ArticlePDF Available

Geomorphic and Hydraulic Unit Richness and Complexity in a Coastal Plain River

Authors:

Abstract and Figures

Geomorphic and hydraulic units in river channels are closely linked to geodiversity and habitats, and thus to biodi- versity. In a ~ 200 km reach of the lower Sabine River, in the northern Gulf of Mexico Coastal Plain, 72 different hydraulic units (HU) were identified in six geomorphic zones or river styles. Richness–area relationships indicate a linear or logarithmic increase of HUs, as opposed to the less steep power functions generally found in biogeographic species–area curves or in soil richness– area analyses. Different results are obtained when starting from the upstream or downstream end of the study area, indicating the importance of directionality in such analyses. These results show that HUs (and related habitats and biotopes) are both richer and more variable than a repeated sequence of units. The number of HUs inundated increases linearly with flow stage categories, indi- cating the importance of high within-bank flows in maintaining and activating HUs. Aggregated HUs (AHUs) associated with similar geomorphic units are highly connected, both with respect to patterns of spatial adjacency and potential connectivity at similar flow levels. Spectral graph theory metrics applied to a graph representation of spatial adjacency shows a highly complex network with a high potential for rapid propagation of changes—and even more so for a graph based on flow connectivity. The flow connectivity graph shows far higher synchronization as indicated by algebraic connectivity. Thus suggests more rapid and coherent changes for processes driven by river flow, as opposed to phenomena driven by other factors between flow events. These findings have im- portant implications for understanding relationships between geodiversity and habitat diversity, managing habitat and biodiversity, and linking the latter to instream flows.
Content may be subject to copyright.
USING e-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION
Required software to e-Annotate PDFs: Adobe Acrobat Professional or Adobe Reader (version 11
or above). (Note that this document uses screenshots from Adobe Reader DC.)
The latest version of Acrobat Reader can be downloaded for free at: http://get.adobe.com/reader/
Once you have Acrobat Reader open on your computer, click on the Comment tab
(right-hand panel or under the Tools menu).
This will open up a ribbon panel at the top of the document. Using a tool will place
a comment in the right-hand panel. The tools you will use for annotating your proof
are shown below:
1. Replace (Ins) Tool for replacing text.
Strikes a line through text and opens up a text
box where replacement text can be entered.
How to use it:
Highlight a word or sentence.
Click on .
Type the replacement text into the blue box that
appears.
2. Strikethrough (Del) Tool for deleting text.
Strikes a red line through text that is to be
deleted.
How to use it:
Highlight a word or sentence.
Click on ..
3.
Commenting Tool for highlighting a section
to be changed to bold or italic or for general
comments.
How to use it:
Click on .
Type any instructions regarding the text to be
altered into the box that appears.
4.
Insert Tool for inserting missing text
at specific points in the text.
Use these 2 tools to highlight the text
where a comment is then made.
How to use it:
Click on .
Click at the point in the proof where the comment
should be inserted.
Type the comment into the box that
appears.
Marks an insertion point in the text and
opens up a text box where comments
can be entered.
Click and drag over the text you need to
highlight for the comment you will add.
The text will be struck out in red.
Click on .
Click close to the text you just highlighted.
USING e-ANNOTATION TOOLS FOR ELECTRONIC PROOF CORRECTION
For further information on how to annotate proofs, click on the Help menu to reveal a list of further options:
5. Attach File Tool for inserting large amounts of
text or replacement figures.
Inserts an icon linking to the attached file in the
appropriate place in the text.
How to use it:
Click on .
Click on the proof to where you’d like the attached
file to be linked.
Select the file to be attached from your computer
or network.
Select the colour and type of icon that will appear
in the proof. Click OK.
The attachment appears in the right-hand panel.
6. Add stamp Tool for approving a proof if no
corrections are required.
Inserts a selected stamp onto an appropriate
place in the proof.
How to use it:
Click on .
Select the stamp you want to use. (The Approved
stamp is usually available directly in the menu that
appears. Others are shown under Dynamic, Sign
Here, Standard Business).
Fill in any details and then click on the proof
where you’d like the stamp to appear. (Where a
proof is to be approved as it is, this would
normally be on the first page).
7.
Drawing Markups Tools for drawing shapes, lines, and freeform
annotations on proofs and commenting on these marks.
Allows shapes, lines, and freeform annotations to be drawn on proofs and
for comments to be made on these marks.
How to use it:
Click on one of the shapes in the Drawing
Markups section.
Click on the proof at the relevant point and
draw the selected shape with the cursor.
To add a comment to the drawn shape,
right-click on shape and select Open
Pop-up Note.
Type any text in the red box that
appears.
Drawing
tools
available on
comment
ribbon
Research Article
Geomorphic and hydraulic unit richness and complexity in a coastal plain river
Jonathan D. Phillips
72 different hydraulic units (HU) were identified in six geomorphic zones or river styles. Richness-area relationships in-
dicate a linear or logarithmic increase of HUs, as opposed to the less steep power functions generally found in biogeo-
graphic species-area curves or in soil richness-area analyses. Spectral graph theory metrics applied to a graph
representation of spatial adjacency shows a highly complex network with a high potential for rapid propagation of
changesand even more so for a graph based on flow connectivity.
Wiley Online Library Graphical TOC
Author Query Form
Journal: Earth Surface Processes and Landforms
Article: esp_4211
Dear Author,
During the copyediting of your paper, the following queries arose. Please respond to these by annotating your proofs with the
necessary changes/additions.
If you intend to annotate your proof electronically, please refer to the E-annotation guidelines.
If you intend to annotate your proof by means of hard-copy mark-up, please use the standard proofing marks. If manually
writing corrections on your proof and returning it by fax, do not write too close to the edge of the paper. Please remember
that illegible mark-ups may delay publication.
Whether you opt for hard-copy or electronic annotation of your proofs, we recommend that you provide additional clarification
of answers to queries by entering your answers on the query sheet, in addition to the text mark-up.
Query No. Query Remark
Q1 AUTHOR: Please confirm that given names (red) and surnames/family names (green)
have been identified correctly.
Q2 AUTHOR: Please verify that the linked ORCID identifiers are correct for each author.
Q3 AUTHOR: Frissell et al., 1986 correct as amended from Frissell, 1986?
Q4 AUTHOR: Ref. Coffman et al., 2011is cited in text but not provided in the reference
list. Please provide details in the list or delete the citation from the text.
Q5 AUTHOR: The citation Anderson et al., 2008has been changed to Anderson and
Rodriguez, 2008to match the author name/date in the reference list. Please check if
the change is fine in this occurrence and modify the subsequent occurrences, if
necessary.
Q6 AUTHOR: The citation TCB, Inc., 2006has been changed to TCB, Inc, 2006to
match the author name/date in the reference list. Please check if the change is fine in
this occurrence and modify the subsequent occurrences, if necessary.
Q7 AUTHOR: Ref. Phillips (2012)is cited in text but not provided in the reference list.
Please provide details in the list or delete the citation from the text.
Q8 AUTHOR: Sand sheets are more or less uniform tabular sand sheets?
Q9 AUTHOR: Phillips et al., 2005 not on ref. list.
Q10 AUTHOR: The citation Phillips, 2008, 2013, 2015has been changed to Phillips,
2008a, b, 2013, 2015to match the author name/date in the reference list. Please
check if the change is fine in this occurrence and modify the subsequent occurrences,
if necessary.
Q11 AUTHOR: Is -electron energycorrect? pdf file has box instead of p.
Q12 AUTHOR: "Coffman et al, 2010" has not been cited in the text. Please indicate where it
should be cited; or delete from the Reference List.
Q13 AUTHOR: "Gumbricht et al, 2004" has not been cited in the text. Please indicate where
it should be cited; or delete from the Reference List.
Q14 AUTHOR: Phillips (2008b) unable to open web page. Please check URL and give
access date.
Q15 AUTHOR: Phillips (2011) unable to open web page. Please check URL and give
access date.
Q16 AUTHOR: Revised figures 3 to 5 still contains poor quality of text.
Q17 AUTHOR: Ibàñez et al., 2013 correct as amended from Ibanez and Bockheim, 2013?
Journal Code Article ID Dispatch: 17.08.17 CE:
E S P 4 2 1 1 No. of Pages: 17 ME:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
OK
Yes
OK
OK
ref added
corrected
ref added
OK
deleted
corrected
corrected
sorry
yes
Geomorphic and hydraulic unit richness and
complexity in a coastal plain river
Q1 Q2 Jonathan D. Phillips*
Tobacco Road Research Team, Department of Geography, University of Kentucky, Lexington, KY 40506-0027, USA
Received 12 April 2017; Revised 20 July 2017; Accepted 25 July 2017
*Correspondence to: Jonathan D. Phillips, Tobacco Road Research Team, Department of Geography, University of Kentucky, Lexington, KY 40506-0027, USA. E-mail:
jdp@uky.edu
ABSTRACT: Geomorphic and hydraulic units in river channels are closely linked to geodiversity and habitats, and thus to biodi-
versity. In a~ 200 km reach of the lower Sabine River, in the northern Gulf of Mexico Coastal Plain, 72 different hydraulic units
(HU) were identified in six geomorphic zones or river styles. Richnessarea relationships indicate a linear or logarithmic increase
of HUs, as opposed to the less steep power functions generally found in biogeographic speciesarea curves or in soil richness
area analyses. Different results are obtained when starting from the upstream or downstream end of the study area, indicating the
importance of directionality in such analyses. These results show that HUs (and related habitats and biotopes) are both richer and
more variable than a repeated sequence of units. The number of HUs inundated increases linearly with flow stage categories, indi-
cating the importance of high within-bank flows in maintaining and activating HUs. Aggregated HUs (AHUs) associated with similar
geomorphic units are highly connected, both with respect to patterns of spatial adjacency and potential connectivity at similar flow
levels. Spectral graph theory metrics applied to a graph representation of spatial adjacency shows a highly complex network with a
high potential for rapid propagation of changesand even more so for a graph based on flow connectivity. The flow connectivity
graph shows far higher synchronization as indicated by algebraic connectivity. Thus suggests more rapid and coherent changes
for processes driven by river flow, as opposed to phenomena driven by other factors between flow events. These findings have im-
portant implications for understanding relationships between geodiversity and habitat diversity, managing habitat and biodiversity,
and linking the latter to instream flows. Copyright © 2017 John Wiley & Sons, Ltd.
KEYWORDS: hydraulic units; geomorphic units; richnessarea analysis; connectivity; complexity
Introduction
River and stream channels may contain numerous geomorphic
units (GUs), each of which may encompass one or more
hydraulic units (HUs), associated with aquatic habitats and
biotopes. Many channels may have multiple units within a sin-
gle cross-section or short reach. Further, the activation (via in-
undation) or even presence/absence of the GUs and HUs
often varies with flow levels. This paper explores the number,
type and connectivity of GUs and HUs in three contexts: longi-
tudinal variation along the channel in association with varia-
tions in geomorphic settings, complexity in terms of spatial
adjacency, and complexity with respect to connectivity of units
with varying flows. The study area is the lower Sabine River,
TexasLouisiana, a sand-bed alluvial river that drains to the
northern Gulf of Mexico.
This work can be situated in three general sets of issues in
geomorphology and river science. First, geomorphology is rec-
ognized by geomorphologists, ecologists and other river sci-
entists as a key first-order control of aquatic and riparian
habitats. Thus geomorphically driven geodiversity is strongly
related to habitat diversity in fluvial environments, and thus
to biodiversity. Understanding these geodiversityhabitat di-
versity patterns is thus critical to fluvial biogeomorphology,
ecohydrology and assessments of riverine biodiversity.
Second, connectivity of various types is an important and
ever-growing concern in river science and management (and
in environmental science in general). This paper investigates
the structural complexity of GUs and HUs represented as a
graph/network. This approach complements spatially explicit
studies of hydrologic, sediment, biogeochemical and ecologi-
cal connectivity based on flux and transport with a more
general and robust notion that connects connectivity with
geodiversity and biodiversity. Third, because this study explic-
itly links connectivity to flow levels, and is concerned with re-
lationships between size (area, or length of river reaches) and
geodiversity (as a surrogate for habitat and biodiversity), it is
directly relevant to river management (such as instream flow
programs). These frameworks are addressed more fully below.
GUs are defined as in the hierarchical framework for river
characterization developed by Brierley and Fryirs (2005). The
fundamental reach-scale units, river styles or geomorphic
zones, are defined on the basis of similarities of channel and
valley morphology, channelfloodplain hydrologic exchanges,
dominant hydrologic controls and regimes, and geologic and
other constraints. The geomorphic zonation of the lower
Sabine is described by Phillips (2008a). GUs are specific land-
forms within reaches, e.g. point bars, natural levees, riffle-pool
EARTH SURFACE PROCESSES AND LANDFORMS
Earth Surf. Process. Landforms (2017)
Copyright © 2017 John Wiley & Sons, Ltd.
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/esp.4211
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
sequences. GUs are erosional, depositional or transportational
landforms, referred to by Brierley and Fryirs (2005, p. 26) as
the building blocks of river systems. Each GU represents a dis-
tinct formprocess association. GUs are generally capable of
significant change on the scale of ~1 year, but may range from
ephemeral to persistent due to the episodic, threshold-
dependent nature of geomorphic change.
HUs are the most detailed level in the river styles scheme,
comprising specific hydrologic and ecological elements such
as large woody debris, bedforms, aquatic vegetation and indi-
vidual flow obstructions or roughness elements. These units
are at least potentially capable of significant change over time-
scales of hours to months, but, again, may range from ephem-
eral to persistent. HUs generally comprise the basic habitat
elements for aquatic organisms.
HUs represent relatively uniform patches of flow and sub-
strate characteristics, and are related to biotopes and
mesohabitats. Technical guidance for the Texas Instream Flow
Program, for example, specifies that instream habitats will be
delineated based on mesohabitats(TIFP 2008, p. 69), one of
several similar classifications often used in aquatic biology,
consisting of pools, backwaters, runs or glides, riffles, rapids
and chutes. TIFP (2008) allows for further qualification of
mesohabitats where possible.
The mesohabitats referred to in TIFP (2008) are also called
biotopes. Milan et al. (2010) reviewed five different biotope
classifications from the river and aquatic sciences literature,
none of which are identical to the TIFP classification but all
of which are quite similar, both to each other and to the TIFP
mesohabitats. Beyond problems with imprecision and user
variation in subjective classifications, various nomenclatures
and imprecise terminology (e.g., what constitutes a deep
run?), Milan et al. (2010) identified several other issues with
biotope classification. First, the links between the hydraulic
biotopes and habitat are unclear in many cases (cf. Clifford
et al. 2006; Shoffner and Royall 2008). Second, biotopes are
also stage dependent, though the same can be said of some
HUs. Little research has been done on the effects of flow levels
or stage on biotope assessments, but stage-dependent changes
clearly occur. For instance, areas that are riffles or rapids at
moderate flows may become runs or glides at higher flows.
Third, in larger streams mesohabitats/biotopes are too large to
represent uniform flow and substrate characteristics. The use
of HUs rather than biotopes directly addresses the third issue,
and should also result in tighter relationships between hydro-
geomorphic and ecological characteristics. However (as is
often the case with biotopes), these relationships are often
assumed. This assumption is reasonable, at least to a first
approximation, but is likely to be imperfect and variable. The
stage-dependence issue is directly addressed in this study.
Background
Diversity of GUs and HUs
A key question is the rate and pattern of increase in GUs and
HUs along a stream. If these are mainly common to many or
all geomorphic settings along a fluvial corridor (e.g. sandy point
bars or gravel riffles), then we would expect a pattern of rela-
tively rapid increase as area or corridor length increases ini-
tially, followed by a flattening as the units identified have
already been encountered. This is often the case for species
area curves in biogeography and soil richnessarea relation-
ships in pedology. Biological species richness typically
increases with area as power function (richness = f[A
b
], b<1),
and studies of pedodiversity at various scales have typically
found similar trends (Ibáñez et al. 1995, 1998; Lomolino
2000; Phillips 2001a; Drakare et al. 2006; Phillips and Marion
2007; Ibàñez et a Q17l., 2013). However, if new, not previously en-
countered, units (as well as repeated ones) occur along the cor-
ridor, the richnessarea relationship would not level off.
Several quite different concepts of river ecology and geomor-
phology suggest this could be the case.
Traditional conceptual models of river ecology are based on
a changing continuum of features from upstream to down-
stream (e.g. Vannote et al. 1980 Q3; Frissell et al. 1986; Tomanova
et al. 2007), which could imply a steady increase in GU and
HU diversity as successively longer reaches are considered. Al-
ternatives to this framework based on more complex spatial
variation and hierarchical patch dynamics suggest the possibil-
ity of a more complex pattern of change (e.g. Brierley and Fryirs
2005; Thorp et al. 2008), which would also imply a steady in-
crease. However, the spatial and statistical pattern of the in-
crease is unknown.
Wiens (2002) advocated a landscape ecological approach to
fluvial systems, based on a concern with the quality, bound-
aries, context and connectivity of patches. The patches of land-
scape ecology can be defined and analyzed at various scales,
but HUs are a type of patch in this sense. Though the approach
in this study is not directly concerned with organisms, the ap-
proach is otherwise consistent with Wiens(2002) riverscape
perspective (see Carbonneau et al. 2012, and Wheaton et al.
2015, for explicit examples of a geomorphological approach
to riverscapes).
Scale, particularly with respect to spatial resolution, is always
an issue in geomorphology and landscape ecology GUs and
HUs, patches and other entities can be conceived or perceived
and delineated at multiple scales. The dominant processes,
controls and relationships often vary with spatial scale. With re-
spect to hierarchical relationships, GUs and HUs are defined as
described earlier, consistent with Brierley and Fryirs (2005).
Many schemes for mapping and classifying GUs and similar
features exist; two recent and rigorous examples are Wheaton
et al. (2015) and Belletti et al. (2017). Neither had been
published at the time of data collection for this study, but the
approach here is at least conceptually similar to Belletti et al.
(2017), and the level of detail is comparable to tier 3 in
Wheaton et al.s (2015) four-tier classification. The methods
used in this study are consistent with the characterization of
GUs used by state and federal agencies in Texas Gulf Coastal
Plain rivers for in-stream flow studies and management (Phillips
2008b; Coffman Q4et al., 2011).
Geodiversity and biodiversity
The role of fluvial geomorphology in channel habitats in the
context of river management was discussed by Newson and
Newson (2000), who emphasized the critical control exercised
by geomorphology on abiotic habitats. Newson and Newson
(2000) described the challenges of integrating geomorphology
and ecology, and of incorporating a distinctive spatial formula-
tion and biological validation into studies of interactions of
fluvial geomorphology with biological habitats.
The assumption that the type and quality of aquatic habitats
are directly related to specific landforms and geomorphic pro-
cesses is rarely challenged, and is reflected across the river sci-
ences (e.g. Johnston et al. 2001; Sullivan et al. 2004; Thomson
et al. 2004; Moret et al. 2006; Baker et al. 2016). While the
geomorphic diversityhabitat diversitybiodiversity link is often
implicit or assumed, several studies have explicitly supported
these links (e.g. Sullivan et al. 2004; Smith and Mather 2013;
Baker et al. 2016; Rice 2017). Geomorphology is often used
1GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
as a surrogate for habitat in floodplain and other wetland envi-
ronments, too (e.g. Johnson 2005; Hamilton et al. 2007).
Connectivity
There exists a large and rapidly expanding literature on con-
nectivity in geomorphology, and particularly in fluvial systems
(see reviews by Bracken et al. 2015; Wohl 2017; Wohl et al.
2017). An even more extensive body of work on connectivity
exists in landscape ecology; Eros and Campbell Grant (2015)
presented a synthesis in a river ecology context. The linkages
among hydrologic, geomorphic and biogeochemical connec-
tivity are highlighted by Covino (2017), who emphasized the
range of scales at which connectivity occurs (echoing Bracken
et al. 2015). Wohl et al. (2017, p. 2) identified five research
gaps with respect to connectivity in geomorphic systems. This
paper contributes directly to two of those, quantifying different
facets of connectivity, and quantifying how those different
facets interact. It also indirectly addresses two other gaps
highlighted by Wohl et al. (2017), related to feedbacks and en-
hance or limit connectivity, and quantifying thresholds that
substantially alter system function.
Rather than the spatially explicit connectivity typically
assessed in geomorphology, hydrology and landscape ecology,
this study addresses structural connectivity. This approach is
based on a network or graph representation of system compo-
nents (in this case HUs), with links based on patterns of spatial
adjacency, functional relationships or massenergy fluxes. The
structural connectivity approach has been applied to ecologi-
cal food webs; state-and-transition models of geomorphic, soil,
and ecological systems; and soil landscapes (see review by
Phillips and Van Dyke 2017).
Analogous to the connectivity in biological communities,
where changes in one component may reverberate through a
food web or ecosystem, fluvial HUs and GUs are often strongly
interconnected. Thus other key questions relate to the com-
plexity of these networks of interaction. Specifically, do pertur-
bations tend to be amplified by the network, or absorbed and
damped? To what extent are such changes synchronous?
Complexity is addressed here using spectral graph theory. Iden-
tified aggregated HUs are considered the nodes of a graph, with
links between them defined on the basis of spatial adjacency,
and alternatively on whether they are activated at the same
flow levels.
In a fluvial system, spatially adjacent units are capable of ex-
changing mass, energy and biota. However, water flows may
link even non-contiguous units. Water, sediment and ecologi-
cal connectivity are closely related to hydrologic connectivity,
which varies with flow levels. At very low flows, for example,
only landforms and habitats associated with the stream thalweg
are connected, while at banktop or overbank flows many other
units are linked. Just as successively higher (relative to the
thalweg) units may be activated as stages increase, others
may be deactivated for example, an algal mat on a bar sur-
face is a distinct HU at low flows, but is no longer significant
under several meters of water.
Study Area
The study area is the Sabine River along the TexasLouisiana
border from Toledo Bend Dam to the Sabine Lake estuary near
Beaumont, Texas (FigureF1 1). The Sabine River has a total drain-
age area of 25 267 km
2
, of which 6676 km
2
(26%) is down-
stream of the Toledo Bend dam. The climate is humid
subtropical, and the river valley in the study area is
predominantly forested, except in the lowermost, tidally
influenced areas where marshes are dominant. Like any river,
the lower Sabine has its idiosyncrasies, but the general environ-
mental framework and types of landforms present are typical of
Gulf of Mexico Coastal Plain rivers between the Mississippi/
Atchafalaya and Rio Grande Q5(Anderson and Rodriguez 2008;
Coffman et al., 2011). Though the purpose and operation of
the impoundments differ, all those in Texas are affected by up-
stream dams, five of those including dams within the coastal
plain reaches of the river.
The lower Sabine is an active alluvial river. It has a generally
meandering planform, with active lateral migration and fre-
quent meander cutoffs. Some reaches include active and
semi-active anabranches. The geomorphic zonation, as well
as the Quaternary geomorphic history, is described by Phillips
(2008a, 2013).
Discharge in the lower Sabine River, as in all rivers, is influ-
enced by the climate and hydrologic response of the drainage
basin. However, also important are releases from Toledo Bend
Reservoir, water withdrawals, and tidal and coastal backwater
effects (e.g. temporary ponding or upstream flow). Toledo Bend
Reservoir has a controlled storage capacity of 5.522 km
3
. The
primary purposes are water supply, hydroelectric power gener-
ation, and recreation; there is no flood control function. The
design flow of the Toledo Bend spillway is 8212 m
3
s
1
.A
minimum constant flow of about 5.7 m
3
s
1
(200 ft
3
s
1
)is
maintained via the spillway, but most of the flow is passed
through the hydroelectric turbines. Maximum recorded release
was 3239.5 m
3
s
1
, and a typical flow during turbine operation
is 200300 m
3
s
1
.
Mean and median flows and the 1% and 10% probability
flows increase as expected downstream within the study reach.
The flood stage discharges, however, and thus the recurrence
interval of overbank flow, decline downsteam (Phillips 2008a,
2008b). Therefore, overbank flow occurs more often with dis-
tance downstream from the dam, and channelfloodplain con-
nectivity is greater. Cross-sectional stream power (the product
of discharge, slope and specific weight of water) for a given dis-
charge at flood stage also generally decreases downstream, and
this plus the floodplain inundation reduces sediment transport
capacity and increases alluvial deposition. These trends are
not unusual for the lower reaches of low-gradient coastal plain
rivers (Phillips and Slattery 2006, 2007; Phillips 2013).
Previous studies show that releases from Toledo Bend Dam
have not significantly changed the discharge regime at
Deweyville or inputs into Sabine Lake (Solis et al. 1994;
Phillips 2003 Q6; TCB, Inc 2006), and that peak flows and mean
flows have been minimally influenced. However, dam releases
do clearly influence flows on hourly and daily timescales, and
the seasonality of flow. Dam release effects on hydrology di-
minish downstream from Toledo Bend, and are more important
at lower discharges.
Diurnal tidal ranges in the northern Gulf of Mexico are small
generally less than 0.6 m and are further filtered by the
Sabine Lake estuary. Nevertheless, the Sabine River channel
is cut to below sea level upstream of Deweyville (where the
gage datum is 1.8 m below sea level), to at least Big Cypress
and perhaps Nicholls Creek. The tidal signal in the discharge
record at Deweyville is barely discernible as a subtle sawtooth
pattern superimposed on the discharge and stage record.
Geomorphic zones
The six major geomorphic zones (river styles) identified in pre-
vious work (Phillips 2008a) are shown in Tables T1I and T2II, and
their locations in Figure 1. These represent reaches with distinct
2 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Figure 1. Lower Sabine River with key features and place names shown. Base map is a density map derived from digital elevation data. Boxes in-
dicate geomorphic zones (river styles) described in Tables I and II.
Table I. major reaches (river styles) of the lower Sabine River. Locations are in river distance upstream of Sabine Lake in kilometers
Reach Location Distinguishing Characteristics Primary Geomorphic Controls
1 Toledo Bend to Burr Ferry 213192 Incision, steep slope, bedrock control, valley
constriction, low sediment loads, pulsed flows
Geologic framework; Toledo Bend
Dam releases
2 Burr Ferry to Bon Wier 192131 Active lateral migration, ubiquitous large
point bars, wider valley, larger sediment load
Valley width; avulsion
3 Bon Wier to Big
Cow Creek
131103 Active lateral migration, ubiquitous large
point bars, wider valley, larger sediment load;
high floodplain/channel connectivity; low slope
Valley width; avulsion; neotectonics
4 Big Cow Creek to
Shoats Creek lower
10379 Active lateral migration, fewer point bars, high
floodplain/channel connectivity, low slope
Neotectonics; valley width; coastal
plain paleogeography
5 Shoats Creek to
Cutoff Bayou
7947 Few and finer-grained point bars, high
floodplain/channel connectivity with multiple
high flow distributary channels, high sinuosity,
embayed tributary mouths
Holocene sea-level rise; geology and
coastal plain paleogeography;
Pleistocene stream capture
6 Cutoff Bayou to
Sabine Lake
470 Rare point bars; distributary flow network; very
high sinuosity; deltaic; tidal influence
Holocene sea-level rise; tidal and coastal
influences; Pleistocene stream capture
3GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
hydrologic characteristics in terms of the relative importance of
dam releases and coastal backwater effects, single versus multi-
channel flow patterns, frequency of overbank flow and
channel-floodplain connectivity.
Zones 1 and 2 are single-channel meandering planforms,
while zones 5 and 6 have multiple-channel, anastomosing pat-
terns. In zones 3 and 4 there exists a single-channel planform at
low to median flows, but high-flow channels are activated at
sub-banktop flows via backwater flooding of tributaries and
flow through low points in the natural levee.
Methods
Data collection
GUs and HUs were inventoried based on a combination of field
observations, a database of continuous ground or river-level
photography over much of the study area, low-altitude oblique
aerial photographs and high-resolution aerial photography.
The entire Sabine River and Old River channels from Burr
Ferry to the Interstate 10 bridge was examined by boat at
various times between 2005 and 2008 some reaches on
multiple occasions. Continuous digital photography (i.e. pho-
tographs covering the entire river channel) was analyzed to
identify HUs. In addition, much of the reach from Toledo
Bend Dam to Burr Ferry was also examined by small boat,
canoe or via land access in 20002001 and in 2006, with
photographic records from those trips also available. Field
notes from these previous observations were also utilized,
which included detailed field mapping of specific
cross-sections, and general assessments of bed substrate and
bank material, bank stability and vegetation, and the geometry
and bedforms at tributary junctions. Measurements of bank
height and channel width at selected cross-sections were
made with a LaserTech
laser level, and of depth with a Fish
Ray
hand-held SONAR depth finder. Bank-attached and
channel features were assessed on the basis of morphology,
composition and vegetation indicators of erosiondeposition
processes and hydroperiod. Field observations at selected sites
were also conducted at various times between 2000 and
2012, as shown in Table T3III.
In October, 2007, in connection with studies of bars in the
lower Sabine, several sections of the river were covered by boat
in low-water conditions, with continuous photography of the
channel using a GPS-enabled digital camera. This included
the entire reach from Deweyville to the Sabine River Authority
of Texas canal, several reaches between Burr Ferry and Bon
Wier, and several reaches between Bon Wier and Deweyville.
Also in October 2007, the river from Toledo Bend to
Deweyville was flown during clear-sky, clear-water, low-flow
conditions, and digitally photographed. This oblique photogra-
phy was obtained from a variable altitude of <200 m (~600 ft).
In March, 2010, an 8 km reach upstream of Burr Ferry was
re-examined by canoe to evaluate reported geomorphic
changes in this zone. In this same period, a number of oxbows,
cutoffs, sloughs and paleochannels throughout the study area
were examined by kayak to obtain more detail on HUs outside
the main channel.
High-resolution (1 ft or 0.3 m) vertical color aerial photogra-
phy from the US Army Corps of Engineers was obtained, cover-
ing the study area from upstream of Bon Wier to Sabine Lake.
The photography was flown shortly after Hurricane Rita struck
southeast Texas/southwest Louisiana, making landfall in the
Sabine Lake area in September 2005. This imagery was
previously used to inventory tree blowdowns from the
Table II. Hydrologic regimes in major reaches of the lower Sabine River
Reach Dam pulses Channels Overbank
flow
Channelfloodplain
connectivity
Coastal effects
1 Toledo Bend to
Burr Ferry
Flow dominated by dam
releases
Single channel Rare Low None
2 Burr Ferry to Bon Wier Flow strongly influenced by
dam releases
Single channel Occasional Low None
3 Bon Wier to Big
Cow Creek
Flow strongly influenced by
dam releases
Multiple channels at
high flows
Occasional Moderate None
4 Big Cow Creek to
Shoats Creek lower
Strong influence of dam
releases at low flow
Multiple channels at
high flows
Occasional High None
5 Shoats Creek to Cutoff Bayou Minor influence Multiple channels Common Extensive Minor
6 Cutoff Bayou
to Sabine Lake
Minor influence at low flows Multiple distributary
channels
Common Extensive Significant
Note:Coastal effectsrefers to influence of tides and coastal backwater effects.
Table III. Methods applied to the geomorphic zones (reaches) and years undertaken
Reach Boat traverse Fieldwork Low-altitude aerial photography
1 Toledo Bend to Burr Ferry 2007
1
20001, 2007, 2010 2007
2 Burr Ferry to Bon Wier 2007 20001, 2010 2007
3 Bon Wier to Big Cow Creek 2007 20001, 2007, 2008, 2010 2007
4 Big Cow Creek to Shoats Creek lower 2005, 2007 20001, 2008, 2010 2007
5 Shoats Creek to Cutoff Bayou 2005, 2007 2001, 2008, 2010, 2012 2007
6 Cutoff Bayou to Sabine Lake 2005, 2007,
a
2010,
a
2012
a
2001, 2005, 2010, 2012 None
a
Partial coverage.
Note:Boat traverseindicates travel by small boat with continuous photography of both banks. Fieldworkindicates field measurements and
observations at specific sites within the reach. Low-altitude oblique aerial photographyis by the author, during low- and clear-water conditions.
High-resolution vertical aerial photography and satellite images for all reaches are available for the period 20012014 (earlier and later images are
available, but were not used in this study).
4 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
hurricane in the lower Sabine and Neches rivers (Phillips and
Park 2009). Other aerial and satellite imagery taken at various
times between 2001 and 2014 was also accessed through
Google Earth
.
The various methods used to identify GUs and HUs are
summarized in Table III.
Geomorphic and hydraulic units
In some cases there is a one-to-one correspondence between
GUs and HUs; for example, in the study area the GU of an
unvegetated convex bank is associated with a single HU. In
other cases, there are multiple HUs associated with a single GU.
GUs and HUs were evaluated with respect to five reference
flows. These are based on critical stages of channel inundation
rather than reference flows based on recurrence intervals.
Flows of a given probability vary widely within the study area
with respect to their relationship to the channel. For example,
banktop flow has a recurrence interval of about 0.1% (based
on mean daily discharge) at the gaging station near the bound-
aries of zones 1 and 2, 3% at a station near the upper end of
zone 3 and 13% at the station within zone 5 (Phillips 2008a).
This variability, and its relationship to HU diversity and con-
nectivity, is addressed in the discussion section.
The lowest reference flow, thalweg connectivity, is the mini-
mum discharge required to maintain continuous downstream
water movement. Discharges less than thalweg connectivity
stage are essentially zero flow. Gage station records for the
lower Sabine do not record any zero-flow situations. Bed
inundation is the flow level at which the entire channel bed is
underwater and all mid-channel features are at least partially
inundated. The high within-bank (also called sub-banktop)
level is the higher range of flows that can occur before
overbank flow begins. Channelfloodplain connectivity (CFC)
flows are those that result in river-to-floodplain flow via
crevasse and tie channels, high-flow distributaries and
anabranches, and tributary backwater flooding. Depending
on local channel and levee morphology, this can occur at
sub-banktop levels. Flood or overbank flows are defined as
those that result in levee overtopping. The term bankfull is
avoided here because it is often used in compound or incised
channels to refer to stages below the morphological banktop,
and it is also associated (with varying degrees of justification)
with recurrence intervals of the order of 12 years.
The thalweg connectivity stage inundates only the thalweg
and pools. Bed inundation stage covers all channel units except
higher portions of marginal bars. High within-bank stages are
considered to bring flow in contact with all channel and
bank-attached units. Levels between bed inundation and sub-
banktop partially submerge bank-attached GUs. The CFC stage
can occur considerably below banktop, where distributaries,
semi-active anabranches, tie channels and levee gaps exist.
This is relevant to floodplain GUs, but not directly to the chan-
nel and bank units studied here.
Although some GUs and HUs are spatially discrete, any
categorization should be understood as a simplification of
factors that may vary continuously in space, and/or be sepa-
rated by fuzzy boundaries or transition zones rather than sharp
demarcations.
Richnessarea relationships
In bio- and pedodiversity studies richness versus area relation-
ships are established by enumerating the number of taxa in a
small sample area, and then successively enlarging the area
or adding sample sites. Here, due to the nested relationship be-
tween geomorphic zones (river styles), GUs and HUs, the geo-
morphic zones served as the sampling units. Thus, for example,
starting at Toledo Bend dam, zone 1 extends 21 km down-
stream and includes 46 HUs. Zone 2 extends to 82 km down-
stream and includes an additional 10 units (56 total). In this
way a distance (as a surrogate for area along the channel) ver-
sus richness (number of HUs) relationship was established. To
test the effect of directionality on the relationship, separate
analyses were performed starting at the upstream and down-
stream ends of the study area.
Connectivity and complexity
Because HUs are contained within GUs, in some cases the
connectedness of related HUs is the same; for example,
thalweg and bar surface HUs distinguished by different sub-
strates are equivalent in terms of their relationships to other
HUs. Thus the complexity analysis was based on aggregated
HUs. The analysis was confined to units associated with the
sand-dominated portion of the lower Sabine. This excludes
some bedrock units confined to zone 1, and some associated
with coastal backwater effects that occur only in zone 6.
Two graphs were constructed. In one, the nodes (aggregated
HUs) were considered connected if they occurred adjacent to
each other within the channel at least twice (to avoid spurious
influence of any single coincidences). In the second, units ac-
tive at similar flow levels were considered connected. The cat-
egories were thalweg connectivity (TC), TC to bed inundation
(BI), >BI but less than HWB, high within-bank (HWB), HWB
to banktop (BT) and overbank (OB).
An adjacency matrix Awas constructed for each graph, with
entries of zero if the row and column units were not connected,
and 1 if they were. A node was not considered to be connected
to itself. This represents a simple, unweighted graph with N
nodes (32 AHUs) and medges (connections). Spectral radius,
graph energy, Laplacian spectral radius and algebraic connec-
tivity were calculated for each.
The adjacency matrices have Neigenvalues λ
I
, such that
λ
1
λ
2
λ
N
. The largest eigenvalue (λ
1
) is called the spectral
radius, a standard and widely used measure of graph or net-
work complexity. λ
I
is sensitive to N,mand the nature of the
wiringof the edges, particularly the number of cycles (se-
quences of links that start and end at the same component).
λ
1
is highest for a fully connected graph, where every node or
component is connected to every other (λ
1
=N1 for that
case). The maximum largest eigenvalue for a given N,mis
λmax ¼2mN1ðÞ=N½
0:5(1)
Phillips Q7(2012) developed a method for determining the rela-
tive importance (ζ) of the number of connections (edges) in a
graph and the specific network of connections (wiring)to
the spectral radius. That is:
ζconnection ¼N1ðÞλmax
½=λmax λ1
ðÞ(2)
ζwiring ¼1ζconnection (3)
where λ
1
is the observed spectral radius, and ζ
wiring
+ζ
connection
=1.
Graph energy (the name is derived from its original applica-
tion in physical chemistry) is the sum of the absolute value of
the entire graph spectrum:
EgðÞ¼∑∣λi(4)
Positive eigenvalues indicate the strength of signal
amplification by the network (e.g. the rate at which, say,
5GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
drought-induced changes in one or more units would be prop-
agated). Negative eigenvalues denote the strength of signal
damping (negative feedbacks limiting the overall effects of, for
example, flood scour). Thus graph energy is an overall measure
of the intensity with which the network of connections re-
sponds via amplification and/or damping to changes.
The Laplacian matrix of Ais
L¼DA(5)
where Dis the degree matrix. In Dthe diagonal elements are
the degree of each node (number of edges incident to that
node), with all other elements 0. The largest eigenvalue λ
1
(L)
is called the Laplacian spectral radius, which is inversely re-
lated to dynamical stability. For graphs of the type analyzed
here, for example, the stability condition is
h<2=λ1LðÞ (6)
where his a time step for network change (Gupta et al. 2003).
This indicates that, as the Laplacian spectral radius increases,
the time step necessary to observe stability must be reduced.
The second-smallest eigenvalue of L(the largest non-zero
eigenvalue, as λ
N
[L]=0) is called algebraic connectivity
(α[G]) and is a measure of graph synchronization. Higher
α[G] indicates a greater propensity for changes to occur simul-
taneously or in rapid succession, and vice-versa.
Geomorphic applications of these spectral graph theory
methods are reviewed by Heckmann et al. (2015) and Phillips
et al. (2015). Spatially explicit graph and network analysis has
been extensively applied to study connectivity in landscape
ecology, and has also been applied recently to study geomor-
phic connectivity in fluvial systems (e.g. Czuba and Foufoula-
Georgiou 2015; Passalacqua 2017).
Results
Here only GUs relevant to banktop flows and lower are de-
scribed (i.e. floodplain and valley-bottom units not considered);
some examples are shown in FigureF2 2. Some of the GUs are
semi-permanent, in the sense of being relatively slowly
changed, present at low flows, continuing to be influential at
higher flows. Examples include pool and thalweg units. Others
are persistent: they are present at all flow levels, but only acti-
vated over part of the flow range. For instance, channel bank
units are only activated by flows significantly above bed inun-
dation. Some channel units, by contrast, may be inundated at
all but the lowest flows, but become irrelevant (do not influ-
ence hydraulics, sediment transport, channel morphology,
etc.) as they become deeply inundated.
Other GUs and associated HUs are ephemeral or transient to
varying extents. These include units associated with large
woody debris, and with algal or biotic mats and coatings.
While some areas have persistent log jams, most debris accu-
mulations are mobile at high within-bank and higher flows.
The biotic surfaces may chronically reappear in similar loca-
tions, but are ephemeral in that establishment and removal of
the biofilms is rapid and recurrent.
Technical reports (Phillips 2008b, 2011) provide more de-
tailed descriptions and a photographic example of each GU.
Channel units
Specific features within channels represent specific process
form associations and/or diagnostics of fluvial processes and
evolution. The channel units can be roughly categorized as
thalwegs, bedrock outcrops, bars, pool-related units and large
woody debris jams. Channel GUs and associated HUs are
shown in Table T4IV.
Resistant exposures or outcrops of bedrock locally limit rates
of bed and bank incision or erosion, and generally indicate re-
cent erosional removal of Quaternary alluvium. They occur
only in the reach from Toledo Bend to Burr Ferry (zone 1),
due mainly to scour following dam construction. Bedrock
GUs include mid-channel, channel margin and cross-channel
features.
Bars in the lower Sabine River are dominantly sandy, though
some mud (fine-grained) point bars occur in the lowermost
reaches, and bars in zones 1, 2 and 3 may include small
amounts of gravel. Bars may be marginal, mid-channel or con-
nector type.
Marginal (point and lateral) bars may occasionally be
breached by chute channels. Tributary mouth bars are delta-
like features that may occur as deltas per se at the tributary
mouth, or as spits aligned with the river channel and oriented
downstream. These bars may be breached by tributary or river
flow, and are associated with backwater effects on the tribu-
tary from the river. Diagonal bars are usually bank attached
in the Sabine, but may also occur as cross-channel features.
They are oriented diagonally to banks, with elongate, oval or
rhomboid planform shapes. Diagonal bars are formed where
flow is oriented obliquely to the longitudinal axis of the bar
and may indicate reworking of riffles. While diagonal bars
are usually associated with gravel or mixed-bed channels,
those in the Sabine are predominantly sandy. Forced bars are
associated with sediment trapping behind obstructions, and
may occur in mid-channel or attached to banks. All forced
bars observed in the study area were associated with large
woody debris.
Linguoid bars are often lobate in shape and have a slip face
on the downstream end. They are found at points of relatively
abrupt flow expansion, and in the lower Sabine often occur just
downstream (or at the downstream end of) flow constrictions
associated with point or lateral bars. Linguoid bars are associ-
ated with diverging flow with high availability of sand. Longitu-
dinal bars are mid-channel features oriented parallel to flow
and more or less streamlined, often with a downstream-
oriented teardrop shape. Longitudinal bars are deposited when
transport capacity is exceeded by sediment supply in mid
channel.
Changing flow and sediment transport conditions may lead
to the formation of several generations of different types of
bars in the same location. Further, downstream translation
of mid-channel bars may result in the welding together of
various combinations of point, lateral, diagonal, linguoid
and longitudinal bars. In either case the result may be com-
pound bars, characterized by traits of two or more of the
types described above.
Sand sheets are more or less uniform tabular sand sheets Q8oc-
cupying the entire channel. They are associated with bedload
deposition where sediment supply exceeds transport capacity,
and may exhibit a variety of bedforms. At low water, they
may resemble braided channels with multiple intertwining
subchannels. Sand sheets are readily reworked and may be
translated downstream during floods.
The bar types discussed above are generally recognized in
the geomorphology literature (see, for example, Brierley and
Fryirs 2005). In the lower Sabine an additional type occurs,
termed connector bars. Connector bars extend from the down-
stream end of a point or lateral bar to the upstream end of a
point or lateral bar downstream. They are distinct from the
linguoid bars that sometimes occupy the gaps between mar-
ginal bars in that they lack obvious downstream slip faces,
and are oriented parallel or diagonal to flows.
6 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Pools
In the study area riffles may be associated with linguoid or con-
nector bars or sand sheets, while pools are often associated
with the outer portion of meander bends. A glide or run is a
plane-bed section of channel that is neither pool nor riffle, asso-
ciated with an approximate balance between transport capac-
ity and sediment supply. Forced pools are associated with
flow obstructions such as large woody debris. These units
may be scour features downstream of resistant bedrock out-
crops or large debris pieces, or backwater pools from ponding
behind these obstructions. Backwater forced pools are also
found immediately downstream of some point bars.
Circular meander pools are, in planform, approximately cir-
cular enlargements at the apices of tight meander bends. They
are unusually deep, more so than normal meander pools as
much as three times the maximum depth of adjacent sections.
At least one circular meander pool was identified in the field
in the lower Sabine, and aerial photography suggests other
possible occurrences in reach 6.
Large woody debris (LWD; logs, trees, large limbs) is gener-
ally considered as HUs or microhabitat units rather than a
GU. However, significant LWD accumulations (jams), as
opposed to individual pieces of wood, represent form/process
interactions and are thus legitimate GUs. The largest LWD
jams occur in tributary mouths, where they may pond or de-
flect tributary inflow, and reflect backwater flooding and
recirculating eddies at high flows, where floating wood is de-
posited as flows recede. The second largest class of LWD jams
occurs along eroding river banks, where rapid recruitment of
toppled trees, coupled with entanglement of floating debris,
creates the jams. Mid-channel LWD jams are associated with
entanglement of LWD with large trees embedded in the bed.
These are both smaller and less frequent than the tributary
mouth or bank jams.
Figure 2. Some examples of GUs and HUs. (A) At 1 is flooded riparian forest, and just below 2 is a forced mid-channel sand bar. The portion of the
channel near 3 includes the thalweg and outer channel. An unvegetated convex bank (sand) is at 4, with vegetated convex banks in the background.
(B) Vegetated, stabilized fine-grained point bar with a convex bank that is normally unvegetated (at the time of the photo an extended low-flow period
allowed some vegetation establishment. The muddy channel margin is also exposed. [Colour figure can be viewed at wileyonlinelibrary.com]
Colour online, B&W in print
7GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Bank-attached units
GUs and HUs associated with the channel margins (TableT5 V)
include the channel banks themselves, and significant subunits
along the banks. GUs that lie partly within the channel (such as
marginal bars) are treated in the section on mid-channel units,
while GUs on bank tops connecting the banks and floodplain
(such as natural levees and crevasses) are considered under
floodplain/valley units.
Benches and ledges are low-relief shelf-like features along
channel banks and margins. These features are sometimes
termed channel shelves, particularly when no inferences
about their origins are drawn. Benches are depositional
features related to infilling. They are composed of the same
general type of sediments normally comprising the channel
bed, bars and banks, which is typically sand in the Sabine.
Ledges are morphologically similar, but are erosional features.
Ledges may also occur where an episode of incision cuts a
narrower channel into the former channel bed. Remnants of
the former bed appear as ledges inset into the channel banks.
Ledges of the latter type were not observed on the Sabine, but
have been documented on tributaries of the nearby Trinity and
Angelina Rivers (Phillips 2001b; Phillips et al., 2005 Q9), and may
exist on some lower Sabine tributaries, particularly in reaches
1 and 2.
Compositionally, GUs associated with sandy and muddy ma-
terial indicate the variety of soils and sediments in the Sabine
valley. Banks reflect a large number of combinations of mate-
rial, morphology and vegetation cover. The units were identi-
fied primarily on the basis of profile (bank top to channel bed)
shape, which reflects the cumulative impacts of the interactions
among channel and riparian hydrology, bank materials, vegeta-
tion and slope processes (mainly concave or convex GUs).
Undercut banks a subcategory of convex banks where a
portion of the bank overhangs and shades the water may be
of particular interest for aquatic habitats.
Table IV. Channel GUs and HUs
Geomorphic unit Hydraulic unit(s) Location Substrate Q
cr
Channel Outer bed Channel bed to bank transition A. Sand BI
B. Mud
Channel Central bed Mid-channel other than thalweg A. Sand >TC
B. Mud
Channel Coastal backwater pool Fluvialestuarine transition zone A. Mud TC
B. Sand
Thalweg Thalweg Mid-channel A. Sand TC
B. Sand and gravel
C. Mud
Thalweg Thalweg pool: Within thalweg Sand, mud, gravel TC
A. Sand
B. Sand and gravel
C. Mud
Sand sheet Sand sheet: Mid-channel Sand BI
A. Flat or plane bed
B. Ripples
C. Dunes
Mid-channel bars (forced,
linguoid, longitudinal, compound)
Bar surface: Mid-channel Sand BI
A. Flat
(a) Rippled
(b) Unrippled
B. Convex
(a) Rippled
(b) Unrippled
Mid-channel bars (forced,
linguoid, longitudinal, compound)
Mud bar Mid-channel Mud >BI
Linguoid bars Prograding front Downstream end of bar Sand BI
Mid-channel and lateral bars Cross-bar channel Bisecting mid-channel and
lateral bars
Sand BI to HWB
Mid-channel + mid-channel bars Shallow pool: Mid-channel between bars Mud, sand >TC
A. Mud overlying sand
B. Sand
C. Algal mat overlying
sand or mud
Forced pool Forced shallow pool: Mid-channel other than thalweg Sand, mud >TC
A. Sand
B. Mud
Channel, large woody debris (LWD) LWD Mid-channel Sand, mud BI to HWB
All bar types Biotic mat Lower bar surfaces Sand BI to HWB
Channel rock outcrop Convex bedrock Channel Bedrock BI
Channel rock outcrop Bedrock pool Channel Bedrock BI to HWB
Thalweg and meander cutbank Meander pool: Outer edge of meander bends Sand, mud TC
A. Sand
B. Mud
Anabranches High-flow subchannels Cypress swamp Regularly flooded subchannels Mud BI
Note:Q
cr
indicates the minimum flow at which the unit is inundated. TC, thalweg connectivity; BI, bed inundation; HWB, high within-bank;
BT, banktop.
8 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Bald cypress (Taxodium distichum) is a common riparian and
wetland tree in the lower Sabine (the name Sabine is derived
from the Spanish word for cypress), which can grow in satu-
rated or flooded conditions and develops characteristic
rampart-like subaerial roots called knees. The wide, buttressed
cypress trunks and knees, where they occur along banks,
provide a measure of erosion protection. Bald cypress is an
obligate wetland plant (grows naturally only in wetlands) but
cannot germinate in inundated conditions. Thus cypress
growing in conditions of normally standing water indicates a
local rise in water level subsequent to tree establishment, or
distinct seasonal variations in water level.
Features termed here flooded riparian forests also occur
within river banks (as opposed to floodplain swamp forests).
These occur on lateral bars, and are inundated at stages well
below banktop. Some cypress may occur here, but a variety
of other flood-tolerant species such as black willow (Salix nigra)
are present. Field observations suggest that these occur when
droughts and associated prolonged periods of low water allow
seedlings to become established during periods of limited
inundation and substrate mobility. Once established, these
river-edge forests may persist. However, there have been no
detailed studies of the ecology or geomorphology of these
flooded riparian forests within the study area, and little is
known of their longevity or persistence in the flooded state.
Rotational slumps may occur along eroding concave banks,
where significant vertical variations in material properties due
to soil strength and/or root mats results in rotational failures.
Active slumps are typically characterized by one or more trees
and associated understory vegetation with a root mat holding
the slumped material together. Eventual removal or dispersal
of the slumped material leaves a characteristic scallop-shaped
slump scar.
Other bank-attached units include bedrock banks in zone 1.
Chute channels are high-water channels across point bars,
which may eventually lead to chute cutoffs. Sand ramps are
sandy bank deposits observed in reach 2 that extend from the
channel to the natural levee. These are distinct from marginal
bars in that the latter do not extend to the top of the banks,
and from point bars in that the sand ramps are much narrower
and do not occur on the inside of meander bends. Little is
known of these features, but their position in the channel and
the presence of organic layers within the sand suggest that they
result from flow obstructions and temporary backwater effects
during high flows.
At least 21 floodplain and valley-bottom GUs occur
within the study area (Phillips 2008b), but these are not
discussed here.
Richness-area analysis
Inevitably, the greater the along-channel distance considered,
the more HUs are encountered. When the analysis is
Table V. Bank-attached GUs and HUs
Geomorphic Unit Hydraulic Unit(s) Location Substrate Q
cr
Convex bank Unvegetated convex bank Banks Cohesive soil, mud BI to HWB
Convex bank Vegetated convex bank Banks Cohesive soil, mud HWB
Convex bank Sand ramp Banks Sand HWB
Concave bank Unvegetated concave bank Eroding banks Sand, cohesive soil >BI
Complex bank A. Sand
B. Cohesive soil
Concave bank Bank overhang root mat Eroding banks Cohesive soil, sand HWB
Overhanging bank A. Cohesive soil
B. Sand
Concave bank Bank large woody debris Banks Sand, mud >BI
A. Sand
B. Mud
Concave bank Vegetated bank slump: Eroding banks Cohesive soil, sand HWB
Bank slump A. Cohesive soil
B. Sand
Bedrock ledge Bank rock outcrop Banks with exposed bedrock Bedrock HWB
Tributary mouth Flooded forest: Tributary mouths and lower
portions of banks
Mud, sand HWB
Various bank GUs A. Mud
B. Sand
Tributary mouth Tributary mouth bar Main channel just outside
tributary mouths
Sand, mud BI to HWB
A. Sand
B. Mud
C. Mud drape over sand
Tributary mouth
Distributary confluence
Recirculating eddy Backwater areas near confluences Mud, organic detritus HWB to BT
Abandoned channels Channel plug Infilled confluence with abandoned
channels
Sand, mud HWB to BT
A. Sand
B. Mud
Large woody debris (LWD) LWD jam Confluences with subchannels
or recently abandoned channels
Sand, mud >BI to HWB
Cypress fringe Cypress fringe Low banks; channelwetland transitions Mud, sand >BI
Cypress buttress Cypress stump fringe Lower banks Sand, mud >BI
Marsh Fringe marsh Channelwetland transitions in fluvial
estuarine transition zone
Mud, sand >BI
A. Mud
B. Mud overlying sand
Note:Q
cr
indicates the minimum flow at which the unit is inundated. TC, thalweg connectivity, BI, bed inundation; HWB, high within-bank;
BT, banktop.
9GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
conducted upstream to downstream, the trend is linear
(FigureF3 3), but with a slope of only about 0.14, as only new
units not encountered in upstream geomorphic zones add to
the total. The relationship is qualitatively similar in the down-
stream to upstream direction, but the best-fit relationship is a
logarithmic function (FigureF4 4). This occurs because 83% (60
of 72 HUs) occur in zones 5 and 6, with a much slower (flatter)
increase moving upstream.
The increase in HUs inundated as flow stages increase was
also examined. This considered only inundation of HUs, and
not deactivation of some units as they become deeply im-
mersed. As shown in FigureF5 5, there is an almost linear in-
crease, interrupted by a disproportionately large number of
HUs engaged at high within-bank flows. While an exponential
function (y= 6.505 ×
1.1844
) provides the best statistical fit
(R
2
= 0.99), a linear trend (y= 10.024×5.107) does almost
as well (R
2
= 0.98).
Graph complexity
In the sand-dominated portion of the lower Sabine River 32 ag-
gregated HUs (AHUs) were identified, and the flows at which
they are active and relevant identified (TableT6 VI). Both graphs
(based on spatial adjacency, and potential connectivity at sim-
ilar flow levels; see FigureF6 6) have N= 32. The number of links
or edges mis 239 for the spatial adjacency and 431 for the flow
connectivity case. The adjacency matrices are shown in
Appendices A and B. A fully connected graph (every AHU is
adjacent or connected to every other) would have m= 496.
Figure 4. HUs versus distance, downstream to upstream.
Figure 5. Number of aggregated HUs inundated at various flow
levels. TC, thalweg connectivity; BI, bed inundation; HWB, high
within-bank; BT, banktop; OB, overbank.
Figure 3.Q16 HUs versus distance, upstream to downstream.
Table VI. Aggregated HUs
Aggregated hydraulic units TC TC-BI BI >BI HWB HWB-BT OB
Point bar surface-mud
or sand
0000 0 1 1
Channel plug 0 0 0 0 0 1 1
Recirculating eddy 0 0 0 0 0 1 1
Bedrock pool
(rock, or mud veneer
over rock); bank
0000 1 1 1
Sand ramp 0 0 0 0 1 1 1
Vegetated convex bank 0 0 0 1 1 1 1
Flooded riparian forest 0 0 0 0 1 1 1
Bank overhang, root mat 0 0 0 1 1 1 1
Bank slump 0 0 0 0 1 1 1
Channel margin large
woody debris
0000 0 1 1
Bank rock outcrop 0 0 0 0 1 1 1
Lateral bar surface
1
0011 1 1 0
Lower point bar,
mud veneer
0000 1 1 0
Cross-bar channel 0 0 0 0 1 1 1
Unvegetated convex bank 0 0 0 1 1 0 0
Gravel veneer 0 0 0 0 1 1 1
Tributary mouth bar 0 0 0 1 1 1 0
Fringe marsh 0 0 0 0 1 1 0
Biotic mat
1
0001 1 1 1
Unvegetated concave bank 0 0 1 1 1 1 1
Cypress fringe 0 1 1 1 0 0 0
Cypress stump fringe 0 0 0 1 1 1 0
Mud bar 0 0 1 1 1 1 1
Cypress swamp 0 0 1 1 1 1 0
Mid-channel large
woody debris
0001 1 1 0
Bedrock pool
(rock, or mud veneer
over rock); mid channel
0011 1 1 1
Prograding front 0 1 1 1 1 1 0
Central bed (sand or mud) 0 1 1 1 0 0 0
Shallow pool (mud/sand,
sand, algal)
0111 0 0 0
Forced pool (sand or mud) 0 0 1 1 0 0 0
Coastal backwater pool 0 1 1 1 0 0 0
Thalweg (sand, sand/gravel,
or mud)
0111 0 0 0
Note: 1, 0 entries indicate whether the AHU is activated at the indi-
cated flows. TC, thalweg connectivity; TC-BI, between TC and BI; BI,
bed inundation; HWB, high within-bank; HWB-BT, between HWB
and bank top (BT); OB, overbank.
10 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
The maximum possible spectral radius for N=32 is λ
1
= 31.
Because
EG
ðÞ
2mN
ðÞ
0:5(6)
the maximum possible graph energy in this case is <125.98.
λ
N
(L) = 0 for any graph of this kind, and for the fully connected
case all other λ
i
(L)=N(and thus algebraic connectivity and
Laplacian spectral radius also equal N). Algebraic connectivity
is less than N1 for other cases, and for graphs of the type
considered here
λN1LðÞ4=N2=2
 (7)
This produces a minimum in this case of 0.008.
Properties of the graphs are shown in TableT7 VII. Mean degree
is the mean number of incident links for each node, and values
in Table VII reflect a high level of connectivity for the spatial
adjacency graph, and even higher for the flow connectivity
graph.
The spectral radius values (16.8 for spatial adjacency; 27.4
for flow connectivity) reflect highly complex networks. If it is
assumed that contiguity or flow connectivity implies that
changes in individual HUs can be propagated to connected
units, then this indicates a large potential for amplification of
changes or disturbances and reorganization of the system.
Ζ
connection
indicates the extent to which the reduction in λ
1
from
the maximum possible for Nis due to the presence of fewer
than the maximum number of edges, as opposed to the specific
network of interconnections (ζ
wiring
). This suggests that
3342% is associated with the specific wiring.
Laplacian spectral radius is insensitive to network structure
for low-Ngraphs (Phillips 2014), but is useful for graphs of
the size considered here. The values indicate a high level of po-
tential instability, consistent with the amplification suggested
by the spectral radius.
Graph energy is higher for the spatial contiguity graph, and is
about 61% of the maximum possible value for N= 32, m= 239.
E(g) of the flow connectivity graph is less than 42% of the
maximum for a graph of that size. The primary cause of this
difference is that the spectrum of the spatial adjacency graph
includes 12 negative eigenvalues with |λ|>1, while the flow
connectivity version has only four. Implications are discussed
below.
The most pronounced difference between the two is alge-
braic connectivity. The far higher value for the flow connectiv-
ity version is consistent with higher synchronization during
specific flow events.
Discussion
Diversity of geomorphic and hydraulic units
Fieldwork and data analysis for this project were completed
well before publication of Belletti et al.s (2017) system for sur-
vey and classification of river GUs, but the definition of GUs
and HUs here is, at least broadly, conceptually similar to their
scheme. In the lower Sabine River there exist a wide variety
of HUs. While ecological and biological aspects were not di-
rectly investigated, it is likely that these are associated with a
high degree of habitat diversity. Although some of these distinc-
tions may not be significant for macrofauna and vascular
plants, they probably are for a variety of smaller fauna, algae,
diatoms, etc. The 72 HUs identified are far more than the
half-dozen or so categories in a typical biotope classification.
Both the number and nature of HUs would no doubt vary
with spatial scale. Here, as a rule of thumb, features that are
typically smaller than about 1 m
2
in surface area were not con-
sidered; thus, for instance, the rippled surface of a point bar was
considered a HU, but not the individual ridges and swales of
the ripples. While sizes of individual units vary greatly, they
are roughly proportional to channel width. Banktop-to-banktop
widths are generally of the order of 100 m, ranging from 80 to
160 m through most of the study area, with occasional locally
narrower or wider sections. Width is greater in the lower delta,
in the vicinity of the port at Orange, TX. The thalweg width
ranges from <10% of banktop channel width to nearly 30%.
Figure 6. The sandy point bar, exposed outer channel margin, thalweg, meander pool and concave cutbank shown here are generally only con-
nected to two other units in terms of spatial contiguity. However, all may be hydrologically connected at higher flow levels. [Colour figure can be
viewed at wileyonlinelibrary.com]
Colour online, B&W in print
Table VII. Aggregated HU graph properties
Spatial adjacency Flow connectivity
Links (m) 239 431
Mean degree (N m
1
) 7.219 13.469
Spectral radius 16.834 27.399
ζ
connection
0.669 0.584
ζ
wiring
0.331 0.416
Laplacian spectral radius 25.430 32.000
Graph energy 75.960 71.092
Algebraic connectivity 1.937 16.288
11GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Bank heights (banktop to outer channel margin) are typically of
the order of 10 m. The maximum width of point bars is typically
approximately half the banktop channel width. Lateral bars
may approach half the channel width at their widest points,
but most range from about 10% to 35% of channel width.
Mid-channel bars and various types of pools vary widely in
size, with the larger ones half or more of the channel width at
their widest points, and several channel widths in length. The
smallest of these GUs are of the order of about 10% of channel
width in both dimensions.
The richnessarea relationships for the HUs are not the typi-
cal power functions found in biogeographic speciesarea
curves or pedodiversity studies, indicating a more linear in-
crease rather than the leveling-off indicated by the usual power
functions. The substitution of upstream drainage area for length
does not substantially influence the relationships and does not
change the shape of the curves. The use of length as a surrogate
for area is more appropriate in this case because only within-
banktop features are included.
This may be related to the inclusion of six distinct geomor-
phic zones or river styles. In biogeographic and pedologic
richness-area analyses, the flattening of the power function re-
lationship is often attributable to the finite number of habitat
types and/or taxa, so that as area or the number of samples in-
creases, the probability of encountering new species or soil
types falls. However, in this study each geomorphic zone, by
definition, presents different environmental controls with re-
spect to geological setting, hydrologic influences, antecedent
morphology and other factors. This, and the nature of GUs
and HUs (as opposed to a classification with a fixed number
of mesohabitats, biotopes or channel types), makes it far more
likely perhaps inevitable that some new types will be en-
countered as area or stream length is increased. Richness-area
analysis within a single zone or style would be instructive.
Any such studies should also account for possible directional-
ity, as examining HU richness in the upstream or downstream
direction produced differing results here. This also raises the
question of the extent to which different starting points might
affect the results of other geomorphic, pedologic or ecological
richness-area analyses.
The richnessarea trends are consistent with either
continuum-based or more complex patch-based concepts of
fluvial environments, with a steady increase in HUs, as op-
posed to the asymptotic power-function trend anticipated if
units simply repeat in different geomorphic contexts. Of course,
these trends need additional examination in other fluvial sys-
tems and environmental settings.
The linear increase in HUs inundated with the flow stage
would presumably appear as more of a step function if
correlated with quantitative stage measurements. The results
underscore the importance of occasional high flows for
maintenance of within-bank HUs, notwithstanding activation
of floodplain units.
Traditionally, reference flows are linked to probabilities or re-
currence intervals. This approach was not used here for two
reasons. First, the specific concern was with what portions of
the channel are inundated at particular flows hence the use
of direct reference flows such as thalweg connectivity, bed in-
undation, etc. Second, in the Sabine and other Texas coastal
plain rivers, key geomorphic and hydrologic threshold flows
are highly variable with respect to probability and not necessar-
ily consistently related to any given recurrence interval (Phillips
and Slattery 2007
Q10 ; Phillips 2008a, 2008b, 2013, 2015). For
example, there are three gaging stations within the study area
where discharge is regularly measured, known as the
Burkeville, Bon Wier and Ruliff stations (the Burkeville station
is near Burr Ferry; the Ruliff station is actually at Deweyville).
Table T8VIII shows the daily flows with 50% (median), 10%, and
1% exceedence probabilities, as well as the bank top or flood
stage discharge. The table also shows the estimated discharges
associated with thalweg connectivity, bed inundation and high
within-bank flows, as well as that associated with a stage where
channelfloodplain connectivity occurs. Table VIII shows that
these have no systematic relationship with the probability of
flows. Note that at Deweyville (Ruliff station) the Sabine has
an anastomosing planform, so that channelfloodplain connec-
tivity is almost constant, except at discharges well below the
median. The thalweg is below sea level here, so bed inundation
requires only a minimal upstream input. Because backwater ef-
fects were not considered in estimates of the benchmark flows,
the thalweg connectivity flow may be an overestimate. At Burr
Ferry/Burkeville, by contrast, the channel is more strongly in-
cised, with only a 0.1% chance of overbank flow, while the
probability of mean daily flows at or above flood stage is about
3% at Bon Wier and 13% at Deweyville. Note that the mean
daily discharge is more relevant to HU inundation than proba-
bilities based on annual peaks because it accounts for the num-
ber of days key flows are exceeded.
This highlights the importance of considering connectivity in
local context, at least at the scale of geomorphic zones or river
styles. For example, a mean daily flow with a 10% chance of
exceedence at Burkeville is only about a third of the discharge
required to activate crevasses, tie channels and backwater ef-
fects in tributaries to achieve channelfloodplain connectivity.
At Bon Wier, the 10% reference flow is very close to the
channelfloodplain connectivity stage, and at Ruliff it is
overbank. This is treated more extensively by Phillips (2008a),
who found low and infrequent channelfloodplain connectiv-
ity in the two geomorphic zones upstream of Bon Wier, moder-
ate and high connectivity in the next two zones, respectively,
and extensive flux connectivity in the downstream-most geo-
morphic zones.
Structural complexity and connectivity
Connectivity networks of AHUs indicate a highly complex
pattern. If every identified AHU were connected to every
other, the corresponding graph would have 32 nodes and
496 edges. With connectivity assessed on the basis of spatial
adjacency or contiguity in the field, the graph has almost half
of this, with a mean node degree of >7, indicating that each
AHU is connected to an average of more than seven others.
The spectral radius indicates a high degree of complexity,
and this, plus the Laplacian spectral radius, suggest dynamical
instability. The graph based on flow connectivity indicates
even greater connectivity and complexity, with a mean node
degree of almost 16.5, and even higher spectral radius and
Laplacian spectral radius values. In general, these findings
Table VIII. Reference flows at three lower Sabine River gaging
stations (m
3
s
1
). The 10% and 1% reference flows refer to mean
daily discharges
Reference flow Burkeville Bon Wier Ruliff
Median daily 75 97 128
10% 433 507 578
1% 940 1093 1272
Flood stage/bank top 1880 793 510
Channelfloodplain connectivity 1300 540 56
High within-bank ~1800 ~750 ~500
Bed inundation 17 99 16
Thalweg connectivity 1.8 8 10
12 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
indicate a high potential for propagation of changes through
the associated geomorphic and hydrologic units, and the po-
tential for rapid rearrangement following, for example, floods,
droughts and other perturbations. This further suggests that
species distributions and biodiversity hotspots are likely to be
constantly shifting.
Algebraic connectivity is much higher for the flow connec-
tivity versus the spatial contiguity graph. This underscores the
importance of connective flow events for synchronous change.
While connections dependent on contiguity can still produce
complex and rapid change, the synchroneity of those changes
will be much lower.
Graph energy is slightly higher for spatial version despite the
larger maximum eigenvalue of the flow connectivity graph,
reflecting higher values of negative eigenvalues in the contigu-
ity graph tending to slow down propagation of changes. This
indicates greater signal damping in the contiguity-based config-
uration, which offsets the greater amplification indicated by the
larger spectral radius and positive eigenvalues of the flow
version.
Graph energy deserves greater attention with respect to its
application to physical and ecological systems. Li et al.
(2012) (one of whose co-authors is the originator of the graph
energy concept) claim that E(G) does not necessarily have any
physical interpretation. In some physical chemistry applica-
tions, and as an approximation, E(G) is related to total -
electron energy
Q11 . Otherwise, according to Li et al. (2012), it is
an abstract graph invariant of primarily mathematical interest.
However, others attribute to the entire graph spectrum (as
opposed to λ
1
) more interpretive significance. For instance,
Tinkler (1972), who did not study graph energy, did analyze
the significance of eigenfunctions of transportation networks
represented as the same type of graph used here. He
showed that the non-principal eigenvalues represented the
growth rates of competing elements in the network (analo-
gous to growing/accreting or shrinking/eroding elements in
the AHU network). The number of positive, negative and
zero eigenvalues of spatial connectivity graphs is also an
important property for geospatial analyses (Griffith and
Luhanga 2011).
The use of graph theory to quantify connectivity in spatially
explicit networks has a long history in landscape ecology, and
has been growing in geomorphology and hydrology. Applica-
tions of algebraic or spectral graph theory to measure structural
complexity and link it to connectivity are still very few. How-
ever, this study suggests strong potential for such methods for
quantifying complexity, diversity and connectivity in fluvial
(and other geomorphic and hydrologic) systems.
Conclusions
GUs and HUs in river channels are closely linked to
geodiversity and habitats, and thus to biodiversity. However,
relationships between GU and HU richness and area (or length
of fluvial corridors) are not well understood. Likewise, the con-
nectivity and complexity of patterns of fluvial units have been
little studied.
In the lower Sabine River 72 different HUs were identified in
six geomorphic zones or river styles. Richnessarea relation-
ships show a linear or logarithmic increase of HUs, as opposed
to the less steep, asymptotic power functions generally found in
biogeographic speciesarea curves or in soil richnessarea
analyses. Different results are obtained when starting from the
upstream or downstream end of the study area, indicating the
importance of directionality in such analyses. These results in-
dicate that HUs (and related habitats and biotopes) are both
richer and more variable than a repeated sequence of units
(e.g. riffles, pools, runs). The number of HUs inundated in-
creases linearly with flow stage categories, accentuating the
importance of high within-bank flows in maintaining and acti-
vating HUs.
AHUs associated with similar GUs are highly connected,
both with respect to patterns of spatial contiguity and flux
connectivity at similar flow levels. Spectral graph theory met-
rics show a highly complex network with a high potential for
rapid propagation of changes for a graph representation of
connectivity based on spatial contiguity and even more so
for a graph based on flow connectivity. The flow connectivity
graph shows far higher synchronization, as indicated by alge-
braic connectivity. Thus suggests more rapid and coherent
changes for processes driven by river flow, as opposed to phe-
nomena driven by other factors between flow events. However,
the contiguity graph has higher graph energy, indicating greater
signal damping.
AcknowledgementsThis work was supported by contract numbers
0704830782 and 1000011022 from the Texas Instream Flow Program,
through the Texas Water Development Board.
References
Anderson JB, Rodriguez AB (eds). 2008. Response of Upper Gulf Coast
Estuaries to Holocene Climate Change and Sea Level Rise. Special
Paper Vol. 443. Boulder, CO.: Geological Society of America.
Baker K, Chadwick MA, Kahar R, Sulaiman ZH, Wahab RA. 2016. Flu-
vial biotopes influence macroinvertebrate biodiversity in South-East
Asian tropical streams. Ecosphere 7: e01479. https://doi.org/
10.1002/ecs2.1479.
Belletti B, Rinaldi M, Bussettini M, Comiti F, Gurnell AM, Mao L et al.
2017. Characterising physical habitats and fluvial hydromorphology:
a new system for the survey and classification of river geomorphic
units. Geomorphology 283: 143157.
Bracken LJ, Turnbull L, Wainwright J, Bogaart P. 2015. Sediment
connectivity: a framework for understanding sediment transfer at
multiple scales. Earth Surface Processes and Landforms 40:
177188.
Brierley GJ, Fryirs K. 2005. Geomorphology and River Management:
Applications of the River Styles Framework. Blackwell: Oxford.
Carbonneau P, Fonstad MA, Marcus WA, Dugdale SJ. 2012. Making
riverscapes real. Geomorphology 137:7486.
Clifford NJ, Harmar OP, Harvey G, Petts GE. 2006. Physical habitat,
eco-hydraulics and river design: a review and re-evaluation of some
popular concepts and methods. Aquatic Conservation: Marine and
Freshwater Science 16: 389408.
Q12Coffman DK, Malstaff G, Heitmuller FT. 2011. Characterization of geo-
morphic units in the alluvial valleys and channels of Gulf Coastal
Plain rivers in Texas, with examples from the Brazos, Sabine, and
Trinity Rivers, 2010. US Geological Survey Scientific Investigations
Report 20115067. Available: https://pubs.er.usgs.gov/publication/
sir20115067 [28 June 2017].
Covino T. 2017. Hydrologic connectivity as a framework for under-
standing biogeochemical flux through watersheds and along fluvial
networks. Geomorphology 277: 133144.
Czuba JA, Foufoula-Georgiou E. 2015. Dynamic connectivity in a flu-
vial network for identifying hotspots of geomorphic change. Water
Resources Research 51: 14011421.
Drakare S, Lennon JJ, Hillebrand H. 2006. The imprint of the geograph-
ical, evolutionary, and ecological context on species-area relation-
ships. Ecology Letters 9: 215227.
Eros T, Campbell Grant EH. 2015. Unifying research on the fragmenta-
tion of terrestrial and aquatic habitats: patches, connectivity and the
matrix in riverscapes. Freshwater Biology 60: 14871501.
Frissell CA, Liss WJ, Warren CE, Hurley MD. 1986. A hierarchical
framework for stream habitat classification: viewing streams in a wa-
tershed context. Environmental Management 10: 199214.
13GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Griffith GA, Luhanga U. 2011. Approximating the inertia of the adja-
cency matrix of a connected planar graph that is the dual of a geo-
graphic surface partitioning. Geographical Analysis 43: 383402.
Q13 Gumbricht TT, McCarthy J, McCarthy TS. 2004. Channels, wetlands
and islands in the Okavango Delta, Botswana, and their relation to
hydrological and sedimentological processes. Earth Surface Pro-
cesses and Landforms 29:1529.
Gupta V, Hassibi B, Murray RM. 2003. Stability analysis of stochasti-
cally varying formations of dynamic agents. Proceedings, 42nd IEEE
Conference on Decision and Control; 504509.
Hamilton SK, Kellndorfer J, Lehner B, Tobler M. 2007. Remote sensing
of floodplain geomorphology as a surrogate for biodiversity in a
tropical river system (Madre de Dios, Peru). Geomorphology 89:
2338.
Heckmann T, Schwanghart W, Phillips JD. 2015. Graph theory: recent
developments and its application in geomorphology. Geomorphol-
ogy 243: 130146.
Ibáñez JJ, De-Alba S, Bermúdez FF, García-Álvarez A. 1995.
Pedodiversity: concepts and measures. Catena 24: 215232.
Ibáñez JJ, De-Alba S, Lobo A, Zucarello V. 1998. Pedodiversity and
global soil patterns at coarse scales (with discussion). Geoderma
83: 171192.
Ibàñez JJ, Vargas RJ, Vázquez-Hoehne A. 2013. Pedodiversity: state of
the art and future challenges. In Pedodiversity, Ibàñez JJ, Bockheim
JG (eds). CRC Press: Boca Raton, FL; 128.
Johnson JB. 2005. Hydrogeomorphic Wetland Profiling: An Approach to
Landscape and Cumulative Impacts Analysis. Report EPA/620/R-05/
001. Corvallis, OR: US Environmental Protection Agency National
Health and Environmental Effects Research Laboratory.
Johnston CA, Bridgham SD, Schurbauer-Berigan JP. 2001. Nutrient dy-
namics in relation to geomorphology of riverine wetlands. Soil Sci-
ence Society of America Journal 65: 557577.
Li X, Shi LY, Gutman I. 2012. Graph Energy. Springer: Berlin.
Lomolino MV. 2000. Ecologys most general, yet protean pattern: the
speciesarea relationship. Journal of Biogeography 27:1726.
Milan DJ, Heritage GJ, Large ARG, Entwhistle NS. 2010. Mapping hy-
draulic biotopes using terrestrial laser scan data of water surface
properties. Earth Surface Processes and Landforms 35: 918931.
Moret SL, Langford WT, Margineantu DD. 2006. Learning to predict
channel stability using biogeomorphic features. Ecological Modelling
191:4757.
Newson MD, Newson CL. 2000. Geomorphology, ecology, and river
channel habitat: mesoscale approaches to basin-scale challenges.
Progress in Physical Geography 24: 195217.
Passalacqua P. 2017. The Delta Connectome: a network-based frame-
work for studying connectivity in river deltas. Geomorphology 277:
5062.
Phillips JD. 2001a. Divergent evolution and the spatial structure of soil
landscape variability. Catena 43: 101113.
Phillips JD. 2001b. Sedimentation in bottomland hardwoods down-
stream of an east Texas dam. Environmental Geology 40: 860868.
Phillips JD. 2003. Toledo Bend Reservoir and geomorphic response in
the lower Sabine River. River Research and Applications 19:
137159.
Phillips JD. 2008a. Geomorphic controls and transition zones in the
lower Sabine River. Hydrological Processes 22: 24242437.
Phillips JD. 2008b. Geomorphic units of the Lower Sabine River. Austin,
TX: Instream Flow Program Available: http://www.twdb.state.tx.us/
RWPG/rpfgm_rpts.asp.0704830782.
Q14
Phillips JD. 2011. Hydraulic units of the Lower Sabine River. Austin, TX:
Instream Flow Program Available: http://www.twdb.state.tx.us/
RWPG/rpfgm_rpts.asp.1000011022.
Q15
Phillips JD. 2013. Hydrological connectivity of abandoned channel
water bodies on a coastal plain river. River Research and Applica-
tions 29: 149160.
Phillips JD. 2014. State transitions in geomorphic responses to environ-
mental change. Geomorphology 204: 208216.
Phillips JD. 2015. Hydrologic and geomorphic flow thresholds in the
lower Brazos River, Texas, USA. Hydrological Sciences Journal 60:
16311648.
Phillips JD, Marion DA. 2007. Soil geomorphic classification, soil tax-
onomy, and effects on soil richness assessments. Geoderma 141:
8997.
Phillips JD, Park L. 2009. Forest blowdown impacts of Hurricane Rita
on fluvial systems. Earth Surface Processes and Landforms 34:
10691081.
Phillips JD, Slattery MC. 2006. Sediment storage, sea level, and sedi-
ment delivery to the ocean by coastal plain rivers. Progress in Physi-
cal Geography 30: 513530.
Phillips JD, Slattery MC. 2007. Downstream trends in discharge, slope,
and stream power in a coastal plain river. Journal of Hydrology 334:
290303.
Phillips JD, Van Dyke C. 2017. Geomorphological state-and-transition
models. Catena 153: 168181.
Phillips JD, Schwanghart W, Heckmann T. 2015. Graph theory in the
geosciences. Earth-Science Reviews 143: 147160.
Rice SP. 2017. Tributary connectivity, confluence aggradation and net-
work biodiversity. Geomorphology 277:616.
Shoffner D, Royall D. 2008. Hydraulic habitat composition and diver-
sity in rural and urban stream reaches of the North Carolina pied-
mont. River Research and Applications 24: 10821103.
Smith JM, Mather ME. 2013. Beaver dams maintain fish biodiversity by
increasing habitat heterogeneity throughout a low-gradient stream
network. Freshwater Biology 58: 15231538.
Solis RS, Longley WL, Malstaff G. 1994. Influence of inflow on sedi-
ment deposition in delta and bay systems. In Freshwater Inflows to
Texas Bays and Estuaries, Longley WL (ed). Texas Water Develop-
ment Board: Austin, TX; 5670.
Sullivan SMP, Watzin MC, Hession WC. 2004. Understanding stream
geomorphic state in relation to ecological integrity: evidence using
habitat assessments and macroinvertebrates. Environmental Manage-
ment 34: 669683.
TCB, Inc. 2006. Sabine Lake: Ecological Condition of the Sabine-
Neches Estuary. Prepared for Sabine River Authority of Texas and
Lower Neches Valley. Authority: Orange, TX.
Thomson JR, Taylor MP, Brierley GJ. 2004. Are River Styles ecologically
meaningful? A test of the ecological significance of a geomorphic
river characterization scheme. Aquatic Conservation: Marine and
Freshwater Ecosystems 14:2548.
Thorp JH, Thoms MC, DeLong MD. 2008. The Riverine Ecosystem Syn-
thesis. Academic Press: London.
TIFP (Texas Instream Flow Program). 2008. Texas Instream Flow Studies:
Technical Overview. Texas Water Development Board Report 369:
Austin, TX.
Tinkler KJ. 1972. The physical interpretation of eigenfunctions of di-
chotomous matrices. Transactions of the Institute of British Geogra-
phers 55:1746.
Tomanova S, Tedesco PA, Campero M, Van Damme PA, Moya N,
Oberdorff T. 2007. Longitudinal and altitudinal changes of macroin-
vertebrate functional feeding groups in neotropic streams: a test of
the river continuum concept. Fundamental and Applied Limnology:
170. https://doi.org/10.1127/1863-9135/2007/0170-0233.
Vannote RL, Minshall GW, Cummins KW, Sedell JR, Cushing CE. 1980.
The river continuum concept. Canadian Journal of Fisheries and
Aquatic Sciences 37: 130137.
Wheaton JM, Fryirs KA, Brierley G, Bangen SG, Bouwes N, OBrien B.
2015. Geomorphic mapping and taxonomy of fluvial landforms.
Geomorphology 248: 273295.
Wiens JA. 2002. Riverine landscapes: taking landscape ecology into the
water. Freshwater Biology 47: 501515.
Wohl E. 2017. Connectivity in rivers. Progress in Physical Geography
41: 345362.
Wohl E, Magilligan FJ, Rathburn SL. 2017. Introduction to the special
issue: connectivity in Geomorphology. Geomorphology 277:15.
14 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Appendix A
Adjacency matrix for aggregated hydraulic units. The letters on the column headings correspond to the AHUs in column 1. AHUs are
considered connected (cell entry of 1) if AHUs are inundated by similar flow levels, and unconnected (0) otherwise.
ABCDEFGHI J KLMNOPQRSTUVWXYZabcde f
Point bar surface 0 1 1 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 11100
Channel plug 1 0 1 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 11100
Recirculating eddy 1 1 0 1 1 1 1 1 1 1 0 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 11100
Sand ramp 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11111
Vegetated convex bank 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11111
Flooded riparian forest 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 1 1 11111
Overhanging root mat 1 1 1 1 1 1 0 1 1 1 0 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 11100
Bank slump 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 1 11111
Channel margin LWD 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11111
Lateral bar surface 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 11111
Lower point bar mud 0 0 0 1 1 1 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11111
Cross-bar channel 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 1 1 1 1 0 0 0 0 0 1 1 11111
Unvegetated convex bank 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11111
Gravel veneer (bar) 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 0 0 0 0 0 1 1 11111
Tributary mouth bar 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 11111
Biotic mat 0 0 0 1 1 1 0 1 1 0 1 0 1 0 1 0 1 1 1 1 1 1 1 1 1 1 1 11111
Unvegetated concave bank 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 11111
Cypress fringe 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 11111
Cypress stump/buttress 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 11111
Mid-channel LWD 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 11111
Prograding front 0 0 0 1 1 0 0 0 1 0 1 0 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 11111
Central bed sand 0 0 0 1 1 0 0 0 1 0 1 0 1 0 1 1 1 1 1 1 1 0 1 1 1 1 1 11111
Shallow pool sand 0 0 0 1 1 0 0 0 1 0 1 0 1 0 1 1 1 1 1 1 1 1 0 1 1 1 1 11111
Shallow pool mud/sand 0 0 0 1 1 0 0 0 1 0 1 0 1 0 1 1 1 1 1 1 1 1 1 0 1 1 1 11111
Shallow pool algal 0 0 0 1 1 0 0 0 1 0 1 0 1 0 1 1 1 1 1 1 1 1 1 1 0 1 1 11111
Forced pool sand 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 11111
Riffle 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 11111
Thalweg sand and gravel 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 01111
Thalweg sand 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 10111
Meander pool sand 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11011
Thalweg pool sand and gravel 0 0 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11101
Thalweg pool sand 0 0 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11110
15GEOMORPHIC AND HYDRAULIC UNIT RICHNESS AND COMPLEXITY
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Appendix B
Adjacency matrix for aggregated hydraulic units. The letters on the column headings correspond to the AHUs in column 1. AHUs are
considered connected (cell entry of 1) if AHUs are directly spatially contiguous, and unconnected (0) otherwise.
ABCDE FGH I J KL MNOPQRSTUVWXYZa bcde f
Point bar surface 0 0 0 0 1 0 1 1 1 0 1 1 1 1 0 0 1 1 1 1 1 1 0 0 0 0 1 00100
Channel plug 0 0 1 0 1 1 1 1 1 0 0 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1 0 11100
Recirculating eddy 0 1 0 0 0 0 0 0 1 1 1 0 0 0 1 1 0 0 0 0 0 0 0 1 1 0 1 00000
Sand ramp 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 00000
Vegetated convex bank 1 1 0 1 0 1 1 0 1 1 1 0 1 0 0 0 1 0 1 0 0 1 1 1 1 1 1 00000
Flooded riparian forest 0 1 0 0 1 0 0 0 1 1 1 0 0 0 0 1 0 0 0 0 0 1 1 1 1 1 0 00000
Overhanging root mat 1 1 0 0 1 0 0 1 1 1 0 1 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 00000
Bank slump 1 1 0 0 0 0 1 0 1 1 1 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 00000
Channel margin LWD 1 1 1 0 1 1 1 1 0 1 1 0 1 0 1 1 1 1 1 1 0 0 1 1 1 1 1 00000
Lateral bar surface 0 0 1 1 1 1 1 1 1 0 0 1 1 1 0 1 0 0 0 1 1 1 1 1 1 1 1 00000
Lower point bar mud 1 0 1 0 1 1 0 1 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 00100
Cross-bar channel 1 0 0 0 0 0 1 0 0 1 1 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 00000
Unvegetated convex bank 1 1 0 0 1 0 0 1 1 1 0 0 0 0 0 0 1 0 0 0 0 1 1 1 1 1 1 00000
Gravel veneer (bar) 1 0 0 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 0 0 0 1 0 0 0 0 1 00000
Tributary mouth bar 0 0 1 0 0 0 1 0 1 0 0 0 0 1 0 1 1 1 1 1 1 1 0 0 0 1 0 11000
Biotic mat 0 1 1 0 0 1 0 0 1 1 1 1 0 0 1 0 1 1 1 0 0 0 1 1 1 1 0 00000
Unvegetated concave bank 1 1 0 0 1 0 1 1 1 0 0 0 1 0 1 1 0 0 1 1 1 1 0 1 1 1 1 11111
Cypress fringe 1 1 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 1 1 0 1 1 1 1 1 1 11111
Cypress stump/buttress 1 1 0 0 1 0 1 0 1 0 0 0 0 0 1 1 1 1 0 1 1 1 1 1 1 1 1 11111
Mid-channel LWD 1 1 0 0 0 0 0 0 1 1 0 0 0 0 1 0 1 1 1 0 1 1 1 1 1 1 1 11111
Prograding front 1 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 1 0 1 1 0 1 1 1 1 1 1 11111
Central bed sand 1 1 0 0 1 1 0 0 0 1 1 0 1 1 1 0 1 1 1 1 1 0 1 1 1 1 1 11111
Shallow pool sand 0 1 0 0 1 1 0 0 1 1 0 0 1 0 0 1 0 1 1 1 1 1 0 0 0 0 1 11000
Shallow pool mud/sand 0 1 1 0 1 1 0 0 1 1 0 0 1 0 0 1 1 1 1 1 1 1 0 0 0 0 1 11000
Shallow pool algal 0 1 1 0 1 1 0 0 1 1 0 0 1 0 0 1 1 1 1 1 1 1 0 0 0 0 1 11000
Forced pool sand 0 1 0 0 1 1 0 0 1 1 0 0 1 0 1 1 1 1 1 1 1 1 0 0 0 0 1 11000
Riffle 1 0 1 0 1 0 0 0 1 1 0 0 1 1 0 0 1 1 1 1 1 1 1 1 1 1 0 11111
Thalweg sand and gravel 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 01111
Thalweg sand 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1 1 1 1 1 1 1 10111
Meander pool sand 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 1 11011
Thalweg pool sand and gravel 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 1 11100
Thalweg pool sand 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 1 11100
16 J. D. PHILLIPS
Copyright © 2017 John Wiley & Sons, Ltd. Earth Surf. Process. Landforms, (2017)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
... The focus is on the Holocene evolution of the lower Sabine River ( Fig. 10.1) along the Texas/Louisiana border. Despite many similarities among the Trinity, Neches, and Sabine, their geographical proximity, and commonalities in avulsion dynamics, landscape evolution differs significantly among the three because of local factors (Phillips, 2009(Phillips, , 2014(Phillips, , 2017. Thus, the stories must be told one river at a time. ...
... There exists a detailed inventory of hydraulic and geomorphic units in the lower Sabine (Phillips, 2017). Here the focus is on the larger forms, and those directly related to consequences and/or causes of avulsions and cutoffs. ...
Chapter
Earth faces serious contemporary and future environmental change. The principles of landscape evolution outlined here provide valuable lessons for understanding, interpreting, managing, and responding to those changes, which are reviewed here. The theory outlined in this book can be summarized as TREE: Transformative, Reciprocal, Emergent Evolution. Transformative evolution signifies continuous change, punctuated by state changes rather than incremental progression. Reciprocal represents the fact that landscapes are characterized by dense networks of interactions and by mutual adjustments. Landscape evolution is emergent because it is strongly directed and constrained by various forms of selection, leading to phenomena that arise spontaneously and are not preordained. The inherent, irreducible individuality (perfection) of landscape means that each has its own individual stories, the telling of which will ideally reveal commonalities with other landscapes while preserving, and even celebrating, their unique aspects. Two examples are presented: the evolution of the lower Sabine River valley, USA, largely driven by geomorphic change; and of the main ridge of the Šumava Mountains, Czech Republic, driven chiefly by ecosystem engineering impacts of vegetation.
... Several studies promoted the use of the framework of hydrogeomorphic patches for studying the spatial arrangement of river systems (Belletti et al., 2017;Phillips, 2017;Milan et al., 2010;Shoffner and Royall, 2008;Newson and Newson, 2000). According to Belletti et al. (2017), the analysis of relationships between patch scale geomorphic units, i.e., physical habitats and biota, can provide a physical basis for biological surveys with respect to habitat heterogeneity, composition, and attributes at a scale that is geomorphologically meaningful. ...
... According to Belletti et al. (2017), the analysis of relationships between patch scale geomorphic units, i.e., physical habitats and biota, can provide a physical basis for biological surveys with respect to habitat heterogeneity, composition, and attributes at a scale that is geomorphologically meaningful. Phillips (2017) points out the importance of patch scale over the repeated sequence of patches (e.g., riffles, pools and runs) by examining richness and diversity of HUs along a river corridor. ...
Article
Full-text available
Most fluvial systems exhibit systematic, continuous upstream to downstream variations in channel morphology and related ecological and hydrological parameters (emphasized by conceptual frameworks such as downstream hydraulic geometry and the river continuum concept), and discontinuous, shorter range variations (emphasized by hierarchical patch dynamics). This study investigates the relative importance of broader-scale upstream to downstream variation and local variation at the hydraulic unit scale in a bedrock-controlled stream in central Kentucky, USA. A nested ANOVA analytical approach was used to determine the relative importance of three nested spatial scales in explaining variations in channel morphology and riparian trees. Results show that channel morphology is largely controlled by local-scale variation explaining about 92% of slope, 46% of bankfull width, 99% of average depth, 54% of width-depth ratio, 86% of channel cross-sectional area, and 100% of the hydraulic radius of the channel. Different categories of substrate characteristics, however, represent anomalies with respect to variance explained at different levels. Furthermore, local-scale controls explain 60% of variations in species richness, 59% of variations in the total number of individual trees, 68% of variation in the proportion of Platanus occidentalis basal area and 43% of variation in the total number of biogeomorphic impacts. These results are consistent with the idea of tight coupling between channel morphology and riparian vegetation, although they do not, by themselves, prove such interactions. The morphological variation of the channel at the local scale is primarily attributable to the geological controls (e.g., faults, bedding planes, joints and fractures) and incision status associated with the study area. The local scale variation in vegetation pattern can be explained by the highly local edaphic differences along the riparian corridor, which is likely to be related to the local scale fluvial process-form variations, and biogeomorphic impacts and feedbacks. These patterns may therefore be common in bedrock rivers strongly influenced by geological controls.
... With respect to hydrological and geomorphological phenomena, this often occurs due to gradient, resistance, and efficiency selection. Pathways with the steepest flux gradients, landscape elements with the highest resistance, and configurations with the greatest efficiency for work (consistent with the least action principle) preferentially occur, recur, growth, and survive (Hunt, 1998;2017;Huang and Nanson, 2000;Phillips, 2010;2011;Smith, 2010;Nanson and Huang, 2017;2018). ...
Preprint
Full-text available
In Earth surface systems (ESS), everything is connected to everything else, an aphorism often called the First Law of Geography and of ecology. Such linkages are not always direct and unmediated, but many ESS, represented as networks of interacting components, attain or approach full, direct connectivity among components. The question is how and why this happens at the system or network scale. The crowded landscape concept dictates that linkages and connections among ESS components are inevitable. The connection selection concept holds that the linkages among components are advantageous to the network and are selected for and thereby preserved and enhanced. These network advantages are illustrated via algebraic graph theory. For a given number of components in an ESS, as the number of links or connections increases, spectral radius, graph energy, and algebraic connectivity increase. While the advantages (if any) of increased complexity are unclear, higher spectral radii are directly correlated with higher graph energy. The greater E(g) is associated with more intense feedback in the system, and tighter coupling among components. This in turn reflects advantageous properties of more intense cycling of water, nutrients, and minerals, as well as multiple potential degrees of freedom for individual components to respond to changes. The increase of algebraic connectivity reflects a greater ability or tendency for the network to respond in concert to changes.
... Examination of maps and images, and field experience in several Gulf of Mexico coastal plain rivers (e.g. Phillips, 2013Phillips, , 2014Phillips, , 2017Phillips & Slattery, 2006 suggests that the phenomena found in the Neuse are widespread in similar coastal plain settings. ...
Article
The fluvial‐estuarine transition zone (FETZ) of the Neuse River, North Carolina features a river corridor that conveys flow in a complex of active, backflooded, and high‐flow channels, floodplain depressions, and wetlands. Hydrological connectivity among these occurs at median discharges and stages, with some connectivity at even lower stages. Water exchange can occur in any direction, and at high stages the complex effectively stores water within the valley bottom and eventually conveys it to the estuary along both slow and more rapid paths. The geomorphology of the FETZ is unique compared to the estuary, or to the fluvial reaches upstream. It has been shaped by Holocene and contemporary sea‐level rise, as shown by signatures of the leading edge of encroaching backwater effects. The FETZ can accommodate extreme flows from upstream, and extraordinary storm surges from downstream (as illustrated by Hurricane Florence). In the lower Neuse—and in fluvial‐to‐estuary transitions of other coastal plain rivers—options for geomorphological adaptation are limited. Landscape slopes and relief are low, channels are close to base level, sediment inputs are low, and banks have high resistance relative to hydraulic forces. Limited potential exists for changes in channel depth,width, or lateral migration. Adaptations are dominated by formation of multiple channels, water storage in wetlands and floodplain depressions, increased frequency of overbank flow (compared to upstream), and adjustments of roughness via vegetation, woody debris, multiple channels, and flow through wetlands.
... Everard and Quinn, 2015). Yet, most of the relevant studies investigated the longitudinal aspect of fluvial geodiversityi.e., changes from upstream to downstreamand associated impacts on the channel's hydrological connectivity (e.g., Ibisate et al., 2011;Okin et al., 2015;Phillips, 2017). At the same time, results of this study demonstrate the importance of vertical geodiversityi.e., changes throughout the channel bed's profileto the stream's hydrological connectivity. ...
Article
The type of stratum in dryland ephemeral channel beds (wadis) determines its hydrological characteristics, which impact vegetation productivity. The objective of this study was to assess some key properties of a cemented , fine-grained bed stratum, previously characterized as a fluvio-pedogenic unit (FPU), as well as those of the overlying non-cohesive gravel bed stratum, known as a fluvially active unit (FAU). While the latter is an outcome of regular fluvial processes of recurring scour and fill, the former is characterized by horizonation, calcium carbonate deposition, oxidation, and accumulation of fine-grained material, defining it more as a soil or pseudo soil than as a sediment. The study was conducted in the hyper-arid southern Negev and Arava Valley of Israel, where the FAU and the underlying FPU were investigated, focusing on geo-ecological functioning. The results revealed considerable differences between the two types of strata. The contents of hygroscopic moisture, total organic carbon, and calcium carbonate were fourfold, 57%, and 36% greater, respectively, in the FPU than these in the FAU. Above all, the fine particle contents, including the clay and silt fractions, were almost eightfold and almost sixfold greater, respectively, for the FPU than those for the FAU, characterizing the former as loamy fine sand and the latter as fine sand. Altogether, these properties determine the stratum's field capacity and permanent wilting point, which were 73% and 50% greater, respectively, in the FPU than those in the FAU. In turn, the last two properties determine the available water capacity, which was almost twofold greater under the FPU than that under the FAU. Considering the (i) FPU's mean thickness of 0.6 m; (ii) FPU's horizonation into A, B, and C horizons; and (iii) a 70-90% decrease in available water capacity between FPU's A and B horizons, as well as between its B and C horizons, then the total available water capacity throughout this stratum's thickness is 20-87% greater under the FPU than that under the FAU. In addition, the FPU's cemented nature seems to provide a safe stratum that remains stable during floods, sustaining the establishment and growth of deep-rooted perennial vegetation. It is concluded that the FUP's properties have positive implications for vegetation growth and survival under the harsh climatic conditions across the region.
Article
A geomorphic unit is a landform that has been created and reworked by a particular set of earth surface processes. Each geomorphic unit has a particular morphology and sediment properties. Characteristic assemblages and patterns of geomorphic units reflect the use of available energy at any particular location in the landscape. In river systems the mix and balance of erosional and depositional processes creates characteristic, and sometimes distinctive, patterns of geomorphic units at the reach scale. As geomorphic units make up all parts of every valley bottom, the analysis of geomorphic units provides a universal resource with which to undertake systematic geomorphic analysis of river systems. In the first instance, this tool helps to interpret river morphodynamics. Particular process‐form associations determine what type of geomorphic unit is found where, how it is formed and/or reworked, and if/how that unit is related to adjacent units in the channel and/or floodplain. From this, particular assemblages of geomorphic units can be used to identify and map reach boundaries along a river course. Each reach has a particular set of process‐form relationships that determine (and/or reflect) the range of behaviour and the capacity for adjustment of that section of river. Framed in a catchment context and in relation to evolutionary trajectory, interpretation of geomorphic unit assemblages, and how they change over time, informs analysis of river condition and the potential for geomorphic recovery of each reach. A scaffolding framework to conduct such analyses and interpretations provides an important bridge between expert manual analysis and machine learning analysis using Big Data, allowing for the identification and interpretation of the distinctive traits of each and every river system.
Article
Full-text available
State-and-transition models (STM) are used to describe, model, interpret, and predict when landscapes will undergo a qualitative state change. Although rangeland ecologists pioneered STMs, geomorphological STM-type models were developed prior to and independently of ecological STMs. This study categorized 47 geomorphological STMs according to whether they were: based on single or multiple study areas; primarily for description and interpretation or predictive and prescriptive use; explicitly concerned with complex system dynamics; and the role of biogeomorphic interactions in the model. Each STM was represented as a graph and the structure identified. Spectral radii were calculated to measure the complexity of each STM. Although STMs are associated with conceptual frameworks that recognize the possibility of nonequilibrium, alternative states, and path dependency, results show that an explicit concern with complexity does not necessarily lead to the identification of more states and transitions, or a more complex transition pattern. The purpose for which a STM was created, as well as the number of study sites it can be applied to, also had little bearing on the models’ complexity. This review suggests that geomorphic STMs, rather than being used to fit explanations about landscape evolution into predefined theoretical categories, are veridical representations of empirical observations. Although STMs are particularly useful for grasping the biogeomorphological dynamics of landscapes, this review indicates their utility is not limited to biogeomorphology or to systems with a strong ecological imprint. Time scales involved in geomorphic change can make it difficult to observe a large number of states and transitions, which may constrain what types of STM structure can be identified, as the number of observed states and transitions required to develop particular graph structures varies widely.
Article
Full-text available
Given the widespread degradation of aquatic systems caused by land-use changes associated with palm oil production in South-East Asia, it is imperative to identify and study the remaining undisturbed rivers and streams. Stream macroinvertebrates are reliable indicators of environmental health. Linking the community structure of these organisms to natural hydraulic and geomorphic conditions (categorized as biotopes) is vital for the conservation and restoration of streams. This study characterizes the effects of biotopes on macroinvertebrate community structure in three streams within Ulu Temburong National Park in northern Borneo. Biotopes within these streams were categorized as either bedrock (waterfalls and cascades) or mixed substrate (riffles and pools). In total, 119 taxa were collected from all sampled biotopes, but not all taxa were collected from each stream. Biotopes were statistically distinct in terms of taxonomic richness, but not mean individual density or average community biomass. There were differences in community structure between waterfalls, cascades, pools, and riffles. The survey suggests that pool and riffle biotopes were more vulnerable to scouring flows and had similar community structure, while waterfalls and cascades likely experienced lower sheer stress during floods and had similar macroinvertebrate communities. This study has found that classification and mapping of macroinvertebrates with biotope theory in pristine, tropical streams is a useful framework for simplifying the many linkages between ecology, geomorphology, and hydrology. These natural patterns increase our understanding of tropical streams and can be used to assess the impacts of forest degradation.
Article
Full-text available
Hydrologic connections can link hillslopes to channel networks, streams to lakes, subsurface to surface, land to atmosphere, terrestrial to aquatic, and upstream to downstream. These connections can develop across vertical, lateral, and longitudinal dimensions and span spatial and temporal scales. Each of these dimensions and scales are interconnected, creating a mosaic of nested hydrologic connections and associated processes. In turn, these interacting and nested processes influence the transport, cycling, and transformation of organic material and inorganic nutrients through watersheds and along fluvial networks. Although hydrologic connections span dimensions and spatiotemporal scales, relationships between connectivity and carbon and nutrient dynamics are rarely evaluated within this framework. The purpose of this paper is to provide a cross-disciplinary view of hydrologic connectivity — highlighting the various forms of hydrologic connectivity that control fluxes of organic material and nutrients — and to help stimulate integration across scales and dimensions, and collaboration among disciplines.
Chapter
Full-text available
This chapter provides an overview of the Riverine Ecosystem Synthesis (RES), which is an integrated model derived from aspects of other aquatic and terrestrial models proposed from 1980 to 2007, combined with the perspectives on functional process zones (FPZs) and other aspects of riverine biocomplexity. The RES pertains to the entire riverine landscape, which includes both the floodscape and the riverscape. This contrasts with many lotic models whose primary emphasis or support focuses on main channel systems within headwaters. This synthesis, which incorporates the ecosystem consequences of spatiotemporal variability across mostly longitudinal and lateral dimensions, has three broad components: a fundamental, physical model describing the hierarchical patchy arrangement of riverine landscapes within longitudinal and lateral dimensions based primarily on hydrogeomorphology and emphasizing a new geomorphic division (an FPZ) between the reach and the valley scale; ecological implications of the physical model in terms of an expandable set of 17 general to specific (testable) hypotheses, or model tenets, on biocomplexity, which is applicable in some form to both pristine and altered riverine landscapes; a framework for studying, managing, and rehabilitating riverine landscapes through the use of the hierarchical physical model and aquatic applications of the terrestrially derived hierarchical patch dynamics (HPD) model. The goal is to provide a framework for the development of a cohesive theory of riverine ecosystems over time rather than to produce a finished product. This chapter draws upon three primary components of river science that contribute to the study of riverine landscapes: lotic ecology, landscape ecology, and fluvial geomorphology.
Article
Connectivity describes the degree to which matter and organisms can move among spatially defined units in a natural system. River connectivity is typically described in longitudinal, lateral, and vertical dimensions within the river corridor and the watershed and can be conceptualized as a continuum from fully connected to disconnected over diverse temporal and spatial scales. Explicit characterization of connectivity helps understanding of disparities among short-term, local rates of flux and inferred fluxes over larger scales. Connectivity also strongly influences the response of rivers to natural and human disturbances. Investigations of connectivity facilitate the recognition of processes that cross traditional disciplinary boundaries, as well as understanding of nonlinear behavior and retention within rivers. Enhancing the ability to identify, quantify, and predict the processes that create and maintain connectivity is integral to the effective management of rivers.
Book
This book presents the most comprehensive model yet for describing the structure and functioning of running freshwater ecosystems. This "riverine ecosystem synthesis" (RES) is a result of combining several theories published in recent decades, dealing with aquatic and terrestrial systems. New analyses are fused with a variety of new perspectives on how river network ecosystems are structured and function, and how they change along longitudinal, lateral, and temporal dimensions. Among these novel perspectives is a dramatically new view of the role of hydrogeomorphic forces in forming functional process zones from headwaters to the mouths of great rivers. Designed as a useful tool for aquatic scientists worldwide whether they work on small streams or great rivers and in forested or semi-arid regions, this book will provide a means for scientists to understand the fundamental and applied aspects of rivers in general and includes a practical guide and protocols for analyzing individual rivers. This combination of theoretical and applied analysis is quite unique in running freshwater ecology. Specific examples of rivers in at least four continents (Africa, Australia, Europe and North America) serve to illustrate the power and utility of the RES concept.. Develops the classic, seminal article in River Research and Applications, "A Model of Biocomplexity in River Networks Across Space and Time" which introduced the RES concept for the first time. Provides a guide to the practical analysis of individual rivers using the Riverine Ecosystems Synthesis and extends its use from pristine ecosystems to modern, human-modified rivers.. An essential aid both to the study fundamental and applied aspects of rivers, such as rehabilitation, management, monitoring, assessment, and flow manipulation of river networks.
Article
Geomorphic units are the elementary spatial physical features of the river mosaic at the reach scale that are nested within the overall hydromorphological structure of a river and its catchment. Geomorphic units also constitute the template of physical habitats for the biota. The assessment of river hydromorphological conditions is required by the European Water Framework Directive 2000/60 (WFD) for the classification and monitoring of water bodies and is useful for establishing links between their physical and biological conditions. The spatial scale of geomorphic units, incorporating their component elements and hydraulic patches, is the most appropriate to assess these links. Given the weakness of existing methods for the characterisation and assessment of geomorphic units and physical habitats (e.g., lack of a well-defined spatiotemporal framework, terminology issues, etc.), a new system for the survey and characterisation of river geomorphic units is needed that fits within a geomorphologically meaningful framework.
Article
With the increased recent attention on understanding the complex pathways and fluxes of material, energy, and organisms and the multidimensional coupling between various geomorphic process domains, the 47th Binghamton Geomorphology Symposium was organized around the theme of Connectivity in Geomorphology. Six key subthemes emerged: sediment connectivity, hydrologic connectivity, geochemical connectivity, riverine connectivity, landscape connectivity, and modeling of connectivity. Papers presented at the symposium and collected in this special issue elucidate the theoretical and conceptual underpinnings of connectivity but also provide field-based and analytical approaches to document the fundamental role of connectivity in geomorphic and ecological functioning. Among the consistent themes that emerged from the papers and from discussion during the symposium were (1) the interconnectedness of diverse forms of connectivity, (2) the appropriateness (or not) of conceptualizing connectivity as a gradient or continuum, and (3) the importance of an explicit recognition of disconnectivity as a mechanism to retain ecologically important materials and organisms within landscapes. Knowledge gaps identified in discussions during the symposium include: identifying questions that can be usefully addressed within a framework emphasizing connectivity; identifying feedbacks that enhance and limit connectivity; and quantification – of diverse aspects of connectivity, of thresholds of connectivity, and of interactions among different types of connectivity.