define the digital database of
114
S2S segment morphology are based on a set of manual and automated techniques that are described in detail below
115
(Table 1).
US
CR
IP
T
106
116
Morphological measurements of the terrestrial realm were derived from the Global Terrestrial Sink Catchment
118
(GTSC) dataset of Nyberg et al. (2018) and digital elevation models (DEM) of SRTM v4.1 (Reuter et al., 2007) and
119
GMTED2010 (Danielson and Gesch, 2011). The GTSC database was developed from hydrologically conditioned
120
river network data based on HydroSHEDS (Lehner et al., 2008) in addition to high-resolution DEMs to show the
121
distribution of rivers to modern terrestrial sinks. A terrestrial sink (or terrestrial sedimentary basin) is defined by the
122
authors as a region that is undergoing net subsidence and has shown sediment accumulation since the Holocene
123
based on the work of Nyberg and Howell (2015).
M
ED
PT
CE
124
AN
117
The parameters in the GTSC database include information on catchment area, river length, maximum relief,
126
percentage of catchment defined as a terrestrial sink, water discharge (Q) and total suspended sediment load (Qs)
127
(Table 1). The river profile gradient is defined as a rivers maximum relief over the length of its longest river
128
channel. The proportion of a catchment defined as low-lying plains with a relief <10 m/km was derived based on
129
DEMs (e.g., Reuter et al., 2007; Danielson and Gesch, 2011). A <10 m/km relief area may or may not include a
130
terrestrial sink representing a low-lying area that does not necessarily have long-term terrestrial sediment storage
131
potential but is otherwise important in influencing sediment transport to the marine realm (Sømme et al., 2009;
132
Allen et al., 2013; Romans et al., 2016; Nyberg et al., 2018) (Fig. 1).
AC
125
133 5
ACCEPTED MANUSCRIPT Delineations of continental shelves, continental slopes and submarine fans were derived from the consistent global
135
seafloor geomorphic features map (GSFM) of Harris et al. (2014). Harris et al. (2014) manually defined the shelf
136
edge and the foot of the continental slope by analyzing 100 m contour grids and 3D viewing at scales of 1:500,000
137
to mark abrupt changes in gradient. To measure shelf width from the continental shelf delineation, a series of lines
138
that represent thiessen polygons surrounding the vertices outlining the continental shelves were constructed into a
139
graph network. The shortest cumulative distance from each river mouth to its closest continental shelf edge through
140
these lines represents bathymetric profiles from river to shelf edge (e.g., Fig. 2).
141
The bathymetric profile may not strictly relate to its continental shelf width by including a significant distance along
142
a coastline waterbody (e.g., fjord, lagoon, estuary or embayment). This is a problematic issue as Harris et al. (2016)
143
noted in a global geomorphic overview of modern continental shelves. To resolve this dilemma, the current study
144
extracted the centerlines of global continental shelves from the graph network mentioned above and intersected each
145
bathymetric profile to create a centroid point on the continental shelf. The proximal distance from each point to the
146
continental landmass and to the continental shelf edge provides a reliable and objective measure of shelf width for
147
the global S2S database. The approach has the added benefit of measuring the waterbody length as the difference
148
between shelf width and the bathymetric profile length from river to shelf edge (Fig. 1; Table 1).
149
Continental slope length is defined by a simple measure from its continental shelf edge to its proximal basin floor as
150
defined by the base of the continental slope definition of Harris et al. (2014). This is appropriate given the simplicity
151
of the continental slope delineation relative to that of the terrestrial coastline. Depth of the continental shelves and
152
continental slopes are based on global bathymetric data from the SRTM 30 Plus product at a 30-arc second
153
resolution (Becker et al., 2009). Information on depth and width of the continental shelves, and depth and length of
154
the continental slopes, along each S2S bathymetric profile (e.g., Fig. 2) define the gradient of those respective
155
segments (Table 1).
156
An averaged sediment thickness along the bathymetric profile of the continental shelf and slope was sampled every
157
5 km based on global maps of Whittaker et al. (2013). The distribution of major submarine canyons were extracted
158
from the database of Harris and Whiteway (2011) who mapped 5,849 large submarine canyons showing a depth
159
range greater than 1 km and a continental shelf/continental slope incision exceeding 100 m. Morphological
160
information on submarine canyon length and frequency of canyon heads were associated to each bathymetric profile
AC
CE
PT
ED
M
AN
US
CR
IP
T
134
6
ACCEPTED MANUSCRIPT within 10 km of a canyon and assigned to a S2S system. Distance from the river mouth to each canyon is defined by
162
the distance along each bathymetric profile (Table 1).
163
Harris et al. (2014) identify global locations of submarine fans using sediment thickness maps on the continental rise
164
at the base of submarine canyons that show a concentric feature spreading seawards as observed by 100 m isobaths.
165
Based on 79 submarine fans delineations by Harris et al. (2014), manual measurements were used to define
166
submarine fan length as the furthest distance from base of the continental slope to the distal fan, and submarine fan
167
width as the widest portion parallel to the base of the continental slope (Fig. 1). Global bathymetry (e.g., Becker et
168
al., 2009) defined submarine fan depth and average submarine fan gradient was defined by bathymetric fall over its
169
submarine fan length (Table 1).
170
3. Accuracy of Morphological Measurements
US
CR
IP
T
161
AN
171
To assess the accuracy of the automated methods in defining morphological parameters of Table 1, the global S2S
173
database was compared to a set of manual observations and previous published literature. The parameters and
174
accuracy of those measurements are discussed in the following sections.
M
172
ED
175 3.1. Terrestrial Realm
177
The parameters of total suspended sediment load (Qs), river length, catchment area and catchment relief are very
178
similar in 28 modern S2S systems that are previously reported in both Somme et al. (2009) and Milliman and
179
Farnsworth (2011) (see Appendix A). The largest source of variability is associated with maximum measured relief
180
between the current study and that of Milliman and Farnsworth (2011) with a positive linear correlation at an r2 of
181
0.58. This is due to lower resolution cartographic surveys that were primarily used to build upon the Milliman and
182
Farnsworth (2011) database as opposed to high-resolution DEMs in the current study. A comparison of the present
183
study to the 28 S2S systems compiled by Sømme et al. (2009) (excluding the Danube system), indicates that where
184
similar DEM data are used, there is a strong positive 1 to 1 linear correlation with an r2 at 0.97. River profile
185
gradient is also similar in the current study to that of Somme et al. (2009) with a positive slope of 0.95 and an r2 of
186
0.87.
AC
CE
PT
176
7
ACCEPTED MANUSCRIPT 187
3.2. Marine Realm
188
The accuracy of the automated method applied in this study to estimate continental shelf width and slope length
189
shows a high degree of confidence when compared to manual measurements of 100 randomly selected S2S systems
190
(see Appendix B). Similarly, continental shelf width and continental shelf gradient are comparable for 28 S2S
191
systems between the current dataset and Somme et al. (2009) (see Appendix B).
IP
T
192
In contrast, a comparison of continental slope length and gradient between the current study and that of 28 systems
194
measured by Somme et al. (2009) shows a higher degree of variability (Fig. 3). One probable reason for the
195
discrepancy may lie in the current methodology where the present study considers the maximum slope length along
196
all catchments that contribute to a particular submarine fan (Fig. 2). The reason for this is that the current highstand
197
positioning of the largest contributing catchment, frequently misaligns with the widest portion of the continental
198
slope or the longest submarine canyon. This may reflect a relict drainage system, longshore drift of sediment or the
199
present-day disconnection of catchments to its shelf edge that spatially may shift during lowstand. Exceptions do
200
occur including the present-day Congo River where the submarine canyon incises significantly into its shelf and is
201
currently positioned directly offshore the river mouth.
ED
M
AN
US
CR
193
202
Also, whereas the continental shelf is defined by a clear shelf edge break, the base of the continental slope is more
204
gradual and reflects a combination of tectonic controls and stratigraphic overprinting, yielding different slope
205
profiles and grain size distributions (Ross et al., 1994; Praher et al., 2017). Identifying the base of the continental
206
slope is also difficult in ancient system reconstructions for similar reasons (Snedden et al. 2018). Small differences
207
in the chosen values for gradient thresholds may yield different analytical results. It does, however, highlight the
208
potential difficulty in applying the slope length for scaling the segment morphology of paleo-S2S systems. While
209
similarity between measurements on the continental slope in this and the study by Sømme et al. (2009) is low, a
210
consistent delineation of continental slope by Harris et al. (2014) used in the present study, provides an opportunity
211
to examine how this impacts morphological relationships on the broader and global perspective of S2S system
212
scaling.
AC
CE
PT
203
213
8
ACCEPTED MANUSCRIPT Morphological measurements of submarine fans are based on the delineations of Harris et al. (2014). A comparison
215
to 8 corresponding Holocene fan dimensions compiled from published literature by Somme et al. (2009) show
216
strong correlations, giving confidence in the current database (see Appendix B). Submarine fan length and area are
217
similar with a positive linear correlation and a slope of 1.08 and 1.18, respectively, though fan width measurements
218
are slightly wider in the current study as it does not separate individual fan aprons by age. This may be appropriate
219
as a first-order basis to scale paleo-S2S systems where individual fan systems will be difficult to constrain. For
220
geometries measured on the basin floor, the largest difference between the present study and that of Somme et al.
221
(2009) appear to be associated with submarine fan depth with an r2 of 0.23, but removing the Orinoco system outlier,
222
the positive linear correlation improves with an r2 of 0.70. On the basis of this discrepancy, we consider it timely to
223
employ strict, rule-based, automated methods to measure S2S segment geometries in order to more accurately derive
224
morphological relationships.
US
CR
IP
T
214
AN
225 226
4. The Morphology of Modern Source-to-Sink Systems
228
In this section, we will focus on morphological relationships that characterize the catchment, continental shelf,
229
continental slope and submarine fan segments of modern S2S systems. The S2S database contains 69,586
230
catchments, however, morphological analyses on the catchment, continental shelf and continental slope have been
231
based on only 1,778 catchments that have a total suspended sediment load (Qs) greater than 1 MT/yr. The
232
catchments therefore represent the largest sedimentary systems on earth and cover 75% of the non-glaciated
233
terrestrial landsurface. In addition, S2S systems have been analyzed based on the larger contributing area of 4,843
234
catchments (including the 1,778 S2S systems) that drain 23% of the non-glaciated landmass to 52 submarine fans
235
(Fig. 4). The S2S systems span a range of scales with a contributing catchment area from 2.99 x 102 to 7.13 x 106
236
km2 and a submarine fan area of 1.71 x 102 to 2.55 x 106 km2, providing a robust analysis of morphological scaling
237
relationships.
238
Morphological relationships have been analyzed in a categorization of main catchment tectonic regime: forearc,
239
foreland, passive margin, strike-slip or extensional setting (Nyberg et al., 2018). In addition, the current study
240
identifies epicontinental seaways as those S2S systems without a continental slope and basin floor configuration,
AC
CE
PT
ED
M
227
9
ACCEPTED MANUSCRIPT based on the work of Nyberg and Howell (2016). The tectonic regime of the continental shelf/slope margin is used
242
to distinguish active versus passive margins. For instance, S2S systems draining to the Red Sea represent passive
243
margin catchments along an active (or young) continental shelf/slope margin, in comparison to passive margin
244
catchments draining to the northern Gulf of Mexico along a passive (or mature) continental shelf/slope. Active
245
versus passive continental slope margins reflect a similar nomenclature used by previous authors (e.g., Inman and
246
Nordstrom, 1971; Sømme et al., 2009; Harris et al., 2014).
247
4.1. The Catchment
248
Figure 5 summarizes relationships between catchment area, river length, maximum relief, % of low relief area, and
249
river gradient that were defined in the original study of Sømme et al. (2009). Catchment area and river length
250
suggest a power-law relationship (Fig. 5) and catchment area (and subsequently its river length) are thought to
251
increase as a system ages by incorporating new drainage areas (Hack, 1957; Milliman and Meade, 1983). However,
252
tectonic movement in the hinterland may add to, divert and/or reroute drainages which is particularly relevant of
253
small basins along active tectonic margins. Regardless, the relationship between catchment area and river length is
254
invariant of scale and remains valid when analyzing any sub-catchment within the entire catchment area (Hack,
255
1957; Rigon et al., 1996; Nyberg et al., 2018).
256
The maximum amount of relief in a catchment does not correlate with catchment area (Fig. 5) and further
257
subdivision by tectonic regime does not improve this correlation. However, a regional based analysis of selective
258
S2S systems with similar tectonic histories (e.g., passive margins along the northern Gulf of Mexico, Amazon,
259
Congo and Niger; Fig. 4), does indicate that relief of a system relates to catchment area (Fig. 6A). By removing the
260
tectonic influence and only analyzing systems with a similar sedimentological history, the results suggest that as a
261
system ages and incorporates new drainage area, additional relief is typically captured as well. This is most apparent
262
by comparing young versus mature passive margins (Fig. 6B) that differ in basin maturity and hence subsidence
263
from crustal thinning and mantle-lithospheric thickening (Woodcock, 2004; Ingersoll, 2012). The lack of correlation
264
of these parameters on the global dataset suggests the individuality of S2S systems that are in different stages of
265
steady state between uplift and erosion. The results reaffirm the connected tectonic and sedimentological history that
266
define the morphology of source-to-sink systems. As reiterated by D’Archy and Whittaker (2014), it is often
AC
CE
PT
ED
M
AN
US
CR
IP
T
241
10
ACCEPTED MANUSCRIPT difficult to decouple tectonic and climatic influences on a system, such as relief and precipitation, given their inter
268
related nature.
269
The proposal that the proportion of low relief area (10 m/km) increases with catchment size (Sømme et al., 2009) is
270
in disagreement with the current study (Fig. 5). General observations indicate that larger systems (> ~100 x 10 3 km2)
271
have a larger low-relief area (> ~10%) whereas smaller catchments have a more variable range. This increased
272
variability likely reflects that smaller systems incorporate both intra-montane high relief regions and systems
273
originate on plains that are of lower relief, while large catchments naturally incorporate more low-relief terrain.
274
Furthermore, smaller tectonically active systems could represent a range of systems in different stages of basin
275
development whereas larger systems are typically older and may then represent later stages of basin development.
276
The topographic distribution in relief of small catchments can significantly influence the along-strike sediment load
277
contribution to the marine environment as Pechlivanidou et al. (2017) have observed in a Holocene S2S study of the
278
Serchios rift in central Greece. Similarly, Sharman et al. (2017) have concluded that a major early Cenozoic increase
279
in sediment volume delivery to the Gulf of Mexico was caused by a river capture of a catchment that was modest in
280
size but included an area of high topographic relief.
281
The tectonic history of a system controls the magnitude and distributions of uplift, and the creation of
282
accommodation by a combination of crustal thinning, mantle-lithospheric thickening, sedimentary and volcanic
283
loading, tectonic loading, subcrustal loading, asthenospheric flow and crustral densification, which are different
284
though not unique to any specific tectonic regime (Ingersoll, 2012). Combined with basin infilling history and sea
285
level fluctuations, the relation of accommodation and sediment supply control the boundaries of the low relief
286
terrestrial sink area (Blum and Tornqvist, 2000; Nyberg and Howell, 2015). Thus, the controlling factors of
287
catchment relief and low relief area vary greatly from basin to basin and the results here consequently show that it is
288
difficult to define catchment relief by morphological relationships alone (e.g., Fig. 5). Defining relief and
289
topography in eroded catchments of ancient sedimentary systems remains a source of high uncertainty in the
290
reconstruction of paleo-S2S morphology (e.g., Sømme et al., 2013; Eide et al., 2017; Sharman et al., 2017).
291
The results seem to contradict findings that the average river gradient of a S2S system profile decreases with
292
increasing catchment area (e.g., Sømme et al., 2009) (Fig. 5). This data suggests a larger low-gradient plain for
293
larger S2S systems. However, considering that catchment area is strongly correlated to river length and river profile
AC
CE
PT
ED
M
AN
US
CR
IP
T
267
11
ACCEPTED MANUSCRIPT gradient relates to fall over river length, this result reflects that larger catchments have longer rivers relative to the
295
amount of catchment relief. The river profile gradient does not consider the proportion of high and low relief areas,
296
which, as discussed above, is quite variable (Fig. 5).
297
4.2. The Continental Shelf
298
Morphological relationships specific to continental shelf margins linked to 52 submarine fans are summarized in
299
Figure 7. The results suggest that river length and thereby also catchment area and sediment discharge (e.g.,
300
Milliman and Syvitski, 1992; Syvitski and Milliman, 2007) have weak power law correlation with increasing shelf
301
width. Shelf width does not correlate with continental slope length either and correlations do not improve when
302
analyzing the larger S2S database. These results disagree with the previous studies of Somme et al. (2009), Carvajal
303
et al. (2009) and Helland-Hansen et al. (2012) by suggesting sediment discharge is not a determining control on
304
shelf width. While sediment supply may have a considerable influence on shelf growth (e.g., Carvajal et al., 2009;
305
Ingersoll, 2012), the same plate-tectonic controls that influence the variability in proportion of low-lying relief areas
306
(Fig. 4) extend to the continental shelf (e.g., Nyberg and Howell, 2016). Factors controlling subsidence and relative
307
sea level change (e.g., accommodation versus sediment supply) is a key role in the transient position of the present-
308
day shoreline and hence shelf width (e.g., Blum and Womack, 2009). This influences morphological relationships
309
between catchment length, catchment area, sediment discharge, and a measurement of continental shelf width.
310
However, absolute measurement of shelf width do differ between active versus passive continental shelf/slope
311
margins (Fig. 8). This reflects previous studies that have shown that passive margins and foreland settings have, in
312
general, larger continental shelves than short-lived extensional, strike-slip or forearc settings (e.g., Nyberg and
313
Howell, 2016; Harris and Macmillan-Lawler 2016). This would suggest that the interplay between tectonics and
314
sedimentation have a greater influence on width of the continental shelf than previously assumed by S2S models.
315
For instance, passive continental margins are characteristic of crust and lithosphere thinning and thermal cooling as
316
it matures, while later sediment loading influences its subsidence (Osmundsen and Redfield, 2011; Ingersoll, 2012).
317
Forearc settings are controlled by rates, orientation and conditions of the subsiding plate combined with sediment
318
supply to its basin and trench (Dickinson and Seely, 1997; Fuller et al., 2006; Ingersoll, 2012). Thus, comparing the
319
dynamic factors of basin creation to one aspect of its morphological character and relation to sediment supply is too
320
simplistic.
AC
CE
PT
ED
M
AN
US
CR
IP
T
294
12
ACCEPTED MANUSCRIPT 321 Gradient of the continental shelf shows a slight power law increase with increasing river gradient (Fig. 7). These
323
results suggest that steeper, smaller S2S systems along active tectonic margins (e.g., Red Sea, Taiwan, Andes) have
324
a greater tectonic influence on its continental shelf thereby potentially steepening and narrowing the continental
325
shelf thus reducing the available accommodation for the supplied sediment. In addition, catchment systems that
326
experience infrequent but high magnitude discharge events may directly bypass sediment on the continental to its
327
basin floor sink (Milliman and Syvitski, 1992; Helland-Hansen et al., 2016). This is supported by sediment thickness
328
maps that generally suggest higher sediment accumulations along long-lived passive versus short-lived active
329
continental shelf/slope margins (Woodcock, 2004; Harris and Whiteway, 2011; Ingersoll, 2012) (Fig. 8).
330
In comparison, larger catchments that contain larger terrestrial sinks have longer transient times of sediment to its
331
marine realm (Allen et al., 2013; Romans et al., 2016; Nyberg et al., 2018). This produces a steady discharge of
332
sediment to create a series of low gradient clinoform packages (Carvajal et al., 2009; Helland-Hansen et al., 2016)
333
on continental shelves with available accommodation. Furthermore, the continental shelf may be an expression of
334
sedimentary wedges produced by its fluvial counterpart during periods of sea level lowstand (Blum and Tornqvist,
335
2000). The morphology of the onshore catchment and its continental shelf in relation to the nature of base-level
336
lowering will influence the response and character of the fluvial system (Schumm, 1993; Blum and Tornqvist,
337
2000). The gradient of the continental shelf is therefore a combined expression of tectonic, sedimentological and
338
base level change.
339
Estuarine length is shown to increase with increasing continental shelf width (Fig. 7). This would be expected
340
considering wide shelves amplify tidal currents, which in turn increase restricted and funnel-shaped shoreline
341
morphologies (e.g., Cram, 1979; Ainsworth et al., 2011). Systems with large low gradient terrestrial sink may easily
342
accommodate the remobilization of coastal sediments into funneled shaped shorelines, in comparison to steeper
343
gradient catchment systems (Nyberg and Howell, 2016). A longer estuary will increase the potential sediment transit
344
times along the sediment routing system to its submarine fan (e.g., Meade 1969; Kineke et al., 1996; Goodbred and
345
Kuehl, 1999). This is especially relevant during sea level highstands and might help explain high sediment trapping
346
efficiencies on present-day continental shelf (Kineke et al., 1996; Goodbred and Kuehl, 1999). For instance, Bostock
347
et al. (2007) calculated that more than 50% of the total annual suspended sediment load of the Fitzroy River in
AC
CE
PT
ED
M
AN
US
CR
IP
T
322
13
ACCEPTED MANUSCRIPT Queensland, Australia, is trapped in lower floodplains and estuaries. The Ganges/Brahmaputra river system traps
349
more than 30% of its total annual sediment load in its lower floodplains and estuaries (Goodbred and Kuehl, 1999).
350
4.3. The Continental Slope
351
Morphological parameters measured on the continental slope show a lack of any statistical correlation to other
352
segments of the S2S system (Fig. 9). Continental slope length is not empirically related to river length, river
353
gradient, catchment area or shelf width. The lack of correlation between morphological parameters, and hence
354
sedimentation rates and continental slope length, suggests a more complex interplay between plate tectonics and
355
sedimentation history than previously asserted in S2S models (e.g., Sømme et al., 2009; Helland-Hansen et al.,
356
2016). In a review of modern and paleo-continental slopes, Prather et al. (2017) notes the wide range of controls on
357
morphology of continental slope profiles including: sediment progradational patterns, erosional continental slope
358
margins, rapid sea level fluctuations, tectonic over-steepening and alternating siliclastics versus carbonate
359
deposition. The gradient of the continental slopes varies considerably from 1˚ to 10˚ and the base-of-slope transition
360
to the basin floor is often imprecisely defined (Prather et al., 2017). Our current study therefore discourages use of
361
the continental slope dimensions for scalingother segments.
ED
M
AN
US
CR
IP
T
348
362
Interestingly, absolute values of continental slope length and depth do not seem to differ significantly either by
364
tectonic regime or active versus passive continental shelf/slope margins with the exception of forearc settings (Fig.
365
10). Similar slope lengths and depths yield comparable gradients though the destruction and reworking of
366
sedimentary wedges along subducting zones of forearc systems considerably deepen and steepen those continental
367
slopes (Fig. 10). The lack of a clear morphological distinction of continental slope length may then reflect the
368
difficulty in defining the transition between slope and rise (Fig. 3).
AC
CE
PT
363
369 370
Harris and Whiteway (2011) showed that submarine canyons that contribute to the abyssal plains have higher
371
frequency of canyons and shorter lengths along active continental margins compared to canyons on passive margins.
372
Extending this analysis in the current study by active versus passive continental shelf/slope margin for submarine
373
canyons contributing to the 52 submarine fans suggest comparable results (Fig. 11). Active margins have a higher
14
ACCEPTED MANUSCRIPT frequency of canyon heads in comparison to passive continental slopes. The lengths of those canyons are also on
375
average shorter on active versus passive continental slope margins when comparing S2S systems of similar
376
catchment tectonics. Longer submarine canyons along passive continental slope margins are typically incised on the
377
continental shelf (Harris and Whiteway, 2011) and run the length of its continental slope and rise. More active
378
continental slopes show shorter submarine canyons on steeper slopes. However, active faulting, particularly during
379
early rifting, may divert and lengthen submarine canyons along relay ramps (e.g., Athmer and Luthi, 2011), which
380
may explain the considerable variability in submarine canyon lengths. Given that slope lengths are comparable
381
between active versus passive continental slope margins (Fig. 10) while canyon lengths differ, the data suggest that
382
canyons along active margins may not extend the length of its corresponding continental slope (Fig. 11).
CR
IP
T
374
US
383 4.5. The Basin Floor
385
Measurements in this study show comparable first-order relationships of submarine fan dimensions to catchment
386
properties as the ones proposed by Wetzel (1993) and Sømme et al. (2009). Submarine fan width and length show a
387
strong power law correlation (Fig. 12), in agreement with the original work by Wetzel (1993). The length of the
388
submarine fan increases by a power law correlation to the length of the longest river channels of contributing
389
catchments (e.g., Fig. 2) and submarine fan area shows a power law relationship with increasing contributing
390
catchment area (Fig. 12). In an analysis of 40 Paleocene to Pleistocene deepwater fan systems in the Gulf of Mexico,
391
Snedden et al. (2018), showed that submarine fan dimensions mapped from seismic reflections and well data
392
correlate to paleogeogaphic reconstructions of onshore catchment area, in agreement with modern empirical scaling
393
relationships. Given the strong correlation between catchment area and suspended sediment load (Syvitski and
394
Milliman|, 2007), submarine fan area expands with increasing sediment load of a system (Qs) (Fig. 12). The distal
395
depths of submarine fans do not differ by tectonic setting, however, longer fans associated with larger S2S systems
396
with similar fan depths result in a power law decrease in fan gradient with increasing fan area (Fig. 12).
397
An important observation of the current study shows proximal to distal S2S morphological and thus
398
sedimentological correlations should relate the basin floor to its broader contributing terrestrial region (Fig. 2),
399
rather than its largest catchment, as originally proposed by Wetzel (1993). In certain circumstances, the total
400
contributing catchment area can be significant. For example, during the current sea level highstand, 46% of the
AC
CE
PT
ED
M
AN
384
15
ACCEPTED MANUSCRIPT catchment area contributing to the Bengal Fan and 16% of the estimated sediment load (calculated using the
402
BQART approach of Milliman and Syvitski 2007) that reaches the Bengal Fan comes from catchments other than
403
that of the Ganges/Brahmaputra Rivers (Fig. 2). During lower than at present sea levels, paleogeographic
404
reconstructions suggest new drainage patterns can similarly increase sediment contribution to a basin floor region
405
from multiple river drainage systems rather than a single coalesced river. For instance, Holocene sea level
406
reconstructions of the Sunda Shelf have shown that lower sea levels than at present supported four large river
407
systems contributing to the South China Sea (Voris, 2000). Calculating the merger of catchments at the continental
408
shelf edge based on present-day bathymetry of the Sunda Shelf, indicates that over 30% of the contributing
409
catchment area to the South China Sea is derived from catchments other than the Mekong River. The increased
410
contributing catchment area will likely increase the sediment input to the basin floor. U–Pb and Pb–Pb dating of
411
detrital zircons can provide a valuable tool to assess the different potential source regions of the contributing
412
catchment area (e.g., Blum and Pecha, 2014; Sharman et al., 2015; Helland-Hansen et al., 2016).
413
5. The Source-to-Sink Profile
414
The morphology of each source-to-sink segment defines the topographic and bathymetric profiles of 52 S2S systems
415
stretching from the highest hinterland point to the most distal point of the submarine fan (Figure 13; see Fig. 4 for
416
location; see also supplementary kmz). The data show the variable morphological nature of S2S systems across the
417
spectrum of tectonic regimes. Visually, the relative river length (i.e., terrestrial sink, < 10/km relief zone and source
418
region) increases with increasing catchment area while the relative continental shelf width and slope length
419
decreases. In contrast, smaller systems that are common along tectonically-active margins are represented by a
420
proportionally larger continental shelf and slope length. These results adhere to the earlier findings showing a strong
421
power law correlation of increasing catchment area and river length (Fig. 5) whereas absolute continental shelf (Fig.
422
7) and continental slope (Fig. 9) dimensions do not significantly change with the size of the catchment.
423
These observations are reiterated by mapping the proportional distance (0-100%) of the river, continental shelf,
424
continental slope and submarine fan S2S profile by catchment area (Fig. 14). The negative power-law correlations
425
between catchment area and proportion of continental shelf and continental slope length suggest that smaller
426
systems have proportionally a longer distance to transport sediment from its river mouth to its basin floor (i.e., a
427
longer relative continental shelf and continental slope component). On the other hand, the relative proportion of the
AC
CE
PT
ED
M
AN
US
CR
IP
T
401
16
ACCEPTED MANUSCRIPT submarine fan length to catchment area is more varied despite strong absolute correlations of contributing catchment
429
area and fan dimensions (Fig. 12). This suggests that the relative length of submarine fans in unconfined basin floors
430
do not proportionally correlate to the length of the entire S2S system as the submarine fan dimensions only show a
431
strong positive correlation to the contributing catchment region (Fig. 12) rather than continental shelf width or slope
432
length (Fig. 9).
433
5.1 The Transient Source-to-Sink Segment Morphology
434
Relative sea level fluctuations on timescales less than the lifespan of most basins (< 106 yrs) may significantly
435
influence the S2S topographic/bathymetric profile and morphology. During a relative sea level fall, the contributing
436
catchment area will expand, resulting in an increase in river length, low-relief terrestrial area, sediment discharge
437
and sediment bypass on the continental shelf (e.g., Schumm, 1993; Blum and Tornqvist, 2000) (Fig. 15). River
438
profile gradient will decrease as the length of the river increases relative to the amount of additional relief obtained
439
from any sea level fall. The volume of sediment stored on the terrestrial realm may thus increase due to the increase
440
in low-lying catchment area, while sediment stored on the continental shelf will decrease. This will change the
441
proportion between semi-transient terrestrial sink and continental shelf sediment storage.
442
The amount of sediment stored within the terrestrial realm and continental shelf can be significant. For instance,
443
present-day storage within the Ganges-Brahmaputra delta/floodplain may be as high as 30% of its total suspended
444
sediment load, while ~60% of sediment that reaches the Bengal Sea is stored on its continental shelf and a remaining
445
~40% reaches the Swatch of No Ground canyon (Goodbred and Kuehl, 1999). Aalto et al. (2006) proposed that 50%
446
of the total Andean sediment budget is stored in the terrestrial Amazon basin while as much as 100% of sediment
447
that reaches the Atlantic is stored on its continental shelf (Kineke et al., 1996). In consequence, early work of
448
Shanmugam et al., 1985 and Posamentier et al., 1991 suggested that submarine fans are principally derived during
449
periods of sea level lowstands when a river may extend to the continental shelf edge and deliver sediment directly to
450
the basin floor.
451
Observations of modern rivers support a low stand increase in sediment load due to an increase in catchment area
452
(e.g., Milliman and Syvitski, 1992; Syvitski and Milliman, 2007) despite the variability of low relief area (Fig. 5).
453
This suggests that while terrestrial sediment storage may increase during a relative sea level fall, a larger catchment
AC
CE
PT
ED
M
AN
US
CR
IP
T
428
17
ACCEPTED MANUSCRIPT area will typically increase annual sediment load to the marine environment. A decrease in sea level can furthermore
455
promote increased catchment denudation upstream as the river tries to restore an equilibrium caused by local
456
steepening from channel incision and avulsion (Schumm, 1993; Blum and Tornqvist, 2000). However, more recent
457
studies suggest that submarine fans can build regardless of sea level stand and are defined by high magnitude
458
tectonic and climatic perturbations transporting new, and remobilizing previously stored sediments (e.g., Covault
459
and Graham, 2010; Simpson and Castelltort, 2012; Shanmugam, 2016). A larger low stand terrestrial sink may then
460
increase the amount of terrestrial sediments available for remobilization and directly connect the system to the
461
continental shelf edge but not necessarily provide the millennial scale events that are important for the stratigraphic
462
record.
463
S2S systems with large terrestrial sinks and shallow wide continental shelves, especially along epicontinental
464
seaways (e.g., Gulf of Thailand, Gulf of Carpanteria, Adriatic Sea), would experience the greatest morphological
465
change because of sea level change (Fig. 15). In contrast, short and steep systems with higher gradient catchments
466
and continental shelves would only marginally alter catchment area, river length and continental shelf width with a
467
similar sea level change (Fig. 15). Hence, the continental shelf may be better considered as a transient geomorphic
468
feature rather than a segment of a S2S system.
469
Tectonic activity that typically occurs within the source region operates on a variety of temporal and spatial scales,
470
may alter the relative sea level, either connect or disconnect drainage area, while relief of the system may increase or
471
decrease to reflect that tectonic history (e.g., Pechlivanidou et al., 2017; Sharman et al., 2017). Morphological
472
relationships in the current study suggest that variations in catchment area will influence the distal submarine fan
473
area (Fig. 12) and sediment discharge will change given the strong influence of catchment area and relief on river
474
sediment load (Syvitski and Milliman, 2007). Small tectonically active systems (e.g., strike-slip, extensional) would
475
be most susceptible to these changes due to active faulting in these short-lived systems. In addition, the region of
476
tectonic influence may be proportionally greater in small systems compared to larger passive/foreland systems
477
thereby greatly influencing the morphology of those S2S system (Fig. 15).
AC
CE
PT
ED
M
AN
US
CR
IP
T
454
478 479
18
ACCEPTED MANUSCRIPT 480 6. The Source-to-Sink Model
482
Helland-Hansen et al. (2016) proposed a simplified three end members of S2S systems that drain to the ocean:
483
“steep, short and deep”, “wide and deep” and “wide and shallow” systems. Here we expand and revise the
484
classification based on modern morphological characteristics and relationships presented in the current study.
485
Furthermore, we discuss the change in S2S segment morphology on timescales influencing a sedimentary basin
486
infill history (>104 to <106 yrs).
487
The classification of S2S systems below considers individual catchment - continental shelf - continental slope and
488
basin floor configurations. In studying S2S systems, it is imperative to consider the larger contributing catchment
489
area (e.g., Fig. 12) with the understanding that downstream segments may represent an amalgamation of different
490
source regions. The larger contributing catchment area may subsequently represent a combination of different S2S
491
models. For example, the western part of the Bay of Bengal fan (Fig. 2) is fed by a number of wide and deep passive
492
tectonic regimes whereas the eastern parts are increasingly represented by wide and deep foreland, and steep, short
493
and narrow forearc tectonic regime S2S systems. The variability in S2S profiles contributing to one basin-floor fan
494
will influence the distribution and mode of sediment transport to the submarine fan.
495
A general rule, as with using any modern analog, is to view the morphological relationships and characteristics as a
496
model of S2S system scaling and apply that knowledge to system-specific conditions. Furthermore, it is important to
497
note that some of the morphological relationships would potentially break down in systems of topographically
498
confined basin floors, such as along subduction zones (e.g., Andes), of terminal oceanic basins (e.g., Mediterranean
499
Sea), or salt-confined basins (e.g., Plio-Pleistocene, Gulf of Mexico), where the submarine fan is being restricted or
500
destroyed.
501
6.1 Steep, Short and Deep Systems
502
Steep, short and deep systems are found on active margins and contain steep and short catchments bordering a basin
503
floor (Helland-Hansen et al., 2016) of a young or contracting ocean basin (Harris and Macmillan-Lawler, 2018). All
504
steep, short and deep systems are characterized by a high sediment yield (Milliman and Syvitski, 1992) with narrow
505
shelves (<35 km; Nyberg and Howell, 2016), shallow shelf breaks (<180 m; Harris and Macmillan-Lawler, 2016),
AC
CE
PT
ED
M
AN
US
CR
IP
T
481
19
ACCEPTED MANUSCRIPT medium long continental slope lengths (~60 km) and deep basin floors (< 3km; Fig. 16A). Steep, short and deep S2S
507
systems of forearc tectonic regimes are distinguished by small terrestrial sinks (Nyberg and Howell, 2015) fed by
508
few but large source catchments relative to the total source region area (Nyberg et al., 2018). The S2S systems are
509
long-lived (~107 - 108 yrs; Woodcock, 2004; Ingersoll 2012), fed by numerous and steep submarine canyons that are
510
shelf and slope incised (e.g., Andes; Hagen et al., 1994; Harris and Whiteway, 2011) (Fig. 11) and associated with a
511
basin floor with limited preservation potential (Harris and Macmillan-Lawler, 2018). The areally limited terrestrial
512
sinks suggests that these systems will likely deposit sediment directly on the continental shelves. A lack of sediment
513
thickness on the continental shelf and slope (Fig. 8) suggests either an immaturity of those systems and/or the
514
destruction of bypassed sediments on the basin floor.
515
Steep, short and deep systems along oceanic transform margins including strike-slip and extensional tectonic
516
regimes are short lived (< 107 yrs; Woodcock; 2004; Ingersoll, 2012) although the young basin floor may have
517
longer-term preservation (> 107 yrs; Harris and Macmillan-Lawler, 2018). These systems are characterized by small
518
terrestrial sinks fed by few but large source catchment regions relative to the total source region area (Nyberg et al.,
519
2018) that transport sediment to the basin floor through several steep but short submarine canyons (Figs. 11, 16B).
520
The shorter terrestrial sediment transit times in the areally insignificant terrestrial sinks (Nyberg and Howell, 2015)
521
will likely bypass sediment to the marine realm as suggested by thinner sediment packages (Fig. 8). The distal parts
522
of the S2S system will furthermore experience sediment routing pathway avulsions as the systems develops and
523
matures (e.g., Red Sea; Gulf of California; Ingersoll, 2012).
524
Steep and short endorheic systems are furthermore present in continental extensional and strike-slip tectonic
525
regimes, characterized as short lived (< 107 yrs) given the continual tectonic destruction and creation of new
526
accommodation (Gawthorpe and Leeder, 2000; Ingersoll, 2012). In this circumstance, the system will terminate on a
527
relatively larger terrestrial sink (Nyberg and Howell, 2015) fed by numerous yet smaller source catchment regions
528
when comparing to the entire basin drainage (Nyberg et al., 2018).
529
On shorter timescales (< 106 yrs), the primary control on morphological change of steep, short and deep systems is
530
tectonic activity where faulting may influence a large proportion of the overall S2S system to change relief and
531
catchment area (Fig. 15). Tectonics furthermore play a key role in the sediment routing pathways of these systems
532
by controlling initiation, diversion and potentially termination of sediment transport to the basin floor (Gawthorpe
AC
CE
PT
ED
M
AN
US
CR
IP
T
506
20
ACCEPTED MANUSCRIPT and Leeder, 2000; Athmer and Luthi, 2012). In comparison, relative sea level change on narrow and steep
534
continental shelves with smaller low-lying terrestrial sinks will not significantly alter the scale and morphology of
535
the S2S system, as illustrated in Figure 15B.
536
6.2 Wide and Deep Systems
537
Wide and deep systems are typical of the mature remnants of pre-existing rifted margins (Kingston et al., 1983;
538
Osmundsen and Redfield, 2011) or at margins of convergent plate tectonic boundaries (Ingersoll, 2012). These S2S
539
systems differ in that they are long-lived (> 108 yrs; Woodcock, 2004) and develop large catchments producing
540
significant sediment discharge (Syvitski and Milliman, 2007), medium-wide continental shelves (>50 km; Nyberg
541
and Howell, 2016) with deep shelf breaks (>200 m; Harris and Macmillan-Lawler, 2016), medium continental slope
542
lengths (~60 km), few but long submarine canyons (Harris and Whiteway, 2011) (Fig. 11) and deep basin floors
543
(~3.9 km; Fig. 16C). Foreland tectonic regimes (Fig. 16C) are portrayed by a relatively large source region, a small
544
low gradient sediment bypass zone (~30% of total catchment area < 10m/km) that includes a large terrestrial sink
545
fed by numerous smaller source catchments (Nyberg and Howell, 2015; Nyberg et al., 2018). In contrast, passive
546
margins (Fig. 16D) have a continental hinterland composed of a relatively small mountainous source region, a large,
547
low gradient sediment bypass zone (~48% of total catchment area <10 m/km) that includes a medium-sized
548
terrestrial sink fed by few but large source catchments (Nyberg et al., 2018).
549
Relative sea level change will have a greater influence on catchment area, river length and averaged river gradient
550
profile of wide and deep systems given the larger terrestrial sinks and wider continental shelves. The amount of
551
morphological change will be less significant for larger systems where the relative length of its continental shelf
552
may be insignificant compared to the length of the entire system size (Figs. 13, 15, 16). Tectonic activity on shorter
553
timescales (< 106 yrs) resulting in small relief variations or source catchment diversions would not significantly
554
change the overall morphology of wide and deep S2S systems (Figs. 13, 15).
555
6.3 Wide and Shallow Systems
556
Wide and shallow systems are typical of epicontinental seaways without a traditional shelf-slope-basin floor-fan
557
configuration. These systems are long-lived, have large catchments and wide shelves of low gradient bathymetry
558
(Harris and Macmillan-Lawler, 2016). Intracratonic systems along high-latitude regions, (e.g., Hudson Bay, North
AC
CE
PT
ED
M
AN
US
CR
IP
T
533
21
ACCEPTED MANUSCRIPT Sea, Arctic Sea) tend to have a low relief hinterland, no terrestrial sink (Nyberg and Howell, 2015), deep glaciated
560
troughs with shelf breaks as deep as 360 m (Harris and Macmillan-Lawler, 2016) and low sediment discharge
561
systems (Milliman and Farnsworth, 2011). The lack of sediment discharge, and small terrestrial sinks, is in part
562
related to present-day climate of high-latitude regions that show significant isostatic uplift since the last glacial
563
maximum (Sella et al., 2007). Low to mid-latitude intracratonic systems (e.g., Gulf of Carpanteria) along
564
epicontinental seaways have a similar morphological character without major glaciated troughs (Harris et al., 2014),
565
medium sized terrestrial sink (Nyberg and Howell, 2015) and typically a higher sediment discharge related to
566
warmer climates (Nyberg et al., 2018). In comparison, present-day configuration of foreland basins within
567
epicontinental seaways tend to have higher catchment relief, higher sediment discharge, larger terrestrial sinks with
568
variable source morphology relating to constricted catchments profiles. Their continental shelves are low gradient
569
and shallow with shelf breaks at approximately 200 m (Harris and Macmillan-Lawler, 2016) with thicker sediment
570
accumulations.
571
Endorheic systems are similar to their marine counterpart. Intracratonic systems show a low relief hinterland but
572
with a medium sized terrestrial sink component sourced by numerous small source catchments. By contrast, large
573
foreland type endoheric systems have a higher relief but are often associated with a large terrestrial sink component
574
fed by small, numerous source catchments (Nyberg et al., 2018). In many circumstance, the lithospheric loading of
575
the tectonic plates fades from the hinterland and the distal part of these systems are characteristic of an intracratonic
576
settings (DeCelles and Giles, 1996).
577
The morphology of wide and shallow systems along epicontinental seaways will fluctuate the most, subject to sea
578
level change. The shallow continental shelves (Harris and Macmillan-Lawler, 2016) that sometimes span hundreds
579
of kilometers (e.g., Gulf of Carpentaria, Gulf of Thailand) can significantly alter the catchment area, river length
580
river profile gradient from sea level change (Fig. 12). This reflects the dynamic nature of the continental shelf. As
581
with wide and deep systems, regional tectonic uplift (< 106 yrs) may have less influence given the size of the
582
absolute contributing catchment area of these S2S systems (Fig. 15).
583
7. Conclusions
AC
CE
PT
ED
M
AN
US
CR
IP
T
559
22
ACCEPTED MANUSCRIPT The holistic S2S approach is an attempt to relate the complexity of trigger and response of sedimentary processes
585
from the complete sediment routing perspective. By studying the scaling relationships between modern S2S
586
segments that have formed on geological timescales (>106 yrs), the long term influence of sedimentary processes on
587
the morphology of sedimentary systems can be understood. Scaling relationships of S2S segments would provide a
588
useful, first-order assessment to support paleogeographic reconstruction of ancient sedimentary systems where the
589
complete sediment routing system is not preserved. A restored paleogeography would thus offer insight into the
590
sedimentary processes and volume budgets contributing to the stratigraphic record. It has been the purpose of this
591
article to revisit the morphology and scaling relationships of modern S2S segments and discuss the implications for
592
scaling segments in ancient sedimentary systems. The results of our study have provided new insight into S2S
593
scaling relationships and sediment routing systems that imply:
CR
IP
1.
US
594
T
584
The properties of catchment area and river length scale internally for the catchment segment. However, characteristics of catchment topography, such as maximum relief and proportion of low relief area, is more
596
variable and system specific, which will introduce uncertainties in estimations of S2S sediment budgets. In
597
addition, the transient nature of the continental shelf will influence the dimensions of the catchment and
598
scale of downstream segments. In application to ancient S2S systems, scaling relationships are defined for
599
systems formed over geological timescales (>106 years) that incorporate higher frequency fluctuations in
600
catchment morphology and sedimentological characteristics.
M
ED
2.
PT
601
AN
595
Continental slope length is difficult to measure objectively due to the gradational transition to the abyssal plain. It is a sedimentological and tectonically controlled boundary that shows considerable variability.
603
Therefore, the use of the continental slope morphology to scale segments of a S2S systems should be
604
exercised with extreme caution. 3.
AC
605
CE
602
The catchment and submarine fan segments of the S2S system correlate for many morphological
606
parameters, suggesting the larger scale fan dimensions relate primarily to sediment transport to the basin
607
floor. The catchment size and its sediment transport capacity may then be a useful first-order parameter to
608
scale proximal to distal S2S segments. First-order assessment of submarine fan dimensions are a
609
contribution from multiple catchments to one basin floor region that are in a continuous state of coalescing
610
and dividing associated with the transient shoreline position. Thus, it is imperative that S2S studies
611
consider the larger region of contributing catchments which supply sediment to a specific submarine fan.
23
ACCEPTED MANUSCRIPT 612
4.
Models characterizing S2S systems as “Steep, short and deep”, “Wide and deep” and “Wide and shallow” show distinct morphological characteristics of its catchment, continental shelf, continental slope and
614
submarine fan by different tectonic regimes. However, the distinctions may not necessarily suggest scaling
615
relationship of morphological dimensions within a segment or between adjacent linked S2S segments. This
616
is particularly important along tectonically bounded S2S features of the continental shelf and continental
617
slope that are not appropriate to scale based on the morphology of other segments. Similar continental shelf
618
widths and continental slope lengths across a range of system sizes suggests a proportionally longer marine
619
sediment transport zone to the basin floor along sediment routing systems with decreasing catchment size.
IP
5.
CR
620
T
613
In the reconstruction of partially preserved S2S segments, morphological and sedimentological scaling relationships should be viewed as a first-order assessment of the proximal and distal portions of a sediment
622
routing system. The relationships may apply to any sea level stand of non-topographically confined basins
623
supplemented by other available analytical techniques in a holistic approach to improve system specific
624
conditions. For instance, detrital zircon analyses can improve knowledge on source region and volume,
625
while interpretation of depositional environment and tectonic reconstruction can further constrain probable
626
dimensions of S2S segments.
ED
M
AN
US
621
Future work needs to concentrate on the role of plate tectonics and basin development on S2S morphology and
628
scaling relationships. These issues appear to be a major control on the observed variability in the present study.
629
Forward modelling of landscapes and sedimentary systems may be an interesting approach to integrate S2S concepts
630
into a process-based framework that considers both sedimentation and tectonic influences. In conclusion, the
631
correlations (and sometimes low correlations), between morphology of S2S segments shows the significant amount
632
of research that remains in order to improve our understanding of sediment routing systems.
633
Acknowledgements
634
Statoil are thanked for supporting the main author under the Spatial-Temporal Reconstruction of Basin Fills project
635
(no. 810127) and RLG acknowledges support by VISTA. Reviewer John Snedden and editor Jasper Knight are
636
thanked for constructive comments that greatly improved this manuscript.
AC
CE
PT
627
637 24
ACCEPTED MANUSCRIPT 638
Appendix A - A comparison of parameters gathered on the catchment segment of modern S2S systems in the
639
current study versus previously reported databases of Sømme et al. (2009) and Milliman and Farnsworth (2011).
640
Sømme - Sømme et al. (2009), M & F (2011) - Milliman and Farnsworth (2011)
641 Appendix B - A comparison of parameters gathered on the continental shelf, continental slope and basin floor
643
segments of modern S2S systems in the current study versus manual observed measurements and previously
644
reported dimensions in Sømme et al. (2009). Sømme - Sømme et al. (2009)
CR
IP
T
642
US
645 References
647
Aalto, R., Dunne, T., Guyot, J.L., 2006. Geomorphic Controls on Andean Denudation Rates. Journal of Geology
649
114, 85–99.
M
648
AN
646
Ainsworth, R., Vakarelov, B., Nanson, R., 2011. Dynamic Spatial and temporal prediction of changes in depositional processes on clastic shorelines: Towards improved subsurface uncertainty reduction and
651
management. American Association of Petroleum Geologists Bulletin 95, 267–297. Allen, P.A., 2008. Time scales of tectonic landscapes and their sediment routing systems. In: Gallagher, K., Jones,
PT
652
ED
650
S.J., Wainwright, J. (Eds.), Landscape Evolution: Denudation, Climate and Tectonics Over Different Time
654
and Space Scales, Geological Society of London Special Publications, 296, pp. 7-28.
656 657
Allen, P.A., 2017. Sediment Routing Systems: The Fate of Sediment from Source to Sink. Cambridge University
AC
655
CE
653
Press, Cambridge.
Allen, P.A., Armitage, J.J., Carter, A., Duller, R.A., Michael, N.A., Sinclair, H.D., Whitchurch, A.L., Whittaker,
658
A.C., 2013. The Qs problem: Sediment volumetric balance of proximal foreland basin systems.
659
Sedimentology 60, 102–130.
660 661
Amante, C., Eakins, B., 2009. ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis. NOAA Techical Memo, NESDIS NGDC-24, 19 pp.
25
ACCEPTED MANUSCRIPT 662 663 664
Athmer, W., Luthi, S.M., 2011. The effect of relay ramps on sediment routes and deposition: A review. Sedimentary Geology 242, 1–17. Barnes N.E., Normark W.R., 1985. Diagnostic Parameters for Comparing Modern Submarine Fans and Ancient Turbidite Systems. In: Bouma A.H., Normark W.R., Barnes N.E. (Eds.), Submarine Fans and Related
666
Turbidite Systems. Frontiers in Sedimentary Geology pp. 13-14. Becker, J.J., Sandwell, D.T., Smith, W.H.F., Braud, J., Binder, B., Depner, J., Fabre, D., Factor, J., Ingalls, S., Kim,
IP
667
T
665
S.H., Ladner, R., Marks, K., Nelson, S., Pharaoh, A., Trimmer, R., von Rosenberg, J., Wallace, G.,
669
Weatherall, P., 2009. Global Bathymetry and Elevation Data at 30 Arc Seconds Resolution: SRTM30_PLUS.
670
Marine Geodesy 32, 355–371.
US
671
CR
668
Blum, M.D., Hattier-Womack, J., 2009. Climate change, sea-level change, and fluvial sediment supply to deepwater depositional systems. In: Kneller, B., Martinsen, O.J., McCaffrey, B. (Eds.), Externernal Controls on Deep-
673
Water Depositonal Systems. Society for Sedimentary Geology 92, p15–39.
676
M
675
Blum, M., Pecha, M., 2014. Mid-cretaceous to paleocene North American drainage reorganization from detrital zircons. Geology 42, 607–610.
ED
674
AN
672
Blum, M.D., Tornqvist, T.E., 2000. Fluvial responses to climate and sea-level change : a review and look forward. Sedimentology 47, 2–48.Bostock, H.C., Brooke, B.P., Ryan, D.A., Hancock, G., Pietsch, T., Packett, R.,
678
Harle, K., 2007. Holocene and modern sediment storage in the subtropical macrotidal Fitzroy River estuary,
679
Southeast Queensland, Australia. Sedimentary Geology 201, 321–340.
681 682 683 684 685 686
CE
Carvajal, C., Steel, R., Petter, A., 2009. Sediment supply: The main driver of shelf-margin growth. Earth-Science
AC
680
PT
677
Reviews 96, 221–248. Covault, J.A., Graham, S.A., 2010. Submarine fans at all sea-level stands: Tectono-morphologic and climatic controls on terrigenous sediment delivery to the deep sea. Geology 38, 939–942. Covault, J.A., Romans, B.W., 2009. Growth patterns of deep-sea fans revisited: Turbidite-system morphology in confined basins, examples from the California Borderland Marine Geology. 265, 51–66. Cram, J.M., 1979. The Influence of Continental Shelf Width on Tidal Range: Paleoceanographic Implications.
26
ACCEPTED MANUSCRIPT
688 689 690 691
Journal of Geology. 87, 441–447. Danielson, J.J., Gesch, D.B., 2011. Global multi-resolution terrain elevation data 2010 (GMTED2010). U.S. Geological Survey Open-File Report 2011–107, pp. 26. D’Arcy, M., Whittaker, A.C., 2014. Geomorphic constraints on landscape sensitivity to climate in tectonically active areas. Geomorphology 204, 366–381.
T
687
DeCelles, P.G., Giles, K.A., 1996. Foreland basin systems. Basin Research 8, 105–123.
693
Dickinson, W.R., and Seely, D.R., 1979. Structure and stratigraphy of forearc regions. American Association of
695
CR
Petroleum Geologists Bulletin 63, 2–31.
US
694
IP
692
Eide, C.H., Klausen, T.G., Katkov, D., Suslova, A.A., Helland-Hansen, W., 2017. Linking an Early Triassic delta to antecedent topography: Source-to-sink study of the southwestern Barents Sea margin. Geological Society of
697
Ammerica Bulletin 130, 263-283.
701 702
M
ED
700
zone earthquakes. Geology 34, 65–68.
Gawthorpe, R.L., Leeder, M.R., 2000. Tectono-sedimentary evolution of active extensional basins. Basin Research 12, 195–218.
PT
699
Fuller, C.W., Willett, S.D., Brandon, M.T., 2006. Formation of forearc basins and their influence on subduction
Goodbred, S.L., Kuehl, S.A., 1999. Holocene and modern sediment budgets for the Ganges-Brahmaputra river
CE
698
AN
696
system: Evidence for highstand dispersal to flood-plain, shelf, and deep-sea depocenters. Geology 27, 559–
704
562.
705 706 707 708 709 710
AC
703
Hack, J.T., 1957. Studies of longitudinal stream profiles in Virginia and Maryland. US Geological Survey of Professional Papers., 294-B, 1. Hagen, R.A., Bergersen, D.D., Moberly, R., Coulbourn, W.T., 1994. Morphology of a large meandering submarine canyon system on the Peru-Chile forearc. Marine Geology 119, 7–38. Harris, P.T., Macmillan-Lawler, M., 2016. Global Overview of Continental Shelf Geomorphology Based on the SRTM30_ LUS 30-Arc Second Databas. In: Finkl, C.W., Makowski, C. (Eds.), Seafloor Mapping along
27
ACCEPTED MANUSCRIPT 711
Continental Shelves: Research and Techniques for Visualizing Benthic Environments. Coastal Research
712
Library 13, pp. 169–190.
718 719 720 721 722
T IP
717
active and passive continental margins. Marine Geology 285, 69–86.
Harris, P.T., Macmillan-Lawler, M., Rupp, J., Baker, E.K., 2014. Geomorphology of the oceans. Marine Geology
CR
716
Harris, P.T., Whiteway, T., 2011. Global distribution of large submarine canyons: Geomorphic differences between
352, 4–24.
Helland-Hansen, W., Steel, R.J., Somme, T.O., 2012. Shelf genesis revisited. Journal of Sedimentary Research 82,
US
715
A., Krastel, S., Savini, A. (Eds.), Submarine Geomorphology. Springer Geology, pp. 111–134.
133–148.
AN
714
Harris, P.T., Macmillan-Lawler, M., 2018. Origin and Geomorphic Characteristics of Ocean Basins. In: Micallef,
Helland-Hansen, W., Sømme, T.O., Martinsen, O.J., Lunt, I., Thurmond, J., 2016. Deciphering Earth’s Natural Hourglasses: Perspectives On Source-To-Sink Analysis. Journal of Sedimentary Research 86, 1008–1033.
M
713
Hovius, N., 1996. Regular spacing of drainage outlets from linear mountain belts. Basin Research 8, 29–44.
724
Ingersoll, R.V., 2012. Tectonics of sedimentary basins, with revised nomenclature. In: Busby, C., Azor, A. (Eds.),
727 728 729 730 731 732 733 734
PT
Inman, D.L., Nordstrom, C.E., 1971. On the Tectonic and Morphologic Classification of Coasts. Journal of Geology 79, 1–21.
CE
726
Tectonics of Sedimentary Basins: Recent Advances. Wiley-Blackwell , Chichester, pp. 1-43.
Kineke, G.C., Sternberg, R.W., Trowbridge, J.H., Geyer, W.R., 1996. Fluid-mud processes on the Amazon
AC
725
ED
723
continental shelf. Continental Shelf Research 16, 667–696. Kingston, D.R., Dishroon, C.P., Williams, P.A., 1983. Global Basin Classification System. American Association of Petroleum Geologists Bulletin 67, 2175–2193. Lehner, B., Verdin, K., Jarvis, A., 2008. New Global Hydrography Derived From Spaceborne Elevation Data. EOS, Transactions, American Geophysical Union 89, 93–94. Meade, R.H., 1969. Landward transport of bottom sediments in estuaries of the Atlantic coastal plain. Journal of
28
ACCEPTED MANUSCRIPT 735 736
Sedimentary Research 39, 222-234. Miall, A., 2014. The Emptiness of the Stratigraphic Record: A Preliminary Evaluation of Missing Time In the
737
Mesaverde Group, Book Cliffs, Utah, U.S.A . Journal of Sedimentary Research 84, 457–469.
738
Milliman, J., Farnsworth, K., 2011. River Discharge to the Coastal Ocean: A Global Synthesis. Cambridge
742 743
T
IP
741
Milliman, J.D., Meade, R.H., 1983. World-Wide Delivery of River Sediment to the Oceans. Journal of Geology. 91, 1–21.
CR
740
University. Press. Cambridge, 392pp.
Milliman, J.D., Syvitski, J.P.M., 1992. Geomorphic/Tectonic Control of Sediment Discharge to the Ocean: The Importance of Small Mountainous Rivers. Journal of Geology. 100, 525–544.
US
739
Moore, G.T. 1969. Interaction of rivers and oceans--Pleistocene petroleum potential. American Association of
745
Petroleum Geologists Bulletin 53, 2421-2430.
746
Normark, W.R., 1970. Growth patterns of deep-sea fans. American Association of Petroleum Geologists Bulletin 54,
747
2170–2195.
748
Nyberg, B., Howell, J.A., 2015. Is the present the key to the past? A global characterization of modern sedimentary
751 752 753 754 755 756
M
ED
PT
Nyberg, B., Howell, J.A., 2016. Global distribution of modern shallow marine shorelines. Implications for
CE
750
basins Geology 43, 643-646.
exploration and reservoir analogue studies. Marine and Petroleum Geology 71, 83-104. Nyberg, B., Gawthorpe, R. L., Helland-Hansen, W., 2018. The distribution of rivers to terrestrial sinks: Implications
AC
749
AN
744
for sediment routing systems. Geomorphology 316, 1-23. Osmundsen, P.T., Redfield, T.F., 2011. Crustal taper and topography at passive continental margins. Terra Nova 23, 349–361. Pechlivanidou, S., Cowie, P.A., Hannisdal, B., Whittaker, A.C., Gawthorpe, R.L., Pennos, C., Riiser, O.S., 2017.
757
Source-to-sink analysis in an active extensional setting: Holocene erosion and deposition in the Sperchios rift,
758
central Greece. Basin Research 30, 522-543.
29
ACCEPTED MANUSCRIPT 759
Posamentier, H.W., Erskine, R.D., and Mitchum, R.M., Jr., 1991, Submarine fan deposition within a sequence
760
stratigraphic framework. In: Weimer, P., Link, M.H., (Eds), Seismic facies and sedimentary processes of
761
subma- rine fans and turbidite systems. Frontiers in Sedimentary Geology, pp. 127-136.
768 769 770 771 772 773
T
IP
CR
767
Rigon, R., Rodriguez-Iturbe, I., Maritan, A., Giacometti, A., Tarboton, D.G., Rinaldo, A., 1996. On Hack’s law. Water Resources Research 32, 3367–3374.
US
766
International Journal of Geographical Information Science 21, 983–1008.
Romans, B.W., Castelltort, S., Covault, J.A., Fildani, A., Walsh, J.P., 2016. Environmental signal propagation in sedimentary systems across timescales. Earth-Science Reviews 153, 7–29.
AN
765
Reuter, H.I., Nelson, A., Jarvis, A., 2007. An evaluation of void‐filling interpolation methods for SRTM data.
Ross, W.C., Watts, D.E. & May, J.A.,1995. Insights from stratigraphic modeling: mud-limited versus sand-limited
M
764
slope systems. Basin Research 29, 394–416.
depositional systems. American Association of Petroleum Geologists Bulletin 79, 231–258.
ED
763
Prather, B.E., O’Byrne, C., Pirmez, C., Sylvester, Z., 2017. Sediment partitioning, continental slopes and base-of-
Schumm, S.A., 1993. River response to baselevel change: implications for sequence stratigraphy river response to baselevel change: implications for sequence stratigraphy, Journal of Geology, 101, 279-294.
PT
762
Sella, G.F., Stein, S., Dixon, T.H., Craymer, M., James, T.S., Mazzotti, S., Dokka, R.K., 2007. Observation of
775
glacial isostatic adjustment in “stable” North America with GPS. Geophysical Research Letters 34, 1–6.
777 778
Shanmugam, G., 2016. Submarine fans: A critical retrospective (1950–2015), Journal of Palaeogeography 5, 110184.
AC
776
CE
774
Shanmugam, G., Moiola, R.J., Damuth, J.E., 1985. Eustatic Control of Submarine Fan Development. In: Bouma,
779
A.H., Normark, W.R., Barnes, N.E. (Eds.), Submariune Fans and Related Turbidite Systems. Frontiers in
780
Sedimentary Geology, pp. 23–28.
781
Sharman, G.R., Graham, S.A., Grove, M., Kimbrough, D.L., Wright, J.E., 2015. Detrital zircon provenance of the
782
late Cretaceous-Eocene California forearc: Influence of laramide low-angle subduction on sediment dispersal
783
and paleogeography. Geological Society of America Bulletin 127, 38–60.
30
ACCEPTED MANUSCRIPT 784
Sharman, G.R., Covault, J.A., Stockli, D.F., Wroblewski, A.F.J., Bush, M.A., 2017. Early Cenozoic drainage
785
reorganization of the United States Western Interior-Gulf of Mexico sediment routing system Geology 45,
786
187–190.
787
Snedden, J.W., Galloway, W.E., Milliken, K.T., Xu, J., Whiteaker, T., Blum, M.D., 2018. Validation of empirical source-to-sink scaling relationships in a continental-scale system: The Gulf of Mexico basin Cenozoic record.
789
Geosphere 14, 1–17.
792
IP
CR
791
Strahler, A.N., 1952. Hypsometric (area-altitude curve) analysis of erosional topography. Geological Society of America Bulletin 63, 1117-1141.
Sømme, T.O., Helland-Hansen, W., Martinsen, O.J., Thurmond, J.B., 2009. Relationships between morphological
US
790
T
788
and sedimentological parameters in source-to-sink systems: A basis for predicting semi-quantitative
794
characteristics in subsurface systems. Basin Research 21, 361–387.
795
AN
793
Sømme, T.O., Martinsen, O.J., Lunt, I., 2013. Linking offshore stratigraphy to onshore paleotopography: The Late Jurassic-Paleocene evolution of the south Norwegian margin. Geological Society of America Bulletin 125,
797
1164–1186.
800 801
ED
of Fluvial Sediment to the Coastal Ocean. Journal of Geology 115, 1–19.
PT
799
Syvitski, J.P.M., Milliman, J.D., 2007. Geology, Geography, and Humans Battle for Dominance over the Delivery
Voris, H.K., 2000. Maps of Pleistocene sea levels in Southeast Asia: Shorelines, river systems and time durations.
CE
798
M
796
Journal of Biogeography 27, 1153–1167. Wetzel, A., 1993. The transfer of river load to deep- sea fans – a quantitative approach. American Association of
803
Petroleum Geologists Bulletin 77, 1679-1692.
804
Whittaker, J.M., Goncharov, A., Williams, S.E., Müller, R.D., Leitchenkov, G., 2013. Global sediment thickness
805
data set updated for the Australian-Antarctic Southern Ocean. Geochemistry, Geophysics. Geosystems 14,
806
3297–3305.
807
AC
802
Woodcock, N.H., 2004. Life span and fate of basins. Geology 32, 685–688.
808 31
ACCEPTED MANUSCRIPT 809
Table 1. Summary of morphological parameters defined in the S2S database by catchment, continental shelf,
810
continental slope and submarine fan segments. The table shows the parameter name, the calculation, a description of
811
the measurement and units, and the original dataset(s) used to define the measurement. See main text for an
812
explanation of the methodology and supplementary information for the S2S database and metadata.
813 Figure 1. The morphological parameters of the source-to-sink system that were analyzed in the global study
815
(modified after Sømme et al., 2009). 1 DEM based on SRTM v4.1 (Jarvis, 2008) and GMTED2010 (Danielson and
816
Gesch, 2011), 2 catchment attributes based on Somme et al. (2009), Milliman and Farnsworth (2011) and Nyberg et
817
al. (2018), 3 terrestrial sinks based on Nyberg and Howell (2015), 4 bathymetry data based on SRTM30 Plus (Becker
818
et al., 2009) and 5 seafloor delineations based on Harris et al. (2014).
US
CR
IP
T
814
AN
819
Figure 2. An example of the semi-automated method to derive morphological measurements of the catchment,
821
continental shelf, continental slope and submarine fan for the contributing catchment area of the Bengal submarine
822
fan. River profile lengths are derived based on DEM and extend to its continental shelf by its centerline profile while
823
the continental slope profile is defined as its most proximal distance to the abyssal plain. The topographic and
824
bathymetric profiles that connect to a submarine fan define the contributing catchment area. Basemap shows a
825
hillshaded SRTM 30 Plus DEM and bathymetry dataset of Becker et al. (2009).
CE
PT
ED
M
820
826
Figure 3. A comparison of continental slope length and slope gradient measured in the current study and that of
828
Somme et al. (2009). The variability in the two datasets demonstrates the subjective nature of interpreting the
829
continental slope.
AC
827
830 831
Figure 4. The distribution of 52 complete source-to-sink systems showing their contributing catchment areas to
832
submarine fans. A comparison of their scale is shown surrounding the global map with a planar view of its river
833
channel, continental shelf, slope and submarine fan profile (see Figs. 2, 13 and supplementary kmz). Note the
834
change in scalebar. Basemap shows a hillshaded SRTM 30 Plus DEM and bathymetry dataset of Becker et al.
32
ACCEPTED MANUSCRIPT 835
(2009). Map is shown with a Robinson projection at a 1:125 000 000 scale. Source-to-sink system comparisons are
836
in a Cylindrical Equal Area projection.
837 Figure 5. Morphological relationships observed in the catchment segment (Qs > 1 MT/yr). The data show the
839
scaling patterns of river length, maximum relief, % of low relief area and river profile gradient by catchment area.
840
Note the lack of correlation between relief and low relief area by system size.
IP
T
838
CR
841
Figure 6. (A) Relief versus catchment area for a regional-based selection of 1 863 catchments (Qs > 0 MT/yr)
843
grouped into different tectonic settings. (B) Relief versus catchment area for a regional-based selection of 870
844
systems (Qs > 0 MT/yr) grouped by mature and young passive margins. See text for discussion.
AN
US
842
M
845
Figure 7. Morphological relationships observed on the continental shelf segment by a simplified tectonic
847
classification for the 52 source-to-sink systems (Fig. 4). The results show a weak power law correlation between
848
river length and shelf width and shelf width and slope length. The gradient on the shelf does appear to correlate to
849
river profile gradient while a correlation is observed between estuarine length and shelf width. ES – epicontinental
850
seaway.
CE
PT
ED
846
851
Figure 8. Box plots showing the distribution of continental shelf width and sediment thickness of the continental
853
shelf/slope for active versus passive continental shelf/slope margins based on 1 778 source-to-sink systems. Box
854
shows 25th and 75th percentile, whiskers show 10th and 90th percentile, square shows the mean and line shows the
855
median of the data.
AC
852
856 857
Figure 9. Morphological relationships observed on the continental slope segment. The results show a weak power
858
law correlation between slope length and catchment area, shelf width, fan area, fan length, length of river channel
33
ACCEPTED MANUSCRIPT 859
and river gradient. The subjective nature of defining the continental slope boundary suggests it is not a good
860
parameter to scale source-to-sink segments. ES – epicontinental seaway.
861 Figure 10. Box plots showing (A) slope length by active versus passive continental shelf/slope margins, and (B)
863
depth at the continental slope base for 1 778 source-to-sink catchments by tectonic setting based on ETOPO1
864
bathymetry (Amante and Eakins, 2009). The results show that slope length does not vary significantly by active
865
versus passive continental shelf/slope margins (or by further tectonic regime subdivision). Depth at the continental
866
slope base is similar across tectonic regimes with the exception of forearc settings. Box shows 25th and 75th
867
percentile, whiskers show 10th and 90th percentile, square shows the mean and line shows the median of the data. ES
868
– Continental slope adjacent to the epicontinental seaway.
AN
869
US
CR
IP
T
862
Figure 11. The number of submarine canyon heads (A) and the longest submarine canyon length (B) for source-to-
871
sink systems that contribute to one of the 52 submarine fan in Figure 4. Box shows 25 th and 75th percentile, whiskers
872
show 10th and 90th percentile, square shows the mean and line shows the median of the data. Sh/SL – Continental
873
shelf / Continental slope, ES - Continental slope adjacent to the epicontinental seaway.
PT
ED
M
870
875
CE
874
Figure 12. Morphological relationships observed on the submarine fan segment. The results show a power law
877
correlation between increasing submarine fan length and fan width. In addition, proximal and distal relationships are
878
observed for fan length and contributing river length, fan area and contributing catchment area, and contributing
879
river sediment load (Qs) and fan area. Depths of the submarine fan are similar across scales while submarine fan
880
slope decreases with a power law relationship as contributing catchment area increases.
881
Figure 13. Relative topographic and bathymetric length and rise profiles of mature passive margins (see Fig. 4 for
882
location) in descending source-to-sink catchment size. The continental shelf/slope segments suggest a relative length
883
increase with decreasing system size. The data demonstrate that while catchment size changes, continental slope
AC
876
34
ACCEPTED MANUSCRIPT length does not significantly alter thereby increasing the relative length of the continental slope with decreasing
885
source-to-sink size. Note, the submarine fan profile does not represent a drainage profile but a perpendicular transect
886
from the base of the continental slope (see Fig. 4 for location; see also supplementary kmz).
887
Figure 13 (cont.). Relative topographic and bathymetric length and rise profiles of young passive margins, foreland,
888
epicontinental seaways and strike-slip source-to-sink profiles (see Fig. 4 for location) in descending source-to-sink
889
catchment size. Note, the submarine fan profile does not represent a drainage profile but a perpendicular transect
890
from the base of the continental slope (see Fig. 4 for location; see also supplementary kmz).
891
Figure 14. Scatter plots showing for 52 S2S systems (see Fig. 4 for location), the proportional length of each
892
segment comparative to its entire source-to-sink profile length (Fig. 13) by catchment size. The results suggest that
893
river length is proportionally a larger component of the source-to-sink profile with increasing catchment size while
894
the length of its continental slope and shelf is an increasingly smaller component. By contrast, smaller systems
895
suggest that the continental shelf and slope length are proportionally of an increasingly larger component of the
896
source-to-sink system. Proportionally, the submarine fan length shows the most amount of variability with a lack of
897
correlation to the entire source-to-sink profile length. See text for additional discussion.
ED
M
AN
US
CR
IP
T
884
898
Figure 15. Illustration showing the change in catchment morphology of relief and catchment area / river length at T1
900
to T2 based on sea level fluctuations (blue) and regional tectonic uplift (red). For systems with wider terrestrial
901
sinks/continental shelves (A) the large low gradient plain/shelf may proportionally fluctuate the morphology of the
902
catchment significantly while tectonic uplift may have a minimal effect. By contrast, smaller and steeper source-to-
903
sink systems (B) would experience minimal morphological change in terms of a similar relative sea level fall from
904
T1 to T2 to (A), whereas regional tectonic uplift from T1 to T2 may significantly divert drainage area and lower
905
relief from the original catchment region. Insert graphs show the relative change in catchment morphology by a
906
relative sea level or regional tectonic uplift influence for A and B Note, the relative change is a simplistic illustration
907
of a relationship that is not linear.
908
Figure 16 – Summary of morphological parameters of modern source-to-sink systems with a total suspended
909
sediment load (Qs) greater than 1 MT/yr. Numbers show mean and standard deviation of: river length (RL), river
AC
CE
PT
899
35
ACCEPTED MANUSCRIPT gradient (RG), continental shelf width (SW), continental slope length (SL), submarine fan length (SFL), continental
911
shelf depth (ShD), continental shelf gradient (ShG), continental slope depth (SlD), continental slope gradient (SlG),
912
submarine fan depth (SFD), submarine fan gradient (SFG), % of catchment area as terrestrial sink (TS Area), and %
913
of catchment area as low relief area < 10 m/km. n = number of source-to-sink systems considered (catchment, shelf
914
and slope / submarine fan). Steep, short and deep systems (A) are separated based on an extensional/strike-slip or a
915
forearc origin. Wide and deep systems (B) are characterized by sedimentary systems of a passive or foreland
916
tectonic origin on the margin of a passive continental shelf/slope. See text for discussion. Note that the illustration is
917
not drawn to scale.
CR
IP
T
910
AC
CE
PT
ED
M
AN
US
918
36
ACCEPTED MANUSCRIPT Catchment Segment Parameter
Calculation
Description
Dataset 2
A
Catchment area (km )
Nyberg et al. (accepted)
Maximum Relief (R)
R
Maximum catchment relief (km)
Nyberg et al. (accepted)
% < 10 m/km (10m_km)
10m_km
% of catchment area with < 10 m/km in relief
Reuter et al. (2007); Danielson and Gesch (2011)
% Terrestrial Sink (Sed_Basin)
Sed_Basin
% of catchment area defined as a terrestrial sink based on Nyberg and Howell (2015)
Nyberg et al. (accepted)
River Length (River_L)
River_L
Longest river channel (m) derived from endorheic/ exorheic sink to furthest reach along the river network
Nyberg et al. (accepted)
River Profile Gradient (River_S)
R/River_L
Average river profile gradient (m/km)
Water Discharge (Q)
0.00237A0.8
Water discharge (km3/yr) based on Q equation of Syvitski and Milliman (2007)
Nyberg et al. (accepted)
Total Suspended Sediment Load (Qs)
BQART
Total suspended sediment load (MT/yr) based on BQART equation of Syvitski and Milliman (2007)
Nyberg et al. (accepted)
CR
IP
T
Area (Area)
US
Continental Shelf Segment BP_L – S_Dist
Shelf Width (S_Dist)
S_Dist
Shelf Depth (Shelf_LowH)
Shelf_LowH
Shelf Profile Gradient (C_Shelf_S)
Shelf_LowH/S_Dist
Sediment Thickness (Sed_Thick)
Sed_Thick
Base of Slope Depth (Slope_LowH)
Average sediment thickness on the continental shelf/slope (m) measured along the bathymetric profile
Whittaker et al. (2013)
Length of the continental slope (m) measured along the bathymetric profile
Slope_LowH
Depth at the base of the continental slope/basin floor transition (m)
Becker et al. (2009)
Slope_TopH
Depth at the top of the continental slope (m) measured along the bathymetric profile
Becker et al. (2009)
CE
Top of Slope Depth (Slope_TopH)
Becker et al. (2009)
Gradient of the continental shelf (m/km) measured along the bathymetric profile
M
Slope_L
PT
Slope Length (Slope_L)
Depth at the continental shelf edge (m) measured along the bathymetric profile
ED
Continental Slope Segment
AN
Waterbody/estuarine length (m) measured as the difference of the bathymetric river to shelf edge profile length (BP_L) and shelf width (S_Dist) Width of the continental shelf (m) measured as distance from continental shelf edge
Waterbody Length (E_Length)
(Slope_TopH – Slope_LowH) / Slope_L
Average gradient of the continental slope profile (m/km) measured along the bathymetric profile
Canyon Length (C_Length)
C_Length
Length of the submarine canyon (km)
Harris and Whiteway (2011)
Canyon Head Frequency (C_Freq)
C_Freq
Number of canyon heads that incise the continental shelf/slope
Harris and Whiteway (2011)
Canyon Distance (C_Distance)
C_Distance
Distance from river mouth to the submarine canyon (m)
Fan Area (F_Area)
F_Area
Area of the submarine fan (km2)
Fan Length (F_Length)
F_Length
Length of the submarine fan (m) from base of continental slope to distal submarine fan
Fan Width (F_Width)
F_Width
Width of the submarine fan (m) measured parallel to the base of the continental slope
Fan Depth (F_Depth)
F_Depth
Depth of the distal submarine fan (m)
AC
Slope Profile Gradient (C_Slope_S)
Submarine Fan Segment
Becker et al. (2009)
37
ACCEPTED MANUSCRIPT (Slope_LowH – F_Depth) / F_Length
Fan Profile Gradient (F_S)
Average gradient of the submarine fan profile (m/km)
AC
CE
PT
ED
M
AN
US
CR
IP
T
919 920 921
38
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13r1
Figure 13r2
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18
Björn Nyberg, William Helland-Hansen, Rob L. Gawthorpe, Pål Sandbakken, Christian Haug Eide, Tor Sømme, Frode HadlerJacobsen, Sture Leiknes PII: DOI: Reference:
S0037-0738(18)30158-1 doi:10.1016/j.sedgeo.2018.06.007 SEDGEO 5361
To appear in:
Sedimentary Geology
Received date: Revised date: Accepted date:
31 March 2018 7 June 2018 12 June 2018
Please cite this article as: Björn Nyberg, William Helland-Hansen, Rob L. Gawthorpe, Pål Sandbakken, Christian Haug Eide, Tor Sømme, Frode Hadler-Jacobsen, Sture Leiknes , Revisiting morphological relationships of modern source-to-sink segments as a firstorder approach to scale ancient sedimentary systems. Sedgeo (2018), doi:10.1016/ j.sedgeo.2018.06.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Revisiting Morphological Relationships of Modern Source-to-Sink
2
Segments as a First-Order Approach to Scale Ancient Sedimentary
3
Systems
4
Björn Nyberg1, William Helland-Hansen1, Rob L. Gawthorpe1, Pål Sandbakken2, Christian Haug Eide1, Tor
5
Sømme3, Frode Hadler-Jacobsen4, Sture Leiknes2
IP
1
6
T
1
CR
Department of Earth Sciences, University of Bergen, P.O. Box 7803, 5020 Bergen, Norway. 2
7
US
Statoil ASA, Sandsliveien 90, 5254 Sandsli, Norway
3
8
Statoil ASA, Martin Linges vei 33, 1364 Fornebu, Norway
4
9
AN
Statoil ASA, Arkitekt Ebbells veg 10, 7053 Ranheim, Norway
Abstract
11
Catchments provide water and sediment to downstream sedimentary systems, and these form individual source-to-
12
sink systems. Source-to-sink systems comprise adjacent linked segments, commonly hinterland catchments, alluvial-
13
and coastal plains, the continental shelf, continental slope and submarine fan. The dimensions of the catchment and
14
how it scales to downstream segments provides insight into the sedimentary and tectonic controls that influence the
15
morphology and sedimentation patterns in a basins evolution. In ancient sedimentary successions, where the
16
sedimentary routing system is buried and inaccessible for study, or fragmented due to uplift and erosion, using
17
scaling relationships can provide a powerful tool to understand the complete sedimentary system.
18
Observational data from modern sedimentary systems provide an opportunity to create morphological and
19
sedimentological scaling relationships of segments on the entire source-to-sink system. However, previous studies
20
on global modern source-to-sink systems have typically been based on a limited number of examples restricted by
21
the data available at the time and the methodology used to analyze large datasets. In the last decade, the volume and
22
quality of remotely sensed information has significantly improved so that it is now timely to revisit scaling
23
relationships of modern source-to-sink systems’ segment morphologies, and discuss the implications of those results
24
for sedimentological parameters and applicability to ancient source-to-sink systems.
AC
CE
PT
ED
M
10
1
ACCEPTED MANUSCRIPT The results of this reanalysis show that dimensions of the catchment and submarine fan segments scale internally in
26
terms of fan width, length and area. In addition, fan area scales to its largest hinterland catchment area in agreement
27
with previous studies, however, it is important to consider all catchments that contribute sediment to a basin floor
28
region. In paleogeographic settings, where individual submarine fans are difficult to tie to a single catchment, and
29
where basin floor systems are amalgamated, the contributing sediment discharge of all catchments may be
30
significant and likely influence the scale of its submarine fan.
31
Accommodation versus sediment supply in relation to relative sea level change are important controls on the
32
position of the shoreline which vary considerably from system to system over time and space, thus influencing
33
morphological relationships between source-to-sink segments. The continental shelf should therefore be viewed as a
34
transient geomorphic feature rather than a segment of a source-to-sink system. Furthermore, the continental slope
35
length should not be used to scale other segments of the source-to-sink system, which contradicts previous research.
36
The underlying tectonic and sedimentological control on the continental shelf and slope segments, in addition to the
37
subjective interpretation of their basinward boundaries, may render those segment unsuitable for scaling the
38
morphology of other segments. The study highlights both the temporal variability and complexity of controls that
39
influence the morphology and scaling relationships of internal and adjacent linked source-to-sink segments, and the
40
need to place this in a framework of both tectonic and sedimentological history.
ED
M
AN
US
CR
IP
T
25
Keywords: source-to-sink; morphology; catchment; continental shelf; continental slope; submarine fan
43
45
1.
Introduction
AC
44
CE
42
PT
41
46
Source-to-sink (S2S) systems consist of linked segments of sediment routing pathways from areas of erosion in the
47
hinterland to areas of deposition in the basin. A segment of a S2S system is a morphologically distinct component,
48
typically in the form of the catchment, continental shelf, continental slope and submarine fan (Somme et al., 2009)
49
(Fig. 1). The morphology of a segment relates to the propagation and response to updip and internal sedimentary
50
signals occurring over different temporal and spatial scales (Moore, 1969; Helland-Hansen et al., 2016).
2
ACCEPTED MANUSCRIPT 51
Understanding morphological and sedimentological characteristics of a segment and how those characteristics scale
52
to adjacent segments, can improve our knowledge of tectonic and climatic forcing on the entire S2S system.
53 In ancient sedimentary systems, preserved stratigraphic successions often represent an incomplete or partial record
55
of segments along the sediment routing pathway (Miall, 2014). A complete perspective of the entire S2S system can
56
constrain information on system scale, morphological characteristics, and tectonic and climatic controls that are
57
important to determine dimensions, volumes and timing of preserved sediments (Allen, 2008; Allen et al., 2013;
58
Helland-Hansen et al., 2016). Morphological and sedimentological characteristics of a segment and its relationship
59
to dimensions of other segments can thus provide a useful method to predict dimensions of a segment not preserved
60
in the stratigraphic record (Sømme et al., 2009). Empirical observations of modern S2S systems offers one method
61
to constrain morphological characteristics of segments, and relationships between segments, along the complete
62
sediment routing system. A significant body of research have previously characterized the morphology,
63
sedimentation characteristics and/or scaling relationships within modern catchments (e.g., Strahler, 1952; Hack,
64
1957; Milliman and Meade, 1983; Hovius, 1996; Blum and Tornqvist, 2000; Syvitski and Milliman, 2007; Romans
65
et al., 2016; Nyberg et al., 2018), continental shelves (e.g., Inman and Nordstrom, 1971; Harris and Macmillan-
66
Lawler, 2016; Nyberg and Howell, 2016), continental slopes (e.g., Harris and Whiteway, 2011; Prather et al., 2017)
67
and submarine fans (e.g., Normark, 1970; Barnes and Normark, 1985; Wetzel, 1993; Covault and Romans, 2009;
68
Covault and Graham, 2010; Shanmugam, 2016; Harris and Macmillan-Lawler, 2018). However, a limited number of
69
studies (e.g., Wetzel, 1993; Sømme et al., 2009) have explored the scaling relationships on the morphology between
70
those segments on a global and quantitative perspective.
IP
CR
US
AN
M
ED
PT
CE
AC
71
T
54
72
One of the first quantitative approaches to correlate parameters of the proximal to distal segments of modern S2S
73
systems was Normark (1970) who discussed river load controls on submarine fan morphology offshore California.
74
Wetzel (1993) expanded this research to a global perspective of 22 modern, 2 lacustrine and 5 ancient submarine fan
75
systems and showed relationships of river load and submarine fan depositional rate, length and width. Subsequent
76
work by Somme et al. (2009) based on 29 modern S2S systems, tied morphological and sedimentological
77
relationships to the entire sediment routing system from its catchment, continental shelf, continental slope and
78
submarine fan. The authors show that the continental slope length may be a particularly useful parameter to scale
3
ACCEPTED MANUSCRIPT segments that are not preserved or inaccessible in an ancient S2S system. The results have been instrumental in a
80
number of studies to provide a first-order estimate of segment scale, where limited subsurface information is
81
available, and in further developing models of S2S systems (see Helland-Hansen et al., 2016; Allen, 2017; Snedden
82
et al., 2018). However, the scaling relationships observed are based on a few modern systems and are often
83
simplistic by mainly considering S2S segment morphology to sedimentological processes as opposed to an inter-
84
disciplinary correlation between tectonics and sedimentation.
T
79
IP
85
Nearly a decade since the completion of the original work of Somme et al. (2009), new global and quantitative
87
research and datasets have improved our understanding of the physical world. Global delineations of river drainage
88
patterns have improved through high resolution digital elevation models (DEM) such as Hydrosheds (Lehner et al.,
89
2008). Milliman and Farnsworth (2011) have continued to compile observations of catchment characteristics and
90
river load attributes around the world. Nyberg and Howell (2015) delineated a global database of modern terrestrial
91
sedimentary basins which are important sites of sediment storage and transfer to the oceans (Romans et al., 2016).
92
Harris and Whiteway (2011) have mapped and analyzed morphological characteristics of 5849 large submarine
93
canyons around the world. Global bathymetry data has improved with the SRTM 30 Plus product at a 30-arc second
94
resolution (Becker et al., 2009) which led to the development of the first global geomorphic seafloor map by Harris
95
et al. (2014). Furthermore, morphological characteristics of global continental shelves and shelf breaks have been
96
analyzed by Harris and Macmillan-Lawler (2016).
PT
ED
M
AN
US
CR
86
CE
97
The volume of new data delineating and characterizing the extent of S2S segments provides an opportunity to revisit
99
morphological scaling relationships on a truly global scale and assess the implications for sediment routing systems.
AC
98
100
The aim of this paper are threefold: (1) to introduce a quantitative database describing the morphology and
101
relationships between modern S2S segments of the catchment, continental shelf, continental slope and submarine
102
fan; (2) to discuss the temporal variability of those relationships; and (3) to revisit and revise previous models of
103
S2S systems based on the findings of the present study.
104 105
2. Overview of Database, Methodology and Sources
4
ACCEPTED MANUSCRIPT A digital database of modern S2S systems is created documenting the morphology of catchments, continental
107
shelves, continental slopes and submarine fans (Fig. 1) to analyze the morphological character and scaling
108
relationships of S2S segments. The S2S database consist of 69,586 systems of which 56,196 catchments describe
109
exorheic (i.e., that drain to the ocean, or a body of water connected to an ocean) and 13,390 catchments describe
110
endorheic systems (e.g., that drain to lakes with no outflow). In addition, the morphological character of 52
111
submarine fans are described that relate to 4,843 up dip catchments that drain 23% of the non-glaciated terrestrial
112
area. The S2S database and a data statement describing the attributes of the dataset are provided as supplementary
113
information to this article. The methodology and sources of geographical data used to define the digital database of
114
S2S segment morphology are based on a set of manual and automated techniques that are described in detail below
115
(Table 1).
US
CR
IP
T
106
116
Morphological measurements of the terrestrial realm were derived from the Global Terrestrial Sink Catchment
118
(GTSC) dataset of Nyberg et al. (2018) and digital elevation models (DEM) of SRTM v4.1 (Reuter et al., 2007) and
119
GMTED2010 (Danielson and Gesch, 2011). The GTSC database was developed from hydrologically conditioned
120
river network data based on HydroSHEDS (Lehner et al., 2008) in addition to high-resolution DEMs to show the
121
distribution of rivers to modern terrestrial sinks. A terrestrial sink (or terrestrial sedimentary basin) is defined by the
122
authors as a region that is undergoing net subsidence and has shown sediment accumulation since the Holocene
123
based on the work of Nyberg and Howell (2015).
M
ED
PT
CE
124
AN
117
The parameters in the GTSC database include information on catchment area, river length, maximum relief,
126
percentage of catchment defined as a terrestrial sink, water discharge (Q) and total suspended sediment load (Qs)
127
(Table 1). The river profile gradient is defined as a rivers maximum relief over the length of its longest river
128
channel. The proportion of a catchment defined as low-lying plains with a relief <10 m/km was derived based on
129
DEMs (e.g., Reuter et al., 2007; Danielson and Gesch, 2011). A <10 m/km relief area may or may not include a
130
terrestrial sink representing a low-lying area that does not necessarily have long-term terrestrial sediment storage
131
potential but is otherwise important in influencing sediment transport to the marine realm (Sømme et al., 2009;
132
Allen et al., 2013; Romans et al., 2016; Nyberg et al., 2018) (Fig. 1).
AC
125
133 5
ACCEPTED MANUSCRIPT Delineations of continental shelves, continental slopes and submarine fans were derived from the consistent global
135
seafloor geomorphic features map (GSFM) of Harris et al. (2014). Harris et al. (2014) manually defined the shelf
136
edge and the foot of the continental slope by analyzing 100 m contour grids and 3D viewing at scales of 1:500,000
137
to mark abrupt changes in gradient. To measure shelf width from the continental shelf delineation, a series of lines
138
that represent thiessen polygons surrounding the vertices outlining the continental shelves were constructed into a
139
graph network. The shortest cumulative distance from each river mouth to its closest continental shelf edge through
140
these lines represents bathymetric profiles from river to shelf edge (e.g., Fig. 2).
141
The bathymetric profile may not strictly relate to its continental shelf width by including a significant distance along
142
a coastline waterbody (e.g., fjord, lagoon, estuary or embayment). This is a problematic issue as Harris et al. (2016)
143
noted in a global geomorphic overview of modern continental shelves. To resolve this dilemma, the current study
144
extracted the centerlines of global continental shelves from the graph network mentioned above and intersected each
145
bathymetric profile to create a centroid point on the continental shelf. The proximal distance from each point to the
146
continental landmass and to the continental shelf edge provides a reliable and objective measure of shelf width for
147
the global S2S database. The approach has the added benefit of measuring the waterbody length as the difference
148
between shelf width and the bathymetric profile length from river to shelf edge (Fig. 1; Table 1).
149
Continental slope length is defined by a simple measure from its continental shelf edge to its proximal basin floor as
150
defined by the base of the continental slope definition of Harris et al. (2014). This is appropriate given the simplicity
151
of the continental slope delineation relative to that of the terrestrial coastline. Depth of the continental shelves and
152
continental slopes are based on global bathymetric data from the SRTM 30 Plus product at a 30-arc second
153
resolution (Becker et al., 2009). Information on depth and width of the continental shelves, and depth and length of
154
the continental slopes, along each S2S bathymetric profile (e.g., Fig. 2) define the gradient of those respective
155
segments (Table 1).
156
An averaged sediment thickness along the bathymetric profile of the continental shelf and slope was sampled every
157
5 km based on global maps of Whittaker et al. (2013). The distribution of major submarine canyons were extracted
158
from the database of Harris and Whiteway (2011) who mapped 5,849 large submarine canyons showing a depth
159
range greater than 1 km and a continental shelf/continental slope incision exceeding 100 m. Morphological
160
information on submarine canyon length and frequency of canyon heads were associated to each bathymetric profile
AC
CE
PT
ED
M
AN
US
CR
IP
T
134
6
ACCEPTED MANUSCRIPT within 10 km of a canyon and assigned to a S2S system. Distance from the river mouth to each canyon is defined by
162
the distance along each bathymetric profile (Table 1).
163
Harris et al. (2014) identify global locations of submarine fans using sediment thickness maps on the continental rise
164
at the base of submarine canyons that show a concentric feature spreading seawards as observed by 100 m isobaths.
165
Based on 79 submarine fans delineations by Harris et al. (2014), manual measurements were used to define
166
submarine fan length as the furthest distance from base of the continental slope to the distal fan, and submarine fan
167
width as the widest portion parallel to the base of the continental slope (Fig. 1). Global bathymetry (e.g., Becker et
168
al., 2009) defined submarine fan depth and average submarine fan gradient was defined by bathymetric fall over its
169
submarine fan length (Table 1).
170
3. Accuracy of Morphological Measurements
US
CR
IP
T
161
AN
171
To assess the accuracy of the automated methods in defining morphological parameters of Table 1, the global S2S
173
database was compared to a set of manual observations and previous published literature. The parameters and
174
accuracy of those measurements are discussed in the following sections.
M
172
ED
175 3.1. Terrestrial Realm
177
The parameters of total suspended sediment load (Qs), river length, catchment area and catchment relief are very
178
similar in 28 modern S2S systems that are previously reported in both Somme et al. (2009) and Milliman and
179
Farnsworth (2011) (see Appendix A). The largest source of variability is associated with maximum measured relief
180
between the current study and that of Milliman and Farnsworth (2011) with a positive linear correlation at an r2 of
181
0.58. This is due to lower resolution cartographic surveys that were primarily used to build upon the Milliman and
182
Farnsworth (2011) database as opposed to high-resolution DEMs in the current study. A comparison of the present
183
study to the 28 S2S systems compiled by Sømme et al. (2009) (excluding the Danube system), indicates that where
184
similar DEM data are used, there is a strong positive 1 to 1 linear correlation with an r2 at 0.97. River profile
185
gradient is also similar in the current study to that of Somme et al. (2009) with a positive slope of 0.95 and an r2 of
186
0.87.
AC
CE
PT
176
7
ACCEPTED MANUSCRIPT 187
3.2. Marine Realm
188
The accuracy of the automated method applied in this study to estimate continental shelf width and slope length
189
shows a high degree of confidence when compared to manual measurements of 100 randomly selected S2S systems
190
(see Appendix B). Similarly, continental shelf width and continental shelf gradient are comparable for 28 S2S
191
systems between the current dataset and Somme et al. (2009) (see Appendix B).
IP
T
192
In contrast, a comparison of continental slope length and gradient between the current study and that of 28 systems
194
measured by Somme et al. (2009) shows a higher degree of variability (Fig. 3). One probable reason for the
195
discrepancy may lie in the current methodology where the present study considers the maximum slope length along
196
all catchments that contribute to a particular submarine fan (Fig. 2). The reason for this is that the current highstand
197
positioning of the largest contributing catchment, frequently misaligns with the widest portion of the continental
198
slope or the longest submarine canyon. This may reflect a relict drainage system, longshore drift of sediment or the
199
present-day disconnection of catchments to its shelf edge that spatially may shift during lowstand. Exceptions do
200
occur including the present-day Congo River where the submarine canyon incises significantly into its shelf and is
201
currently positioned directly offshore the river mouth.
ED
M
AN
US
CR
193
202
Also, whereas the continental shelf is defined by a clear shelf edge break, the base of the continental slope is more
204
gradual and reflects a combination of tectonic controls and stratigraphic overprinting, yielding different slope
205
profiles and grain size distributions (Ross et al., 1994; Praher et al., 2017). Identifying the base of the continental
206
slope is also difficult in ancient system reconstructions for similar reasons (Snedden et al. 2018). Small differences
207
in the chosen values for gradient thresholds may yield different analytical results. It does, however, highlight the
208
potential difficulty in applying the slope length for scaling the segment morphology of paleo-S2S systems. While
209
similarity between measurements on the continental slope in this and the study by Sømme et al. (2009) is low, a
210
consistent delineation of continental slope by Harris et al. (2014) used in the present study, provides an opportunity
211
to examine how this impacts morphological relationships on the broader and global perspective of S2S system
212
scaling.
AC
CE
PT
203
213
8
ACCEPTED MANUSCRIPT Morphological measurements of submarine fans are based on the delineations of Harris et al. (2014). A comparison
215
to 8 corresponding Holocene fan dimensions compiled from published literature by Somme et al. (2009) show
216
strong correlations, giving confidence in the current database (see Appendix B). Submarine fan length and area are
217
similar with a positive linear correlation and a slope of 1.08 and 1.18, respectively, though fan width measurements
218
are slightly wider in the current study as it does not separate individual fan aprons by age. This may be appropriate
219
as a first-order basis to scale paleo-S2S systems where individual fan systems will be difficult to constrain. For
220
geometries measured on the basin floor, the largest difference between the present study and that of Somme et al.
221
(2009) appear to be associated with submarine fan depth with an r2 of 0.23, but removing the Orinoco system outlier,
222
the positive linear correlation improves with an r2 of 0.70. On the basis of this discrepancy, we consider it timely to
223
employ strict, rule-based, automated methods to measure S2S segment geometries in order to more accurately derive
224
morphological relationships.
US
CR
IP
T
214
AN
225 226
4. The Morphology of Modern Source-to-Sink Systems
228
In this section, we will focus on morphological relationships that characterize the catchment, continental shelf,
229
continental slope and submarine fan segments of modern S2S systems. The S2S database contains 69,586
230
catchments, however, morphological analyses on the catchment, continental shelf and continental slope have been
231
based on only 1,778 catchments that have a total suspended sediment load (Qs) greater than 1 MT/yr. The
232
catchments therefore represent the largest sedimentary systems on earth and cover 75% of the non-glaciated
233
terrestrial landsurface. In addition, S2S systems have been analyzed based on the larger contributing area of 4,843
234
catchments (including the 1,778 S2S systems) that drain 23% of the non-glaciated landmass to 52 submarine fans
235
(Fig. 4). The S2S systems span a range of scales with a contributing catchment area from 2.99 x 102 to 7.13 x 106
236
km2 and a submarine fan area of 1.71 x 102 to 2.55 x 106 km2, providing a robust analysis of morphological scaling
237
relationships.
238
Morphological relationships have been analyzed in a categorization of main catchment tectonic regime: forearc,
239
foreland, passive margin, strike-slip or extensional setting (Nyberg et al., 2018). In addition, the current study
240
identifies epicontinental seaways as those S2S systems without a continental slope and basin floor configuration,
AC
CE
PT
ED
M
227
9
ACCEPTED MANUSCRIPT based on the work of Nyberg and Howell (2016). The tectonic regime of the continental shelf/slope margin is used
242
to distinguish active versus passive margins. For instance, S2S systems draining to the Red Sea represent passive
243
margin catchments along an active (or young) continental shelf/slope margin, in comparison to passive margin
244
catchments draining to the northern Gulf of Mexico along a passive (or mature) continental shelf/slope. Active
245
versus passive continental slope margins reflect a similar nomenclature used by previous authors (e.g., Inman and
246
Nordstrom, 1971; Sømme et al., 2009; Harris et al., 2014).
247
4.1. The Catchment
248
Figure 5 summarizes relationships between catchment area, river length, maximum relief, % of low relief area, and
249
river gradient that were defined in the original study of Sømme et al. (2009). Catchment area and river length
250
suggest a power-law relationship (Fig. 5) and catchment area (and subsequently its river length) are thought to
251
increase as a system ages by incorporating new drainage areas (Hack, 1957; Milliman and Meade, 1983). However,
252
tectonic movement in the hinterland may add to, divert and/or reroute drainages which is particularly relevant of
253
small basins along active tectonic margins. Regardless, the relationship between catchment area and river length is
254
invariant of scale and remains valid when analyzing any sub-catchment within the entire catchment area (Hack,
255
1957; Rigon et al., 1996; Nyberg et al., 2018).
256
The maximum amount of relief in a catchment does not correlate with catchment area (Fig. 5) and further
257
subdivision by tectonic regime does not improve this correlation. However, a regional based analysis of selective
258
S2S systems with similar tectonic histories (e.g., passive margins along the northern Gulf of Mexico, Amazon,
259
Congo and Niger; Fig. 4), does indicate that relief of a system relates to catchment area (Fig. 6A). By removing the
260
tectonic influence and only analyzing systems with a similar sedimentological history, the results suggest that as a
261
system ages and incorporates new drainage area, additional relief is typically captured as well. This is most apparent
262
by comparing young versus mature passive margins (Fig. 6B) that differ in basin maturity and hence subsidence
263
from crustal thinning and mantle-lithospheric thickening (Woodcock, 2004; Ingersoll, 2012). The lack of correlation
264
of these parameters on the global dataset suggests the individuality of S2S systems that are in different stages of
265
steady state between uplift and erosion. The results reaffirm the connected tectonic and sedimentological history that
266
define the morphology of source-to-sink systems. As reiterated by D’Archy and Whittaker (2014), it is often
AC
CE
PT
ED
M
AN
US
CR
IP
T
241
10
ACCEPTED MANUSCRIPT difficult to decouple tectonic and climatic influences on a system, such as relief and precipitation, given their inter
268
related nature.
269
The proposal that the proportion of low relief area (10 m/km) increases with catchment size (Sømme et al., 2009) is
270
in disagreement with the current study (Fig. 5). General observations indicate that larger systems (> ~100 x 10 3 km2)
271
have a larger low-relief area (> ~10%) whereas smaller catchments have a more variable range. This increased
272
variability likely reflects that smaller systems incorporate both intra-montane high relief regions and systems
273
originate on plains that are of lower relief, while large catchments naturally incorporate more low-relief terrain.
274
Furthermore, smaller tectonically active systems could represent a range of systems in different stages of basin
275
development whereas larger systems are typically older and may then represent later stages of basin development.
276
The topographic distribution in relief of small catchments can significantly influence the along-strike sediment load
277
contribution to the marine environment as Pechlivanidou et al. (2017) have observed in a Holocene S2S study of the
278
Serchios rift in central Greece. Similarly, Sharman et al. (2017) have concluded that a major early Cenozoic increase
279
in sediment volume delivery to the Gulf of Mexico was caused by a river capture of a catchment that was modest in
280
size but included an area of high topographic relief.
281
The tectonic history of a system controls the magnitude and distributions of uplift, and the creation of
282
accommodation by a combination of crustal thinning, mantle-lithospheric thickening, sedimentary and volcanic
283
loading, tectonic loading, subcrustal loading, asthenospheric flow and crustral densification, which are different
284
though not unique to any specific tectonic regime (Ingersoll, 2012). Combined with basin infilling history and sea
285
level fluctuations, the relation of accommodation and sediment supply control the boundaries of the low relief
286
terrestrial sink area (Blum and Tornqvist, 2000; Nyberg and Howell, 2015). Thus, the controlling factors of
287
catchment relief and low relief area vary greatly from basin to basin and the results here consequently show that it is
288
difficult to define catchment relief by morphological relationships alone (e.g., Fig. 5). Defining relief and
289
topography in eroded catchments of ancient sedimentary systems remains a source of high uncertainty in the
290
reconstruction of paleo-S2S morphology (e.g., Sømme et al., 2013; Eide et al., 2017; Sharman et al., 2017).
291
The results seem to contradict findings that the average river gradient of a S2S system profile decreases with
292
increasing catchment area (e.g., Sømme et al., 2009) (Fig. 5). This data suggests a larger low-gradient plain for
293
larger S2S systems. However, considering that catchment area is strongly correlated to river length and river profile
AC
CE
PT
ED
M
AN
US
CR
IP
T
267
11
ACCEPTED MANUSCRIPT gradient relates to fall over river length, this result reflects that larger catchments have longer rivers relative to the
295
amount of catchment relief. The river profile gradient does not consider the proportion of high and low relief areas,
296
which, as discussed above, is quite variable (Fig. 5).
297
4.2. The Continental Shelf
298
Morphological relationships specific to continental shelf margins linked to 52 submarine fans are summarized in
299
Figure 7. The results suggest that river length and thereby also catchment area and sediment discharge (e.g.,
300
Milliman and Syvitski, 1992; Syvitski and Milliman, 2007) have weak power law correlation with increasing shelf
301
width. Shelf width does not correlate with continental slope length either and correlations do not improve when
302
analyzing the larger S2S database. These results disagree with the previous studies of Somme et al. (2009), Carvajal
303
et al. (2009) and Helland-Hansen et al. (2012) by suggesting sediment discharge is not a determining control on
304
shelf width. While sediment supply may have a considerable influence on shelf growth (e.g., Carvajal et al., 2009;
305
Ingersoll, 2012), the same plate-tectonic controls that influence the variability in proportion of low-lying relief areas
306
(Fig. 4) extend to the continental shelf (e.g., Nyberg and Howell, 2016). Factors controlling subsidence and relative
307
sea level change (e.g., accommodation versus sediment supply) is a key role in the transient position of the present-
308
day shoreline and hence shelf width (e.g., Blum and Womack, 2009). This influences morphological relationships
309
between catchment length, catchment area, sediment discharge, and a measurement of continental shelf width.
310
However, absolute measurement of shelf width do differ between active versus passive continental shelf/slope
311
margins (Fig. 8). This reflects previous studies that have shown that passive margins and foreland settings have, in
312
general, larger continental shelves than short-lived extensional, strike-slip or forearc settings (e.g., Nyberg and
313
Howell, 2016; Harris and Macmillan-Lawler 2016). This would suggest that the interplay between tectonics and
314
sedimentation have a greater influence on width of the continental shelf than previously assumed by S2S models.
315
For instance, passive continental margins are characteristic of crust and lithosphere thinning and thermal cooling as
316
it matures, while later sediment loading influences its subsidence (Osmundsen and Redfield, 2011; Ingersoll, 2012).
317
Forearc settings are controlled by rates, orientation and conditions of the subsiding plate combined with sediment
318
supply to its basin and trench (Dickinson and Seely, 1997; Fuller et al., 2006; Ingersoll, 2012). Thus, comparing the
319
dynamic factors of basin creation to one aspect of its morphological character and relation to sediment supply is too
320
simplistic.
AC
CE
PT
ED
M
AN
US
CR
IP
T
294
12
ACCEPTED MANUSCRIPT 321 Gradient of the continental shelf shows a slight power law increase with increasing river gradient (Fig. 7). These
323
results suggest that steeper, smaller S2S systems along active tectonic margins (e.g., Red Sea, Taiwan, Andes) have
324
a greater tectonic influence on its continental shelf thereby potentially steepening and narrowing the continental
325
shelf thus reducing the available accommodation for the supplied sediment. In addition, catchment systems that
326
experience infrequent but high magnitude discharge events may directly bypass sediment on the continental to its
327
basin floor sink (Milliman and Syvitski, 1992; Helland-Hansen et al., 2016). This is supported by sediment thickness
328
maps that generally suggest higher sediment accumulations along long-lived passive versus short-lived active
329
continental shelf/slope margins (Woodcock, 2004; Harris and Whiteway, 2011; Ingersoll, 2012) (Fig. 8).
330
In comparison, larger catchments that contain larger terrestrial sinks have longer transient times of sediment to its
331
marine realm (Allen et al., 2013; Romans et al., 2016; Nyberg et al., 2018). This produces a steady discharge of
332
sediment to create a series of low gradient clinoform packages (Carvajal et al., 2009; Helland-Hansen et al., 2016)
333
on continental shelves with available accommodation. Furthermore, the continental shelf may be an expression of
334
sedimentary wedges produced by its fluvial counterpart during periods of sea level lowstand (Blum and Tornqvist,
335
2000). The morphology of the onshore catchment and its continental shelf in relation to the nature of base-level
336
lowering will influence the response and character of the fluvial system (Schumm, 1993; Blum and Tornqvist,
337
2000). The gradient of the continental shelf is therefore a combined expression of tectonic, sedimentological and
338
base level change.
339
Estuarine length is shown to increase with increasing continental shelf width (Fig. 7). This would be expected
340
considering wide shelves amplify tidal currents, which in turn increase restricted and funnel-shaped shoreline
341
morphologies (e.g., Cram, 1979; Ainsworth et al., 2011). Systems with large low gradient terrestrial sink may easily
342
accommodate the remobilization of coastal sediments into funneled shaped shorelines, in comparison to steeper
343
gradient catchment systems (Nyberg and Howell, 2016). A longer estuary will increase the potential sediment transit
344
times along the sediment routing system to its submarine fan (e.g., Meade 1969; Kineke et al., 1996; Goodbred and
345
Kuehl, 1999). This is especially relevant during sea level highstands and might help explain high sediment trapping
346
efficiencies on present-day continental shelf (Kineke et al., 1996; Goodbred and Kuehl, 1999). For instance, Bostock
347
et al. (2007) calculated that more than 50% of the total annual suspended sediment load of the Fitzroy River in
AC
CE
PT
ED
M
AN
US
CR
IP
T
322
13
ACCEPTED MANUSCRIPT Queensland, Australia, is trapped in lower floodplains and estuaries. The Ganges/Brahmaputra river system traps
349
more than 30% of its total annual sediment load in its lower floodplains and estuaries (Goodbred and Kuehl, 1999).
350
4.3. The Continental Slope
351
Morphological parameters measured on the continental slope show a lack of any statistical correlation to other
352
segments of the S2S system (Fig. 9). Continental slope length is not empirically related to river length, river
353
gradient, catchment area or shelf width. The lack of correlation between morphological parameters, and hence
354
sedimentation rates and continental slope length, suggests a more complex interplay between plate tectonics and
355
sedimentation history than previously asserted in S2S models (e.g., Sømme et al., 2009; Helland-Hansen et al.,
356
2016). In a review of modern and paleo-continental slopes, Prather et al. (2017) notes the wide range of controls on
357
morphology of continental slope profiles including: sediment progradational patterns, erosional continental slope
358
margins, rapid sea level fluctuations, tectonic over-steepening and alternating siliclastics versus carbonate
359
deposition. The gradient of the continental slopes varies considerably from 1˚ to 10˚ and the base-of-slope transition
360
to the basin floor is often imprecisely defined (Prather et al., 2017). Our current study therefore discourages use of
361
the continental slope dimensions for scalingother segments.
ED
M
AN
US
CR
IP
T
348
362
Interestingly, absolute values of continental slope length and depth do not seem to differ significantly either by
364
tectonic regime or active versus passive continental shelf/slope margins with the exception of forearc settings (Fig.
365
10). Similar slope lengths and depths yield comparable gradients though the destruction and reworking of
366
sedimentary wedges along subducting zones of forearc systems considerably deepen and steepen those continental
367
slopes (Fig. 10). The lack of a clear morphological distinction of continental slope length may then reflect the
368
difficulty in defining the transition between slope and rise (Fig. 3).
AC
CE
PT
363
369 370
Harris and Whiteway (2011) showed that submarine canyons that contribute to the abyssal plains have higher
371
frequency of canyons and shorter lengths along active continental margins compared to canyons on passive margins.
372
Extending this analysis in the current study by active versus passive continental shelf/slope margin for submarine
373
canyons contributing to the 52 submarine fans suggest comparable results (Fig. 11). Active margins have a higher
14
ACCEPTED MANUSCRIPT frequency of canyon heads in comparison to passive continental slopes. The lengths of those canyons are also on
375
average shorter on active versus passive continental slope margins when comparing S2S systems of similar
376
catchment tectonics. Longer submarine canyons along passive continental slope margins are typically incised on the
377
continental shelf (Harris and Whiteway, 2011) and run the length of its continental slope and rise. More active
378
continental slopes show shorter submarine canyons on steeper slopes. However, active faulting, particularly during
379
early rifting, may divert and lengthen submarine canyons along relay ramps (e.g., Athmer and Luthi, 2011), which
380
may explain the considerable variability in submarine canyon lengths. Given that slope lengths are comparable
381
between active versus passive continental slope margins (Fig. 10) while canyon lengths differ, the data suggest that
382
canyons along active margins may not extend the length of its corresponding continental slope (Fig. 11).
CR
IP
T
374
US
383 4.5. The Basin Floor
385
Measurements in this study show comparable first-order relationships of submarine fan dimensions to catchment
386
properties as the ones proposed by Wetzel (1993) and Sømme et al. (2009). Submarine fan width and length show a
387
strong power law correlation (Fig. 12), in agreement with the original work by Wetzel (1993). The length of the
388
submarine fan increases by a power law correlation to the length of the longest river channels of contributing
389
catchments (e.g., Fig. 2) and submarine fan area shows a power law relationship with increasing contributing
390
catchment area (Fig. 12). In an analysis of 40 Paleocene to Pleistocene deepwater fan systems in the Gulf of Mexico,
391
Snedden et al. (2018), showed that submarine fan dimensions mapped from seismic reflections and well data
392
correlate to paleogeogaphic reconstructions of onshore catchment area, in agreement with modern empirical scaling
393
relationships. Given the strong correlation between catchment area and suspended sediment load (Syvitski and
394
Milliman|, 2007), submarine fan area expands with increasing sediment load of a system (Qs) (Fig. 12). The distal
395
depths of submarine fans do not differ by tectonic setting, however, longer fans associated with larger S2S systems
396
with similar fan depths result in a power law decrease in fan gradient with increasing fan area (Fig. 12).
397
An important observation of the current study shows proximal to distal S2S morphological and thus
398
sedimentological correlations should relate the basin floor to its broader contributing terrestrial region (Fig. 2),
399
rather than its largest catchment, as originally proposed by Wetzel (1993). In certain circumstances, the total
400
contributing catchment area can be significant. For example, during the current sea level highstand, 46% of the
AC
CE
PT
ED
M
AN
384
15
ACCEPTED MANUSCRIPT catchment area contributing to the Bengal Fan and 16% of the estimated sediment load (calculated using the
402
BQART approach of Milliman and Syvitski 2007) that reaches the Bengal Fan comes from catchments other than
403
that of the Ganges/Brahmaputra Rivers (Fig. 2). During lower than at present sea levels, paleogeographic
404
reconstructions suggest new drainage patterns can similarly increase sediment contribution to a basin floor region
405
from multiple river drainage systems rather than a single coalesced river. For instance, Holocene sea level
406
reconstructions of the Sunda Shelf have shown that lower sea levels than at present supported four large river
407
systems contributing to the South China Sea (Voris, 2000). Calculating the merger of catchments at the continental
408
shelf edge based on present-day bathymetry of the Sunda Shelf, indicates that over 30% of the contributing
409
catchment area to the South China Sea is derived from catchments other than the Mekong River. The increased
410
contributing catchment area will likely increase the sediment input to the basin floor. U–Pb and Pb–Pb dating of
411
detrital zircons can provide a valuable tool to assess the different potential source regions of the contributing
412
catchment area (e.g., Blum and Pecha, 2014; Sharman et al., 2015; Helland-Hansen et al., 2016).
413
5. The Source-to-Sink Profile
414
The morphology of each source-to-sink segment defines the topographic and bathymetric profiles of 52 S2S systems
415
stretching from the highest hinterland point to the most distal point of the submarine fan (Figure 13; see Fig. 4 for
416
location; see also supplementary kmz). The data show the variable morphological nature of S2S systems across the
417
spectrum of tectonic regimes. Visually, the relative river length (i.e., terrestrial sink, < 10/km relief zone and source
418
region) increases with increasing catchment area while the relative continental shelf width and slope length
419
decreases. In contrast, smaller systems that are common along tectonically-active margins are represented by a
420
proportionally larger continental shelf and slope length. These results adhere to the earlier findings showing a strong
421
power law correlation of increasing catchment area and river length (Fig. 5) whereas absolute continental shelf (Fig.
422
7) and continental slope (Fig. 9) dimensions do not significantly change with the size of the catchment.
423
These observations are reiterated by mapping the proportional distance (0-100%) of the river, continental shelf,
424
continental slope and submarine fan S2S profile by catchment area (Fig. 14). The negative power-law correlations
425
between catchment area and proportion of continental shelf and continental slope length suggest that smaller
426
systems have proportionally a longer distance to transport sediment from its river mouth to its basin floor (i.e., a
427
longer relative continental shelf and continental slope component). On the other hand, the relative proportion of the
AC
CE
PT
ED
M
AN
US
CR
IP
T
401
16
ACCEPTED MANUSCRIPT submarine fan length to catchment area is more varied despite strong absolute correlations of contributing catchment
429
area and fan dimensions (Fig. 12). This suggests that the relative length of submarine fans in unconfined basin floors
430
do not proportionally correlate to the length of the entire S2S system as the submarine fan dimensions only show a
431
strong positive correlation to the contributing catchment region (Fig. 12) rather than continental shelf width or slope
432
length (Fig. 9).
433
5.1 The Transient Source-to-Sink Segment Morphology
434
Relative sea level fluctuations on timescales less than the lifespan of most basins (< 106 yrs) may significantly
435
influence the S2S topographic/bathymetric profile and morphology. During a relative sea level fall, the contributing
436
catchment area will expand, resulting in an increase in river length, low-relief terrestrial area, sediment discharge
437
and sediment bypass on the continental shelf (e.g., Schumm, 1993; Blum and Tornqvist, 2000) (Fig. 15). River
438
profile gradient will decrease as the length of the river increases relative to the amount of additional relief obtained
439
from any sea level fall. The volume of sediment stored on the terrestrial realm may thus increase due to the increase
440
in low-lying catchment area, while sediment stored on the continental shelf will decrease. This will change the
441
proportion between semi-transient terrestrial sink and continental shelf sediment storage.
442
The amount of sediment stored within the terrestrial realm and continental shelf can be significant. For instance,
443
present-day storage within the Ganges-Brahmaputra delta/floodplain may be as high as 30% of its total suspended
444
sediment load, while ~60% of sediment that reaches the Bengal Sea is stored on its continental shelf and a remaining
445
~40% reaches the Swatch of No Ground canyon (Goodbred and Kuehl, 1999). Aalto et al. (2006) proposed that 50%
446
of the total Andean sediment budget is stored in the terrestrial Amazon basin while as much as 100% of sediment
447
that reaches the Atlantic is stored on its continental shelf (Kineke et al., 1996). In consequence, early work of
448
Shanmugam et al., 1985 and Posamentier et al., 1991 suggested that submarine fans are principally derived during
449
periods of sea level lowstands when a river may extend to the continental shelf edge and deliver sediment directly to
450
the basin floor.
451
Observations of modern rivers support a low stand increase in sediment load due to an increase in catchment area
452
(e.g., Milliman and Syvitski, 1992; Syvitski and Milliman, 2007) despite the variability of low relief area (Fig. 5).
453
This suggests that while terrestrial sediment storage may increase during a relative sea level fall, a larger catchment
AC
CE
PT
ED
M
AN
US
CR
IP
T
428
17
ACCEPTED MANUSCRIPT area will typically increase annual sediment load to the marine environment. A decrease in sea level can furthermore
455
promote increased catchment denudation upstream as the river tries to restore an equilibrium caused by local
456
steepening from channel incision and avulsion (Schumm, 1993; Blum and Tornqvist, 2000). However, more recent
457
studies suggest that submarine fans can build regardless of sea level stand and are defined by high magnitude
458
tectonic and climatic perturbations transporting new, and remobilizing previously stored sediments (e.g., Covault
459
and Graham, 2010; Simpson and Castelltort, 2012; Shanmugam, 2016). A larger low stand terrestrial sink may then
460
increase the amount of terrestrial sediments available for remobilization and directly connect the system to the
461
continental shelf edge but not necessarily provide the millennial scale events that are important for the stratigraphic
462
record.
463
S2S systems with large terrestrial sinks and shallow wide continental shelves, especially along epicontinental
464
seaways (e.g., Gulf of Thailand, Gulf of Carpanteria, Adriatic Sea), would experience the greatest morphological
465
change because of sea level change (Fig. 15). In contrast, short and steep systems with higher gradient catchments
466
and continental shelves would only marginally alter catchment area, river length and continental shelf width with a
467
similar sea level change (Fig. 15). Hence, the continental shelf may be better considered as a transient geomorphic
468
feature rather than a segment of a S2S system.
469
Tectonic activity that typically occurs within the source region operates on a variety of temporal and spatial scales,
470
may alter the relative sea level, either connect or disconnect drainage area, while relief of the system may increase or
471
decrease to reflect that tectonic history (e.g., Pechlivanidou et al., 2017; Sharman et al., 2017). Morphological
472
relationships in the current study suggest that variations in catchment area will influence the distal submarine fan
473
area (Fig. 12) and sediment discharge will change given the strong influence of catchment area and relief on river
474
sediment load (Syvitski and Milliman, 2007). Small tectonically active systems (e.g., strike-slip, extensional) would
475
be most susceptible to these changes due to active faulting in these short-lived systems. In addition, the region of
476
tectonic influence may be proportionally greater in small systems compared to larger passive/foreland systems
477
thereby greatly influencing the morphology of those S2S system (Fig. 15).
AC
CE
PT
ED
M
AN
US
CR
IP
T
454
478 479
18
ACCEPTED MANUSCRIPT 480 6. The Source-to-Sink Model
482
Helland-Hansen et al. (2016) proposed a simplified three end members of S2S systems that drain to the ocean:
483
“steep, short and deep”, “wide and deep” and “wide and shallow” systems. Here we expand and revise the
484
classification based on modern morphological characteristics and relationships presented in the current study.
485
Furthermore, we discuss the change in S2S segment morphology on timescales influencing a sedimentary basin
486
infill history (>104 to <106 yrs).
487
The classification of S2S systems below considers individual catchment - continental shelf - continental slope and
488
basin floor configurations. In studying S2S systems, it is imperative to consider the larger contributing catchment
489
area (e.g., Fig. 12) with the understanding that downstream segments may represent an amalgamation of different
490
source regions. The larger contributing catchment area may subsequently represent a combination of different S2S
491
models. For example, the western part of the Bay of Bengal fan (Fig. 2) is fed by a number of wide and deep passive
492
tectonic regimes whereas the eastern parts are increasingly represented by wide and deep foreland, and steep, short
493
and narrow forearc tectonic regime S2S systems. The variability in S2S profiles contributing to one basin-floor fan
494
will influence the distribution and mode of sediment transport to the submarine fan.
495
A general rule, as with using any modern analog, is to view the morphological relationships and characteristics as a
496
model of S2S system scaling and apply that knowledge to system-specific conditions. Furthermore, it is important to
497
note that some of the morphological relationships would potentially break down in systems of topographically
498
confined basin floors, such as along subduction zones (e.g., Andes), of terminal oceanic basins (e.g., Mediterranean
499
Sea), or salt-confined basins (e.g., Plio-Pleistocene, Gulf of Mexico), where the submarine fan is being restricted or
500
destroyed.
501
6.1 Steep, Short and Deep Systems
502
Steep, short and deep systems are found on active margins and contain steep and short catchments bordering a basin
503
floor (Helland-Hansen et al., 2016) of a young or contracting ocean basin (Harris and Macmillan-Lawler, 2018). All
504
steep, short and deep systems are characterized by a high sediment yield (Milliman and Syvitski, 1992) with narrow
505
shelves (<35 km; Nyberg and Howell, 2016), shallow shelf breaks (<180 m; Harris and Macmillan-Lawler, 2016),
AC
CE
PT
ED
M
AN
US
CR
IP
T
481
19
ACCEPTED MANUSCRIPT medium long continental slope lengths (~60 km) and deep basin floors (< 3km; Fig. 16A). Steep, short and deep S2S
507
systems of forearc tectonic regimes are distinguished by small terrestrial sinks (Nyberg and Howell, 2015) fed by
508
few but large source catchments relative to the total source region area (Nyberg et al., 2018). The S2S systems are
509
long-lived (~107 - 108 yrs; Woodcock, 2004; Ingersoll 2012), fed by numerous and steep submarine canyons that are
510
shelf and slope incised (e.g., Andes; Hagen et al., 1994; Harris and Whiteway, 2011) (Fig. 11) and associated with a
511
basin floor with limited preservation potential (Harris and Macmillan-Lawler, 2018). The areally limited terrestrial
512
sinks suggests that these systems will likely deposit sediment directly on the continental shelves. A lack of sediment
513
thickness on the continental shelf and slope (Fig. 8) suggests either an immaturity of those systems and/or the
514
destruction of bypassed sediments on the basin floor.
515
Steep, short and deep systems along oceanic transform margins including strike-slip and extensional tectonic
516
regimes are short lived (< 107 yrs; Woodcock; 2004; Ingersoll, 2012) although the young basin floor may have
517
longer-term preservation (> 107 yrs; Harris and Macmillan-Lawler, 2018). These systems are characterized by small
518
terrestrial sinks fed by few but large source catchment regions relative to the total source region area (Nyberg et al.,
519
2018) that transport sediment to the basin floor through several steep but short submarine canyons (Figs. 11, 16B).
520
The shorter terrestrial sediment transit times in the areally insignificant terrestrial sinks (Nyberg and Howell, 2015)
521
will likely bypass sediment to the marine realm as suggested by thinner sediment packages (Fig. 8). The distal parts
522
of the S2S system will furthermore experience sediment routing pathway avulsions as the systems develops and
523
matures (e.g., Red Sea; Gulf of California; Ingersoll, 2012).
524
Steep and short endorheic systems are furthermore present in continental extensional and strike-slip tectonic
525
regimes, characterized as short lived (< 107 yrs) given the continual tectonic destruction and creation of new
526
accommodation (Gawthorpe and Leeder, 2000; Ingersoll, 2012). In this circumstance, the system will terminate on a
527
relatively larger terrestrial sink (Nyberg and Howell, 2015) fed by numerous yet smaller source catchment regions
528
when comparing to the entire basin drainage (Nyberg et al., 2018).
529
On shorter timescales (< 106 yrs), the primary control on morphological change of steep, short and deep systems is
530
tectonic activity where faulting may influence a large proportion of the overall S2S system to change relief and
531
catchment area (Fig. 15). Tectonics furthermore play a key role in the sediment routing pathways of these systems
532
by controlling initiation, diversion and potentially termination of sediment transport to the basin floor (Gawthorpe
AC
CE
PT
ED
M
AN
US
CR
IP
T
506
20
ACCEPTED MANUSCRIPT and Leeder, 2000; Athmer and Luthi, 2012). In comparison, relative sea level change on narrow and steep
534
continental shelves with smaller low-lying terrestrial sinks will not significantly alter the scale and morphology of
535
the S2S system, as illustrated in Figure 15B.
536
6.2 Wide and Deep Systems
537
Wide and deep systems are typical of the mature remnants of pre-existing rifted margins (Kingston et al., 1983;
538
Osmundsen and Redfield, 2011) or at margins of convergent plate tectonic boundaries (Ingersoll, 2012). These S2S
539
systems differ in that they are long-lived (> 108 yrs; Woodcock, 2004) and develop large catchments producing
540
significant sediment discharge (Syvitski and Milliman, 2007), medium-wide continental shelves (>50 km; Nyberg
541
and Howell, 2016) with deep shelf breaks (>200 m; Harris and Macmillan-Lawler, 2016), medium continental slope
542
lengths (~60 km), few but long submarine canyons (Harris and Whiteway, 2011) (Fig. 11) and deep basin floors
543
(~3.9 km; Fig. 16C). Foreland tectonic regimes (Fig. 16C) are portrayed by a relatively large source region, a small
544
low gradient sediment bypass zone (~30% of total catchment area < 10m/km) that includes a large terrestrial sink
545
fed by numerous smaller source catchments (Nyberg and Howell, 2015; Nyberg et al., 2018). In contrast, passive
546
margins (Fig. 16D) have a continental hinterland composed of a relatively small mountainous source region, a large,
547
low gradient sediment bypass zone (~48% of total catchment area <10 m/km) that includes a medium-sized
548
terrestrial sink fed by few but large source catchments (Nyberg et al., 2018).
549
Relative sea level change will have a greater influence on catchment area, river length and averaged river gradient
550
profile of wide and deep systems given the larger terrestrial sinks and wider continental shelves. The amount of
551
morphological change will be less significant for larger systems where the relative length of its continental shelf
552
may be insignificant compared to the length of the entire system size (Figs. 13, 15, 16). Tectonic activity on shorter
553
timescales (< 106 yrs) resulting in small relief variations or source catchment diversions would not significantly
554
change the overall morphology of wide and deep S2S systems (Figs. 13, 15).
555
6.3 Wide and Shallow Systems
556
Wide and shallow systems are typical of epicontinental seaways without a traditional shelf-slope-basin floor-fan
557
configuration. These systems are long-lived, have large catchments and wide shelves of low gradient bathymetry
558
(Harris and Macmillan-Lawler, 2016). Intracratonic systems along high-latitude regions, (e.g., Hudson Bay, North
AC
CE
PT
ED
M
AN
US
CR
IP
T
533
21
ACCEPTED MANUSCRIPT Sea, Arctic Sea) tend to have a low relief hinterland, no terrestrial sink (Nyberg and Howell, 2015), deep glaciated
560
troughs with shelf breaks as deep as 360 m (Harris and Macmillan-Lawler, 2016) and low sediment discharge
561
systems (Milliman and Farnsworth, 2011). The lack of sediment discharge, and small terrestrial sinks, is in part
562
related to present-day climate of high-latitude regions that show significant isostatic uplift since the last glacial
563
maximum (Sella et al., 2007). Low to mid-latitude intracratonic systems (e.g., Gulf of Carpanteria) along
564
epicontinental seaways have a similar morphological character without major glaciated troughs (Harris et al., 2014),
565
medium sized terrestrial sink (Nyberg and Howell, 2015) and typically a higher sediment discharge related to
566
warmer climates (Nyberg et al., 2018). In comparison, present-day configuration of foreland basins within
567
epicontinental seaways tend to have higher catchment relief, higher sediment discharge, larger terrestrial sinks with
568
variable source morphology relating to constricted catchments profiles. Their continental shelves are low gradient
569
and shallow with shelf breaks at approximately 200 m (Harris and Macmillan-Lawler, 2016) with thicker sediment
570
accumulations.
571
Endorheic systems are similar to their marine counterpart. Intracratonic systems show a low relief hinterland but
572
with a medium sized terrestrial sink component sourced by numerous small source catchments. By contrast, large
573
foreland type endoheric systems have a higher relief but are often associated with a large terrestrial sink component
574
fed by small, numerous source catchments (Nyberg et al., 2018). In many circumstance, the lithospheric loading of
575
the tectonic plates fades from the hinterland and the distal part of these systems are characteristic of an intracratonic
576
settings (DeCelles and Giles, 1996).
577
The morphology of wide and shallow systems along epicontinental seaways will fluctuate the most, subject to sea
578
level change. The shallow continental shelves (Harris and Macmillan-Lawler, 2016) that sometimes span hundreds
579
of kilometers (e.g., Gulf of Carpentaria, Gulf of Thailand) can significantly alter the catchment area, river length
580
river profile gradient from sea level change (Fig. 12). This reflects the dynamic nature of the continental shelf. As
581
with wide and deep systems, regional tectonic uplift (< 106 yrs) may have less influence given the size of the
582
absolute contributing catchment area of these S2S systems (Fig. 15).
583
7. Conclusions
AC
CE
PT
ED
M
AN
US
CR
IP
T
559
22
ACCEPTED MANUSCRIPT The holistic S2S approach is an attempt to relate the complexity of trigger and response of sedimentary processes
585
from the complete sediment routing perspective. By studying the scaling relationships between modern S2S
586
segments that have formed on geological timescales (>106 yrs), the long term influence of sedimentary processes on
587
the morphology of sedimentary systems can be understood. Scaling relationships of S2S segments would provide a
588
useful, first-order assessment to support paleogeographic reconstruction of ancient sedimentary systems where the
589
complete sediment routing system is not preserved. A restored paleogeography would thus offer insight into the
590
sedimentary processes and volume budgets contributing to the stratigraphic record. It has been the purpose of this
591
article to revisit the morphology and scaling relationships of modern S2S segments and discuss the implications for
592
scaling segments in ancient sedimentary systems. The results of our study have provided new insight into S2S
593
scaling relationships and sediment routing systems that imply:
CR
IP
1.
US
594
T
584
The properties of catchment area and river length scale internally for the catchment segment. However, characteristics of catchment topography, such as maximum relief and proportion of low relief area, is more
596
variable and system specific, which will introduce uncertainties in estimations of S2S sediment budgets. In
597
addition, the transient nature of the continental shelf will influence the dimensions of the catchment and
598
scale of downstream segments. In application to ancient S2S systems, scaling relationships are defined for
599
systems formed over geological timescales (>106 years) that incorporate higher frequency fluctuations in
600
catchment morphology and sedimentological characteristics.
M
ED
2.
PT
601
AN
595
Continental slope length is difficult to measure objectively due to the gradational transition to the abyssal plain. It is a sedimentological and tectonically controlled boundary that shows considerable variability.
603
Therefore, the use of the continental slope morphology to scale segments of a S2S systems should be
604
exercised with extreme caution. 3.
AC
605
CE
602
The catchment and submarine fan segments of the S2S system correlate for many morphological
606
parameters, suggesting the larger scale fan dimensions relate primarily to sediment transport to the basin
607
floor. The catchment size and its sediment transport capacity may then be a useful first-order parameter to
608
scale proximal to distal S2S segments. First-order assessment of submarine fan dimensions are a
609
contribution from multiple catchments to one basin floor region that are in a continuous state of coalescing
610
and dividing associated with the transient shoreline position. Thus, it is imperative that S2S studies
611
consider the larger region of contributing catchments which supply sediment to a specific submarine fan.
23
ACCEPTED MANUSCRIPT 612
4.
Models characterizing S2S systems as “Steep, short and deep”, “Wide and deep” and “Wide and shallow” show distinct morphological characteristics of its catchment, continental shelf, continental slope and
614
submarine fan by different tectonic regimes. However, the distinctions may not necessarily suggest scaling
615
relationship of morphological dimensions within a segment or between adjacent linked S2S segments. This
616
is particularly important along tectonically bounded S2S features of the continental shelf and continental
617
slope that are not appropriate to scale based on the morphology of other segments. Similar continental shelf
618
widths and continental slope lengths across a range of system sizes suggests a proportionally longer marine
619
sediment transport zone to the basin floor along sediment routing systems with decreasing catchment size.
IP
5.
CR
620
T
613
In the reconstruction of partially preserved S2S segments, morphological and sedimentological scaling relationships should be viewed as a first-order assessment of the proximal and distal portions of a sediment
622
routing system. The relationships may apply to any sea level stand of non-topographically confined basins
623
supplemented by other available analytical techniques in a holistic approach to improve system specific
624
conditions. For instance, detrital zircon analyses can improve knowledge on source region and volume,
625
while interpretation of depositional environment and tectonic reconstruction can further constrain probable
626
dimensions of S2S segments.
ED
M
AN
US
621
Future work needs to concentrate on the role of plate tectonics and basin development on S2S morphology and
628
scaling relationships. These issues appear to be a major control on the observed variability in the present study.
629
Forward modelling of landscapes and sedimentary systems may be an interesting approach to integrate S2S concepts
630
into a process-based framework that considers both sedimentation and tectonic influences. In conclusion, the
631
correlations (and sometimes low correlations), between morphology of S2S segments shows the significant amount
632
of research that remains in order to improve our understanding of sediment routing systems.
633
Acknowledgements
634
Statoil are thanked for supporting the main author under the Spatial-Temporal Reconstruction of Basin Fills project
635
(no. 810127) and RLG acknowledges support by VISTA. Reviewer John Snedden and editor Jasper Knight are
636
thanked for constructive comments that greatly improved this manuscript.
AC
CE
PT
627
637 24
ACCEPTED MANUSCRIPT 638
Appendix A - A comparison of parameters gathered on the catchment segment of modern S2S systems in the
639
current study versus previously reported databases of Sømme et al. (2009) and Milliman and Farnsworth (2011).
640
Sømme - Sømme et al. (2009), M & F (2011) - Milliman and Farnsworth (2011)
641 Appendix B - A comparison of parameters gathered on the continental shelf, continental slope and basin floor
643
segments of modern S2S systems in the current study versus manual observed measurements and previously
644
reported dimensions in Sømme et al. (2009). Sømme - Sømme et al. (2009)
CR
IP
T
642
US
645 References
647
Aalto, R., Dunne, T., Guyot, J.L., 2006. Geomorphic Controls on Andean Denudation Rates. Journal of Geology
649
114, 85–99.
M
648
AN
646
Ainsworth, R., Vakarelov, B., Nanson, R., 2011. Dynamic Spatial and temporal prediction of changes in depositional processes on clastic shorelines: Towards improved subsurface uncertainty reduction and
651
management. American Association of Petroleum Geologists Bulletin 95, 267–297. Allen, P.A., 2008. Time scales of tectonic landscapes and their sediment routing systems. In: Gallagher, K., Jones,
PT
652
ED
650
S.J., Wainwright, J. (Eds.), Landscape Evolution: Denudation, Climate and Tectonics Over Different Time
654
and Space Scales, Geological Society of London Special Publications, 296, pp. 7-28.
656 657
Allen, P.A., 2017. Sediment Routing Systems: The Fate of Sediment from Source to Sink. Cambridge University
AC
655
CE
653
Press, Cambridge.
Allen, P.A., Armitage, J.J., Carter, A., Duller, R.A., Michael, N.A., Sinclair, H.D., Whitchurch, A.L., Whittaker,
658
A.C., 2013. The Qs problem: Sediment volumetric balance of proximal foreland basin systems.
659
Sedimentology 60, 102–130.
660 661
Amante, C., Eakins, B., 2009. ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis. NOAA Techical Memo, NESDIS NGDC-24, 19 pp.
25
ACCEPTED MANUSCRIPT 662 663 664
Athmer, W., Luthi, S.M., 2011. The effect of relay ramps on sediment routes and deposition: A review. Sedimentary Geology 242, 1–17. Barnes N.E., Normark W.R., 1985. Diagnostic Parameters for Comparing Modern Submarine Fans and Ancient Turbidite Systems. In: Bouma A.H., Normark W.R., Barnes N.E. (Eds.), Submarine Fans and Related
666
Turbidite Systems. Frontiers in Sedimentary Geology pp. 13-14. Becker, J.J., Sandwell, D.T., Smith, W.H.F., Braud, J., Binder, B., Depner, J., Fabre, D., Factor, J., Ingalls, S., Kim,
IP
667
T
665
S.H., Ladner, R., Marks, K., Nelson, S., Pharaoh, A., Trimmer, R., von Rosenberg, J., Wallace, G.,
669
Weatherall, P., 2009. Global Bathymetry and Elevation Data at 30 Arc Seconds Resolution: SRTM30_PLUS.
670
Marine Geodesy 32, 355–371.
US
671
CR
668
Blum, M.D., Hattier-Womack, J., 2009. Climate change, sea-level change, and fluvial sediment supply to deepwater depositional systems. In: Kneller, B., Martinsen, O.J., McCaffrey, B. (Eds.), Externernal Controls on Deep-
673
Water Depositonal Systems. Society for Sedimentary Geology 92, p15–39.
676
M
675
Blum, M., Pecha, M., 2014. Mid-cretaceous to paleocene North American drainage reorganization from detrital zircons. Geology 42, 607–610.
ED
674
AN
672
Blum, M.D., Tornqvist, T.E., 2000. Fluvial responses to climate and sea-level change : a review and look forward. Sedimentology 47, 2–48.Bostock, H.C., Brooke, B.P., Ryan, D.A., Hancock, G., Pietsch, T., Packett, R.,
678
Harle, K., 2007. Holocene and modern sediment storage in the subtropical macrotidal Fitzroy River estuary,
679
Southeast Queensland, Australia. Sedimentary Geology 201, 321–340.
681 682 683 684 685 686
CE
Carvajal, C., Steel, R., Petter, A., 2009. Sediment supply: The main driver of shelf-margin growth. Earth-Science
AC
680
PT
677
Reviews 96, 221–248. Covault, J.A., Graham, S.A., 2010. Submarine fans at all sea-level stands: Tectono-morphologic and climatic controls on terrigenous sediment delivery to the deep sea. Geology 38, 939–942. Covault, J.A., Romans, B.W., 2009. Growth patterns of deep-sea fans revisited: Turbidite-system morphology in confined basins, examples from the California Borderland Marine Geology. 265, 51–66. Cram, J.M., 1979. The Influence of Continental Shelf Width on Tidal Range: Paleoceanographic Implications.
26
ACCEPTED MANUSCRIPT
688 689 690 691
Journal of Geology. 87, 441–447. Danielson, J.J., Gesch, D.B., 2011. Global multi-resolution terrain elevation data 2010 (GMTED2010). U.S. Geological Survey Open-File Report 2011–107, pp. 26. D’Arcy, M., Whittaker, A.C., 2014. Geomorphic constraints on landscape sensitivity to climate in tectonically active areas. Geomorphology 204, 366–381.
T
687
DeCelles, P.G., Giles, K.A., 1996. Foreland basin systems. Basin Research 8, 105–123.
693
Dickinson, W.R., and Seely, D.R., 1979. Structure and stratigraphy of forearc regions. American Association of
695
CR
Petroleum Geologists Bulletin 63, 2–31.
US
694
IP
692
Eide, C.H., Klausen, T.G., Katkov, D., Suslova, A.A., Helland-Hansen, W., 2017. Linking an Early Triassic delta to antecedent topography: Source-to-sink study of the southwestern Barents Sea margin. Geological Society of
697
Ammerica Bulletin 130, 263-283.
701 702
M
ED
700
zone earthquakes. Geology 34, 65–68.
Gawthorpe, R.L., Leeder, M.R., 2000. Tectono-sedimentary evolution of active extensional basins. Basin Research 12, 195–218.
PT
699
Fuller, C.W., Willett, S.D., Brandon, M.T., 2006. Formation of forearc basins and their influence on subduction
Goodbred, S.L., Kuehl, S.A., 1999. Holocene and modern sediment budgets for the Ganges-Brahmaputra river
CE
698
AN
696
system: Evidence for highstand dispersal to flood-plain, shelf, and deep-sea depocenters. Geology 27, 559–
704
562.
705 706 707 708 709 710
AC
703
Hack, J.T., 1957. Studies of longitudinal stream profiles in Virginia and Maryland. US Geological Survey of Professional Papers., 294-B, 1. Hagen, R.A., Bergersen, D.D., Moberly, R., Coulbourn, W.T., 1994. Morphology of a large meandering submarine canyon system on the Peru-Chile forearc. Marine Geology 119, 7–38. Harris, P.T., Macmillan-Lawler, M., 2016. Global Overview of Continental Shelf Geomorphology Based on the SRTM30_ LUS 30-Arc Second Databas. In: Finkl, C.W., Makowski, C. (Eds.), Seafloor Mapping along
27
ACCEPTED MANUSCRIPT 711
Continental Shelves: Research and Techniques for Visualizing Benthic Environments. Coastal Research
712
Library 13, pp. 169–190.
718 719 720 721 722
T IP
717
active and passive continental margins. Marine Geology 285, 69–86.
Harris, P.T., Macmillan-Lawler, M., Rupp, J., Baker, E.K., 2014. Geomorphology of the oceans. Marine Geology
CR
716
Harris, P.T., Whiteway, T., 2011. Global distribution of large submarine canyons: Geomorphic differences between
352, 4–24.
Helland-Hansen, W., Steel, R.J., Somme, T.O., 2012. Shelf genesis revisited. Journal of Sedimentary Research 82,
US
715
A., Krastel, S., Savini, A. (Eds.), Submarine Geomorphology. Springer Geology, pp. 111–134.
133–148.
AN
714
Harris, P.T., Macmillan-Lawler, M., 2018. Origin and Geomorphic Characteristics of Ocean Basins. In: Micallef,
Helland-Hansen, W., Sømme, T.O., Martinsen, O.J., Lunt, I., Thurmond, J., 2016. Deciphering Earth’s Natural Hourglasses: Perspectives On Source-To-Sink Analysis. Journal of Sedimentary Research 86, 1008–1033.
M
713
Hovius, N., 1996. Regular spacing of drainage outlets from linear mountain belts. Basin Research 8, 29–44.
724
Ingersoll, R.V., 2012. Tectonics of sedimentary basins, with revised nomenclature. In: Busby, C., Azor, A. (Eds.),
727 728 729 730 731 732 733 734
PT
Inman, D.L., Nordstrom, C.E., 1971. On the Tectonic and Morphologic Classification of Coasts. Journal of Geology 79, 1–21.
CE
726
Tectonics of Sedimentary Basins: Recent Advances. Wiley-Blackwell , Chichester, pp. 1-43.
Kineke, G.C., Sternberg, R.W., Trowbridge, J.H., Geyer, W.R., 1996. Fluid-mud processes on the Amazon
AC
725
ED
723
continental shelf. Continental Shelf Research 16, 667–696. Kingston, D.R., Dishroon, C.P., Williams, P.A., 1983. Global Basin Classification System. American Association of Petroleum Geologists Bulletin 67, 2175–2193. Lehner, B., Verdin, K., Jarvis, A., 2008. New Global Hydrography Derived From Spaceborne Elevation Data. EOS, Transactions, American Geophysical Union 89, 93–94. Meade, R.H., 1969. Landward transport of bottom sediments in estuaries of the Atlantic coastal plain. Journal of
28
ACCEPTED MANUSCRIPT 735 736
Sedimentary Research 39, 222-234. Miall, A., 2014. The Emptiness of the Stratigraphic Record: A Preliminary Evaluation of Missing Time In the
737
Mesaverde Group, Book Cliffs, Utah, U.S.A . Journal of Sedimentary Research 84, 457–469.
738
Milliman, J., Farnsworth, K., 2011. River Discharge to the Coastal Ocean: A Global Synthesis. Cambridge
742 743
T
IP
741
Milliman, J.D., Meade, R.H., 1983. World-Wide Delivery of River Sediment to the Oceans. Journal of Geology. 91, 1–21.
CR
740
University. Press. Cambridge, 392pp.
Milliman, J.D., Syvitski, J.P.M., 1992. Geomorphic/Tectonic Control of Sediment Discharge to the Ocean: The Importance of Small Mountainous Rivers. Journal of Geology. 100, 525–544.
US
739
Moore, G.T. 1969. Interaction of rivers and oceans--Pleistocene petroleum potential. American Association of
745
Petroleum Geologists Bulletin 53, 2421-2430.
746
Normark, W.R., 1970. Growth patterns of deep-sea fans. American Association of Petroleum Geologists Bulletin 54,
747
2170–2195.
748
Nyberg, B., Howell, J.A., 2015. Is the present the key to the past? A global characterization of modern sedimentary
751 752 753 754 755 756
M
ED
PT
Nyberg, B., Howell, J.A., 2016. Global distribution of modern shallow marine shorelines. Implications for
CE
750
basins Geology 43, 643-646.
exploration and reservoir analogue studies. Marine and Petroleum Geology 71, 83-104. Nyberg, B., Gawthorpe, R. L., Helland-Hansen, W., 2018. The distribution of rivers to terrestrial sinks: Implications
AC
749
AN
744
for sediment routing systems. Geomorphology 316, 1-23. Osmundsen, P.T., Redfield, T.F., 2011. Crustal taper and topography at passive continental margins. Terra Nova 23, 349–361. Pechlivanidou, S., Cowie, P.A., Hannisdal, B., Whittaker, A.C., Gawthorpe, R.L., Pennos, C., Riiser, O.S., 2017.
757
Source-to-sink analysis in an active extensional setting: Holocene erosion and deposition in the Sperchios rift,
758
central Greece. Basin Research 30, 522-543.
29
ACCEPTED MANUSCRIPT 759
Posamentier, H.W., Erskine, R.D., and Mitchum, R.M., Jr., 1991, Submarine fan deposition within a sequence
760
stratigraphic framework. In: Weimer, P., Link, M.H., (Eds), Seismic facies and sedimentary processes of
761
subma- rine fans and turbidite systems. Frontiers in Sedimentary Geology, pp. 127-136.
768 769 770 771 772 773
T
IP
CR
767
Rigon, R., Rodriguez-Iturbe, I., Maritan, A., Giacometti, A., Tarboton, D.G., Rinaldo, A., 1996. On Hack’s law. Water Resources Research 32, 3367–3374.
US
766
International Journal of Geographical Information Science 21, 983–1008.
Romans, B.W., Castelltort, S., Covault, J.A., Fildani, A., Walsh, J.P., 2016. Environmental signal propagation in sedimentary systems across timescales. Earth-Science Reviews 153, 7–29.
AN
765
Reuter, H.I., Nelson, A., Jarvis, A., 2007. An evaluation of void‐filling interpolation methods for SRTM data.
Ross, W.C., Watts, D.E. & May, J.A.,1995. Insights from stratigraphic modeling: mud-limited versus sand-limited
M
764
slope systems. Basin Research 29, 394–416.
depositional systems. American Association of Petroleum Geologists Bulletin 79, 231–258.
ED
763
Prather, B.E., O’Byrne, C., Pirmez, C., Sylvester, Z., 2017. Sediment partitioning, continental slopes and base-of-
Schumm, S.A., 1993. River response to baselevel change: implications for sequence stratigraphy river response to baselevel change: implications for sequence stratigraphy, Journal of Geology, 101, 279-294.
PT
762
Sella, G.F., Stein, S., Dixon, T.H., Craymer, M., James, T.S., Mazzotti, S., Dokka, R.K., 2007. Observation of
775
glacial isostatic adjustment in “stable” North America with GPS. Geophysical Research Letters 34, 1–6.
777 778
Shanmugam, G., 2016. Submarine fans: A critical retrospective (1950–2015), Journal of Palaeogeography 5, 110184.
AC
776
CE
774
Shanmugam, G., Moiola, R.J., Damuth, J.E., 1985. Eustatic Control of Submarine Fan Development. In: Bouma,
779
A.H., Normark, W.R., Barnes, N.E. (Eds.), Submariune Fans and Related Turbidite Systems. Frontiers in
780
Sedimentary Geology, pp. 23–28.
781
Sharman, G.R., Graham, S.A., Grove, M., Kimbrough, D.L., Wright, J.E., 2015. Detrital zircon provenance of the
782
late Cretaceous-Eocene California forearc: Influence of laramide low-angle subduction on sediment dispersal
783
and paleogeography. Geological Society of America Bulletin 127, 38–60.
30
ACCEPTED MANUSCRIPT 784
Sharman, G.R., Covault, J.A., Stockli, D.F., Wroblewski, A.F.J., Bush, M.A., 2017. Early Cenozoic drainage
785
reorganization of the United States Western Interior-Gulf of Mexico sediment routing system Geology 45,
786
187–190.
787
Snedden, J.W., Galloway, W.E., Milliken, K.T., Xu, J., Whiteaker, T., Blum, M.D., 2018. Validation of empirical source-to-sink scaling relationships in a continental-scale system: The Gulf of Mexico basin Cenozoic record.
789
Geosphere 14, 1–17.
792
IP
CR
791
Strahler, A.N., 1952. Hypsometric (area-altitude curve) analysis of erosional topography. Geological Society of America Bulletin 63, 1117-1141.
Sømme, T.O., Helland-Hansen, W., Martinsen, O.J., Thurmond, J.B., 2009. Relationships between morphological
US
790
T
788
and sedimentological parameters in source-to-sink systems: A basis for predicting semi-quantitative
794
characteristics in subsurface systems. Basin Research 21, 361–387.
795
AN
793
Sømme, T.O., Martinsen, O.J., Lunt, I., 2013. Linking offshore stratigraphy to onshore paleotopography: The Late Jurassic-Paleocene evolution of the south Norwegian margin. Geological Society of America Bulletin 125,
797
1164–1186.
800 801
ED
of Fluvial Sediment to the Coastal Ocean. Journal of Geology 115, 1–19.
PT
799
Syvitski, J.P.M., Milliman, J.D., 2007. Geology, Geography, and Humans Battle for Dominance over the Delivery
Voris, H.K., 2000. Maps of Pleistocene sea levels in Southeast Asia: Shorelines, river systems and time durations.
CE
798
M
796
Journal of Biogeography 27, 1153–1167. Wetzel, A., 1993. The transfer of river load to deep- sea fans – a quantitative approach. American Association of
803
Petroleum Geologists Bulletin 77, 1679-1692.
804
Whittaker, J.M., Goncharov, A., Williams, S.E., Müller, R.D., Leitchenkov, G., 2013. Global sediment thickness
805
data set updated for the Australian-Antarctic Southern Ocean. Geochemistry, Geophysics. Geosystems 14,
806
3297–3305.
807
AC
802
Woodcock, N.H., 2004. Life span and fate of basins. Geology 32, 685–688.
808 31
ACCEPTED MANUSCRIPT 809
Table 1. Summary of morphological parameters defined in the S2S database by catchment, continental shelf,
810
continental slope and submarine fan segments. The table shows the parameter name, the calculation, a description of
811
the measurement and units, and the original dataset(s) used to define the measurement. See main text for an
812
explanation of the methodology and supplementary information for the S2S database and metadata.
813 Figure 1. The morphological parameters of the source-to-sink system that were analyzed in the global study
815
(modified after Sømme et al., 2009). 1 DEM based on SRTM v4.1 (Jarvis, 2008) and GMTED2010 (Danielson and
816
Gesch, 2011), 2 catchment attributes based on Somme et al. (2009), Milliman and Farnsworth (2011) and Nyberg et
817
al. (2018), 3 terrestrial sinks based on Nyberg and Howell (2015), 4 bathymetry data based on SRTM30 Plus (Becker
818
et al., 2009) and 5 seafloor delineations based on Harris et al. (2014).
US
CR
IP
T
814
AN
819
Figure 2. An example of the semi-automated method to derive morphological measurements of the catchment,
821
continental shelf, continental slope and submarine fan for the contributing catchment area of the Bengal submarine
822
fan. River profile lengths are derived based on DEM and extend to its continental shelf by its centerline profile while
823
the continental slope profile is defined as its most proximal distance to the abyssal plain. The topographic and
824
bathymetric profiles that connect to a submarine fan define the contributing catchment area. Basemap shows a
825
hillshaded SRTM 30 Plus DEM and bathymetry dataset of Becker et al. (2009).
CE
PT
ED
M
820
826
Figure 3. A comparison of continental slope length and slope gradient measured in the current study and that of
828
Somme et al. (2009). The variability in the two datasets demonstrates the subjective nature of interpreting the
829
continental slope.
AC
827
830 831
Figure 4. The distribution of 52 complete source-to-sink systems showing their contributing catchment areas to
832
submarine fans. A comparison of their scale is shown surrounding the global map with a planar view of its river
833
channel, continental shelf, slope and submarine fan profile (see Figs. 2, 13 and supplementary kmz). Note the
834
change in scalebar. Basemap shows a hillshaded SRTM 30 Plus DEM and bathymetry dataset of Becker et al.
32
ACCEPTED MANUSCRIPT 835
(2009). Map is shown with a Robinson projection at a 1:125 000 000 scale. Source-to-sink system comparisons are
836
in a Cylindrical Equal Area projection.
837 Figure 5. Morphological relationships observed in the catchment segment (Qs > 1 MT/yr). The data show the
839
scaling patterns of river length, maximum relief, % of low relief area and river profile gradient by catchment area.
840
Note the lack of correlation between relief and low relief area by system size.
IP
T
838
CR
841
Figure 6. (A) Relief versus catchment area for a regional-based selection of 1 863 catchments (Qs > 0 MT/yr)
843
grouped into different tectonic settings. (B) Relief versus catchment area for a regional-based selection of 870
844
systems (Qs > 0 MT/yr) grouped by mature and young passive margins. See text for discussion.
AN
US
842
M
845
Figure 7. Morphological relationships observed on the continental shelf segment by a simplified tectonic
847
classification for the 52 source-to-sink systems (Fig. 4). The results show a weak power law correlation between
848
river length and shelf width and shelf width and slope length. The gradient on the shelf does appear to correlate to
849
river profile gradient while a correlation is observed between estuarine length and shelf width. ES – epicontinental
850
seaway.
CE
PT
ED
846
851
Figure 8. Box plots showing the distribution of continental shelf width and sediment thickness of the continental
853
shelf/slope for active versus passive continental shelf/slope margins based on 1 778 source-to-sink systems. Box
854
shows 25th and 75th percentile, whiskers show 10th and 90th percentile, square shows the mean and line shows the
855
median of the data.
AC
852
856 857
Figure 9. Morphological relationships observed on the continental slope segment. The results show a weak power
858
law correlation between slope length and catchment area, shelf width, fan area, fan length, length of river channel
33
ACCEPTED MANUSCRIPT 859
and river gradient. The subjective nature of defining the continental slope boundary suggests it is not a good
860
parameter to scale source-to-sink segments. ES – epicontinental seaway.
861 Figure 10. Box plots showing (A) slope length by active versus passive continental shelf/slope margins, and (B)
863
depth at the continental slope base for 1 778 source-to-sink catchments by tectonic setting based on ETOPO1
864
bathymetry (Amante and Eakins, 2009). The results show that slope length does not vary significantly by active
865
versus passive continental shelf/slope margins (or by further tectonic regime subdivision). Depth at the continental
866
slope base is similar across tectonic regimes with the exception of forearc settings. Box shows 25th and 75th
867
percentile, whiskers show 10th and 90th percentile, square shows the mean and line shows the median of the data. ES
868
– Continental slope adjacent to the epicontinental seaway.
AN
869
US
CR
IP
T
862
Figure 11. The number of submarine canyon heads (A) and the longest submarine canyon length (B) for source-to-
871
sink systems that contribute to one of the 52 submarine fan in Figure 4. Box shows 25 th and 75th percentile, whiskers
872
show 10th and 90th percentile, square shows the mean and line shows the median of the data. Sh/SL – Continental
873
shelf / Continental slope, ES - Continental slope adjacent to the epicontinental seaway.
PT
ED
M
870
875
CE
874
Figure 12. Morphological relationships observed on the submarine fan segment. The results show a power law
877
correlation between increasing submarine fan length and fan width. In addition, proximal and distal relationships are
878
observed for fan length and contributing river length, fan area and contributing catchment area, and contributing
879
river sediment load (Qs) and fan area. Depths of the submarine fan are similar across scales while submarine fan
880
slope decreases with a power law relationship as contributing catchment area increases.
881
Figure 13. Relative topographic and bathymetric length and rise profiles of mature passive margins (see Fig. 4 for
882
location) in descending source-to-sink catchment size. The continental shelf/slope segments suggest a relative length
883
increase with decreasing system size. The data demonstrate that while catchment size changes, continental slope
AC
876
34
ACCEPTED MANUSCRIPT length does not significantly alter thereby increasing the relative length of the continental slope with decreasing
885
source-to-sink size. Note, the submarine fan profile does not represent a drainage profile but a perpendicular transect
886
from the base of the continental slope (see Fig. 4 for location; see also supplementary kmz).
887
Figure 13 (cont.). Relative topographic and bathymetric length and rise profiles of young passive margins, foreland,
888
epicontinental seaways and strike-slip source-to-sink profiles (see Fig. 4 for location) in descending source-to-sink
889
catchment size. Note, the submarine fan profile does not represent a drainage profile but a perpendicular transect
890
from the base of the continental slope (see Fig. 4 for location; see also supplementary kmz).
891
Figure 14. Scatter plots showing for 52 S2S systems (see Fig. 4 for location), the proportional length of each
892
segment comparative to its entire source-to-sink profile length (Fig. 13) by catchment size. The results suggest that
893
river length is proportionally a larger component of the source-to-sink profile with increasing catchment size while
894
the length of its continental slope and shelf is an increasingly smaller component. By contrast, smaller systems
895
suggest that the continental shelf and slope length are proportionally of an increasingly larger component of the
896
source-to-sink system. Proportionally, the submarine fan length shows the most amount of variability with a lack of
897
correlation to the entire source-to-sink profile length. See text for additional discussion.
ED
M
AN
US
CR
IP
T
884
898
Figure 15. Illustration showing the change in catchment morphology of relief and catchment area / river length at T1
900
to T2 based on sea level fluctuations (blue) and regional tectonic uplift (red). For systems with wider terrestrial
901
sinks/continental shelves (A) the large low gradient plain/shelf may proportionally fluctuate the morphology of the
902
catchment significantly while tectonic uplift may have a minimal effect. By contrast, smaller and steeper source-to-
903
sink systems (B) would experience minimal morphological change in terms of a similar relative sea level fall from
904
T1 to T2 to (A), whereas regional tectonic uplift from T1 to T2 may significantly divert drainage area and lower
905
relief from the original catchment region. Insert graphs show the relative change in catchment morphology by a
906
relative sea level or regional tectonic uplift influence for A and B Note, the relative change is a simplistic illustration
907
of a relationship that is not linear.
908
Figure 16 – Summary of morphological parameters of modern source-to-sink systems with a total suspended
909
sediment load (Qs) greater than 1 MT/yr. Numbers show mean and standard deviation of: river length (RL), river
AC
CE
PT
899
35
ACCEPTED MANUSCRIPT gradient (RG), continental shelf width (SW), continental slope length (SL), submarine fan length (SFL), continental
911
shelf depth (ShD), continental shelf gradient (ShG), continental slope depth (SlD), continental slope gradient (SlG),
912
submarine fan depth (SFD), submarine fan gradient (SFG), % of catchment area as terrestrial sink (TS Area), and %
913
of catchment area as low relief area < 10 m/km. n = number of source-to-sink systems considered (catchment, shelf
914
and slope / submarine fan). Steep, short and deep systems (A) are separated based on an extensional/strike-slip or a
915
forearc origin. Wide and deep systems (B) are characterized by sedimentary systems of a passive or foreland
916
tectonic origin on the margin of a passive continental shelf/slope. See text for discussion. Note that the illustration is
917
not drawn to scale.
CR
IP
T
910
AC
CE
PT
ED
M
AN
US
918
36
ACCEPTED MANUSCRIPT Catchment Segment Parameter
Calculation
Description
Dataset 2
A
Catchment area (km )
Nyberg et al. (accepted)
Maximum Relief (R)
R
Maximum catchment relief (km)
Nyberg et al. (accepted)
% < 10 m/km (10m_km)
10m_km
% of catchment area with < 10 m/km in relief
Reuter et al. (2007); Danielson and Gesch (2011)
% Terrestrial Sink (Sed_Basin)
Sed_Basin
% of catchment area defined as a terrestrial sink based on Nyberg and Howell (2015)
Nyberg et al. (accepted)
River Length (River_L)
River_L
Longest river channel (m) derived from endorheic/ exorheic sink to furthest reach along the river network
Nyberg et al. (accepted)
River Profile Gradient (River_S)
R/River_L
Average river profile gradient (m/km)
Water Discharge (Q)
0.00237A0.8
Water discharge (km3/yr) based on Q equation of Syvitski and Milliman (2007)
Nyberg et al. (accepted)
Total Suspended Sediment Load (Qs)
BQART
Total suspended sediment load (MT/yr) based on BQART equation of Syvitski and Milliman (2007)
Nyberg et al. (accepted)
CR
IP
T
Area (Area)
US
Continental Shelf Segment BP_L – S_Dist
Shelf Width (S_Dist)
S_Dist
Shelf Depth (Shelf_LowH)
Shelf_LowH
Shelf Profile Gradient (C_Shelf_S)
Shelf_LowH/S_Dist
Sediment Thickness (Sed_Thick)
Sed_Thick
Base of Slope Depth (Slope_LowH)
Average sediment thickness on the continental shelf/slope (m) measured along the bathymetric profile
Whittaker et al. (2013)
Length of the continental slope (m) measured along the bathymetric profile
Slope_LowH
Depth at the base of the continental slope/basin floor transition (m)
Becker et al. (2009)
Slope_TopH
Depth at the top of the continental slope (m) measured along the bathymetric profile
Becker et al. (2009)
CE
Top of Slope Depth (Slope_TopH)
Becker et al. (2009)
Gradient of the continental shelf (m/km) measured along the bathymetric profile
M
Slope_L
PT
Slope Length (Slope_L)
Depth at the continental shelf edge (m) measured along the bathymetric profile
ED
Continental Slope Segment
AN
Waterbody/estuarine length (m) measured as the difference of the bathymetric river to shelf edge profile length (BP_L) and shelf width (S_Dist) Width of the continental shelf (m) measured as distance from continental shelf edge
Waterbody Length (E_Length)
(Slope_TopH – Slope_LowH) / Slope_L
Average gradient of the continental slope profile (m/km) measured along the bathymetric profile
Canyon Length (C_Length)
C_Length
Length of the submarine canyon (km)
Harris and Whiteway (2011)
Canyon Head Frequency (C_Freq)
C_Freq
Number of canyon heads that incise the continental shelf/slope
Harris and Whiteway (2011)
Canyon Distance (C_Distance)
C_Distance
Distance from river mouth to the submarine canyon (m)
Fan Area (F_Area)
F_Area
Area of the submarine fan (km2)
Fan Length (F_Length)
F_Length
Length of the submarine fan (m) from base of continental slope to distal submarine fan
Fan Width (F_Width)
F_Width
Width of the submarine fan (m) measured parallel to the base of the continental slope
Fan Depth (F_Depth)
F_Depth
Depth of the distal submarine fan (m)
AC
Slope Profile Gradient (C_Slope_S)
Submarine Fan Segment
Becker et al. (2009)
37
ACCEPTED MANUSCRIPT (Slope_LowH – F_Depth) / F_Length
Fan Profile Gradient (F_S)
Average gradient of the submarine fan profile (m/km)
AC
CE
PT
ED
M
AN
US
CR
IP
T
919 920 921
38
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13r1
Figure 13r2
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18