Available via license: CC BY 4.0
Content may be subject to copyright.
Water 2022, 14, 3845. https://doi.org/10.3390/w14233845 www.mdpi.com/journal/water
Article
Sedimentation and Erosion Patterns of the Lena River
Anabranching Channel
Sergey Chalov 1,2,* and Kristina Prokopeva 1
1 Faculty of Geography, Lomonosov Moscow State University, 119991 Moscow, Russia
2 Institute of Ecology and Environment, Kazan Federal University, 4200097 Kazan, Russia
* Correspondence: srchalov@geogr.msu.ru
Abstract: Lena River is one of the largest “pristine” undammed river systems in the World. In the
middle and low (including delta) 1500 km course of the Lena main stem river forms complex ana-
branching patterns which are affected by continuous permafrost, degradation of the frozen ground
and changes in vegetation (taiga and tundra). This study provides a high-resolution assessment of
sediment behavior along this reach. Comprehensive hydrological field studies along the anabranch-
ing channel located in the middle, low and delta courses of the Lena River were performed from
2016 to 2022 including acoustic Doppler current profiler (ADCP) discharge measurements and sed-
iment transport estimates by gravimetric analyses of sediment concentration data and surrogate
measurements (optical by turbidity meters and acoustic by ADCP techniques). These data were
used to construct regional relationships between suspended sediment concentrations (SSC, mg/L),
turbidity (T, NTU) and backscatter intensity (BI, dB) values applicable for the conditions of the Lena
River. Further, field data sets were used to calibrate the seasonal relationships between Landsat
reflectance intensities and field surface sediment concentration data. Robust empirical models were
derived between the field surface sediment concentration and surface reflectance data for various
hydrological seasons. Based on the integration of in situ monitoring and remote sensing data we
revealed significant discrepancies in the spatial and seasonal patterns of the suspended sediment
transport between various anabranching reaches of the river system. In the middle course of the
Lena River, due to inundation of vegetated banks and islands, a downward decrease in sediment
concentrations is observed along the anabranching channel during peak flows. Bed and lateral scour
during low water seasons effects average increase in sediment load along the anabranching chan-
nels, even though a significant (up to 30%) decline in SSC occurs within the particular reaches of
the main channel. Deposition patterns are typical for the secondary channels. The anabranching
channel that was influenced by the largest tributaries (Aldan and Viluy) is characterized by the
sediment plumes which dominate the spatial and temporal sediment distribution. Finally, in the
distributary system of the Lena delta, sediment transport is mostly increased downwards, predom-
inantly under higher discharges and along main distributary channels due to permafrost-domi-
nated bank degradation.
Keywords: suspended sediment concentration; turbidity; ADCP; sediment budget; remote sensing;
anabranching channel; flow distribution
1. Introduction
The evolution of large Eurasian fluvial systems since the end of the 19th century has
been determined, according to recent studies, by changes that have affected the drainage
basins, notably climate changes at the end of the Little Ice Age, agricultural practices, and
successive phases of channel management involving embankments and dams [1,2]. How-
ever, the respective roles of these various control factors in modifying water and sediment
budget along the downstream reaches have still not been identified and compared. Due
Citation: Chalov, S.; Prokopeva, K.
Sedimentation and Erosion Patterns
within an Anabranching Lena River
Channel. Water 2022, 14, 3845.
https://doi.org/10.3390/w14233845
Academic Editor: Maria Mimikou
Received: 12 October 2022
Accepted: 22 November 2022
Published: 26 November 2022
Publisher’s Note: MDPI stays neu-
tral with regard to jurisdictional
claims in published maps and institu-
tional affiliations.
Copyright: © 2022 by the authors. Li-
censee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and con-
ditions of the Creative Commons At-
tribution (CC BY) license (https://cre-
ativecommons.org/licenses/by/4.0/).
Water 2022, 14, 3845 2 of 27
to complicated impacts of the anabranching channel pattern, inputs from tributaries
(some of which are among the largest rivers on Earth), existence of huge and complex
floodplains or in some case of artificial straightening of the channels and intensive
changes in channel processes, a number of gaps remain in our knowledge of erosion and
deposition, hydrophysical and ecological functioning and the evolution of the channel–
floodplain system of the largest rivers of Eurasia [3–5]. Knowledge on the sediment
transport of these rivers mainly relies on a very limited number of gauging stations, most
of which are downstream. The undammed rivers in this regard are even less studied
mostly due to their location in the remote northern territories of Russian Siberia [3,6–9].
Here, few rivers represent a reference for the natural state of fluvial processes and sedi-
ment transport, and the largest is the Lena which its main stem is unregulated and its total
volume of the regulated river runoff in the tributaries is less than 3% of the total flow.
In recent decades, the Lena River system was subject to a variety of studies investi-
gating different aspects of the hydrological as well as the geomorphological system com-
ponents [5,10,11], particular along Lena’s delta [12–14]. As far as degradation of the frozen
ground has been shown to be the main cause of the marked increase in water discharge
[15–17], the links between change hydrology, fluvial processes and sediment transport are
of particular interest. Some studies have identified how the aggradation patterns during
floodplain inundation and fluvial erosion patterns have a huge effect on the sediment
transport along the anabranching channel reaches [2]. Moreover, bank erosion due to com-
plex ice-triggered abrasion processes represent effective sediment sources, especially in
melting periods [14,18]. At the Lena River the recent channel changes have proven that
the global change deeply impacts the periglacial river [19]. Under the dominant role of
thermal erosion in bank retreat over sections of the anabranching fluvial system [20], the
Lena River offers an opportunity to investigate the impacts of ongoing re-adjustment of
the channel form to sediment transport.
Considering the notable shifts in hydromorphology along the middle and low
courses of the river, we aimed to understand the continuum in sediment transport features
along the river main stem. For that, we aimed at combining in situ and remote sensing
techniques which can support investigation at much broader spatial and temporal scales
than previously possible for the largest rivers. In situ measurements relied on gravimetric
and surrogate measurements based on optical and acoustic principles. The application of
surrogate measurements for the extended territory significantly enhances the in situ ob-
servations, but requires specific regional calibration curves for each drainage and stream
[21]. Among the most applicable approaches is the turbidity (T). T is a unit of the degree
of light scattering through a water column containing suspended organic (e.g., algae) and
inorganic particles. Light scattering increases with a greater concentration of suspended
particles. Turbidity is commonly measured in nephelometric turbidity units (NTU). An-
other promising tool is the acoustic technologies which are based on commercial acoustic
Doppler current profilers (ADCPs), recognized as potential tools for the quantification of
sediment transport in natural streams using echo intensity levels as a measure of acoustic
backscattering strength (backscatter intensity, BI). The remote sensing data also provides
a quantitative value of suspended sediment loads through earth observation data [22–29]
and relies on surface reflectance data (reflectance coefficient, ρ).
Altogether, a combination of these approaches is an efficient tool for identifying long-
term sediment concentration patterns over extended river reaches of such remote and
poorly monitored rivers as Lena. The present study was based on field measurements and
the application of remote sensing approaches to study spatio-temporal variation of sus-
pended sediment transport. As far as all of the mentioned approaches (T, BI and ρ) have
never been used in investigating the Lena River, we selected three case study areas located
in the middle and low course of the Lena River representing various types of anabranch-
ing permafrost-affected channel patterns to develop and further apply a set of in situ and
remote sensing tools to study the sedimentation patterns of the Lena River. The paper
focuses on three main objectives:
Water 2022, 14, 3845 3 of 27
1. To construct a set of regional empirical models between the surface sediment con-
centration and T, BI and ρ;
2. Examine the variability in surface sediment distribution revealed by satellite images
and total sediment budget by ADCP along the Lena anabranching channel;
3. Provide a comparative study of the erosion and sedimentation patterns between var-
ious parts of the river system with a particular focus on the impacts of the thermal
erosion processes on suspended sediment transport.
2. Study Area
The Lena River is the largest Arctic River, originating near Lake Baikal entering into
the Arctic Ocean via the Laptev Sea, crosses East Siberia from south to north, and has a
length of 4400 km. The Lena basin extends throughout almost all of East Siberia (Yakutia)
with an area of 2.5 × 106 km2. The Lena River contributes 524 km3 of freshwater per year,
or about 15% of the total freshwater flow into the Arctic Ocean [30,31] (Figure 1). Almost
the entire (approximately 78–93%) basin is located within the permafrost zone. Various
types of permafrost exist in the Lena basin, including sporadic or isolated permafrost in
the source regions, and discontinuous or continuous permafrost in the downstream re-
gions [32]. Unfrozen sediments (taliks) underlie the main channel of the Lena River due
to the warming of the large water mass [5]. In the middle and lower reaches of the Lena
River, the talik occupies only 10–15% of the total area of the channel [33]. It is in this nar-
row strip in the core zone of the interflow that the most intense changes are localized,
without capturing the entire channel as a whole [34]. River runoff in the study area is
mainly formed during the warm period from May to July (about 75–95%). River floods
start in the southern part of the basin in late April to early May and in the northern part
in late May to early June. The flood duration on most of the rivers lasts for 35–40 days.
About 20–30% of the spring flood is formed due to liquid precipitation. A distinct winter
low-flow period lasts for roughly 200 days on average, early November till early May,
with a long-term mean discharge of 1520 m3s−1. Air temperatures frequently reach the
−50 °C mark during the winter season, ice thickness is on average 0.32 m, and it frequently
exceeds 2.0 m in secondary branches. Rapidly, normally within 10 to 15 days, streamflow
increases due to melting.
In the upper valley the Lena River incises into the Trans-Baikal Highlands and the
Archean Aldan Shield and flows through narrow valleys with relatively steep gradients.
In the middle and low valley, the river drains the plateaus of the Siberian Platform and
forms few extended reaches of anabranching channels between Pokrovsk and Zhigansk
and then into the main downstream along the Lena delta. The large islands (1–5 km in
length in the middle course and up to 50 km in the delta) are the predominant feature of
the fluvial landscape, but the morphology and dynamics of these islands are different due
to gradients in the permafrost [32] and vegetation (from taiga forests to tundra), as well
as distinctions in their origin. In particular, islands of the middle course are formed by
alluvial processes and consist of alluvium; the islands of the delta are mainly composed
of a sandy material and alluvial peat. Sandy sequences covered by the so-called ice com-
plex (IC) form on the third terrace of islands. Ice complex are permafrost areas which usu-
ally consist of fine-grained loess-like sediments within organic-rich to peaty formations.
They have a high content of segregated ice and polygonal ice wedges several meters in
height and width [35]. The particular modes of fluvial processes within this anabranching
channel are formed under the impact of the largest tributaries—Aldan and Viluy Rivers.
Water 2022, 14, 3845 4 of 27
Figure 1. Lena River basin and the locations of case studies.
Our study was conducted at three main sections of the Lena River. Case study 1 (Fig-
ure 1) was located in the middle reach of the Lena River, between Pokrovsk and the con-
fluence of the Lena and Aldan Rivers. Upstream of Yakutsk, the floodplain widens, and
the river develops a complex branching of the river channel, numerous channels surround
the largest sand bars and forested islands. The width of the channel with islands can reach
10 km and the width of the river valley is up to 25 km. The river’s longitudinal gradient
sharply reduces, and the sediment observed on the bars consists of medium and fine sands
[10]. The mean annual water discharge recorded at the Tabaga gauging station located
here is around 8 300 m s−1 for the last decade. This is typical of an anabranching pattern in
the middle course of the Lena River. Alluvial vegetation on the Lena islands here is very
homogeneous and is dominated by willows.
Case study reach 2 (Figure 1) is the Lena channel from Aldan River to Zhigansk. Two
main tributaries flow into the Lena River in this section: Aldan (right) and Viluy (left). The
Aldan River is a higher energy system, transporting a larger fraction of sediment and finer
gravel up to the junction with the main river. Downstream of the Viluy River junction, the
Lena turns north and keeps this direction up to the Laptev Sea. The floodplain is asym-
metric, the right side of the river is flanked by the Verkhoyansky Mountains, while on the
left side, the floodplain can be more than 20 km wide with numerous complex anabranch-
ing and unstable channels, swamps, and thermokarst lakes.
The most downstream anabranching channel (case study 3) is located within the dis-
tributary delta system of the Lena River. With 16,500 m3 s−1 average annual discharge at
Kyusyur (gauging station located at the entrance of the Lena delta), here the river can be
defined as a “mega-river” [36]. The Lena River delta is the largest Arctic delta with an area
of about 29,630 km2 [12], with more than 6000 branches with a total length of 14,600 km
[37]. Here, four main distributaries represent the main sectors of the channel structure: the
largest is the Trofimovskaya branch; the second-largest branch by water discharge is the
Bykovskaya branch that turns sharply to the east after Sardakh Island and flows into the
Buor Khaya Gulf. The third and fourth-largest distributaries are the Olenekskaya branch
which flows west into the Kuba Gulf and the Tumatskaya branch which flows north into
Water 2022, 14, 3845 5 of 27
the Laptev Sea. The origin of this area is particularly different from the anabranching
channel of the middle course: delta terraces were not formed as deposits made by the
flooding river, i.e., river terraces [12]; rather, they were formed by Laptev sea-level fluc-
tuations over a long period of time and thus the surface is significantly higher, hampering
possible submerging.
3. Materials and Methods
3.1. Field Observations
Acoustic Doppler current profiler (ADCP) measurements along case study 1 were
performed from 25 June to 2 July 2020, 23 to 25 September 2021 and 10–20 June 2022 at
over 30 branches of the Yakutsk anabranching reach. The measurements were further pro-
cessed to calculate eight total river discharges as the sum of particular branches in the
valley cross-section (Figure 2a). The ADCP measurements within case study 2 were per-
formed from 3 to 10 July 2020 and from 3 to 6 July 2021 at 17 cross-sections within the
most extended anabranching reach below the confluence of the Viluy River (Saham ana-
branching reach) (Figure 2b). ADCP measurements in the Lena delta were conducted be-
tween 10 and16 August 2022 and included 25 ADCP discharge measurements in the
Bykovskaya, Bulkurskaya, Olenekskaya, Tumatskaya, Arinskaya branches and in the
main channel (Figure 2c).
Water 2022, 14, 3845 6 of 27
(b)
(a)
(c)
Figure 2. Schematic map of the sections and profiles, where SSC changes were calculated: (a) num-
bers represent the profile locations of the SSC sections in case study 1; (b) numbers represent the
profile locations of the SSC sections in case study 2; (c) numbers represent channels in the Lena River
delta (case study 3): 1—main channel; 2—Bykovskaya, 3—Trofimovskaya, 4—Tumatskaya, 5—
Olenekskaya branches; grey boxes are polygons for calculating average sediment concentrations.
The main part of the field data used in this study was related to gravimetric SSC
(mg/L) and water turbidity (NTU) measurements (Table 1). Water samples were pumped
out with a filterless, submersible 12 V pumps from three layers: surface, mid-section, and
near-bottom to account for the vertical distribution of the suspended sediment. For SSC
determination water samples were filtered through pre-weighed and pre-dried (at 105 °C
for two hours) membrane filters (pore size 0.45 µm) from the “Millipore” filtration system
(Germany manufacturer). The filter samples were then oven-dried (at 105 °C for two
hours) and re-weighed to determine suspended sediment concentrations. The mass of dry
sediment was determined by gravitational differences. The SSC was deduced from the
measured mass of the dry sediment and the water sample volume:
Water 2022, 14, 3845 7 of 27
𝑆𝑆𝐶 =𝑚2 − 𝑚1
𝑉
(1)
where m2 is the mass of the filter and dried residue (mg), m1 is the mass of the clean filter
(mg), and V is the volume of the filtered water sample (L). The mass of the membrane
filter and the wet sediment were measured with an accuracy of less than 0.001 g.
Grain-size analysis of the suspended sediments was conducted with the Laser Parti-
cle Sizer Fritsch Analysette 22 (Germany manufacturer). The average 50% (D50, µm) was
calculated for each sample using the measured grain-size distribution curves linearized
to double logarithmic coordinates and interpolated. The turbidity data was obtained in
NTU using a portable Hach turbidity meter (USA manufacturer). The 2100 P portable tur-
bidimeter combines microprocessor-controlled operations and Hach’s patented ratio op-
tics. By providing a direct digital readout in nephelometric turbidity units (NTUs), the
need for calculations or interpolation of calibration charts is completely eliminated. We
took three water samples at each river sampling point, measured the turbidity in situ, and
calculated the mean turbidity.
Table 1. Data characteristics used in empirical SSC models of the Lena River.
Data Type
Case
Study
Date of
Sampling
Samples
Water Discharges Range, m3s−1
D50, µm
Aver
age
Max
Min
Turbidity
(NTU)
1
20–29 June
and 8–10 July
2016
82
12,200–26,400 (by Tabaga gauging
station)
1
25 June–10
July 2020
58
14,400–25,500 (by Tabaga gauging
station)
21.8
25.3
13.0
2
27 June 2020
17
42,900 (authors measurement in
Saham anabranching system)
34.7
68.0
21.7
2
3–5 July 2021
20
19,700 (authors measurement in
Saham anabranching system)
14.7
21.8
6.89
1
20–23
September
2021
86
12,600 (authors measurement in
Yakutsk anabranching system)
11.3
16.3
6.33
1
10–19 June
2022
38
20,200–25,200 (authors
measurement in Yakutsk)
25.6
64.6
16.7
3
10–16 August
2022
22
31,300 (authors measurement in 4.7
km upper Stolb island)
15.0
18.6
10.9
Surface
water
samples
(SSC)
1
20–28 June
and 9 July
2016
39
12,400–26,400 (by Tabaga gauging
station)
2
7 July 2020
8
16,400 (by Tabaga gauging station)
21.8
25.3
13.0
3
13–15 August
2022
32
31,300 (authors measurement in 4.7
km upper Stolb island)
14.7
17.1
11.0
ADCP
1
25 June and 2
July 2020
8
20,300–25,500 (by Tabaga gauging
station)
21.8
25.3
13.0
1
23–25
September
2021
6
12,600 (authors measurement in
Yakutsk anabranching system)
11.3
16.3
6.33
1
10–19 June
2022
12
20,200–25,200 (authors
measurement in Yakutsk
anabranching system)
25.6
64.6
16.7
SSC and turbidity measurements were in many cases coupled with simultaneous dis-
charge measurements with ADCP using a Teledyne RDInstruments (TRDI) RioGrande
WorkHorse 600 kHz ADCP (USA manufacturer) unit mounted on a moving boat. For each
Water 2022, 14, 3845 8 of 27
sample at a depth profile, the boat was repositioned to its original location and sampling
was performed while drifting at the river water velocity. These measurements were used
to calibrate the relationship between the backscatter and suspended sediment concentra-
tions using the power-law least-squares fitting between the raw backscatter values (BI)
and the measured suspended sediment concentration (SSC) for the specific rivers and hy-
drological seasons [38]. For this purpose, only profiles with a sufficient amount of simul-
taneous SSC gravimetric and ADCP-based BI measurements carried out under constant
discharge conditions were considered. Further, an ASET software package for calibrating
and processing TRDI ADCP data to compute the suspended sediment load WR was used
to count suspended sediment loads measured by ADCP [39].
Particular in situ SSC measurements were collected concurrently during field cam-
paigns in the middle of the Lena River from Yakutsk to the Viluy River on 9 July 2016 (21
measurements corresponding to the flood water period) and from 20 June to 9 July 2020
(26 measurements corresponding to the end of the flood water period) (Figure 3a,b) and
were used for the calibration of Landsat images from 9 July 2016 and from 20 June and 9
July 2020. We also used SSC data during field campaigns in the Lena River delta from 13
to 15 August 2022 (32 measurements) and the available satellite images for these dates (16
August 2022) (Figure 3c). Surface water samples were collected at a depth of 0–0.5 m using
a pump water sampler, preserved in 5 L clean bottles at various locations along the river
channel.
(a)
(b)
(c)
Figure 3. Locations of the field measurements of the SSC in the middle reach of the Lena River: 9
July 2016 (a), 20 June till 9 July 2020 (b) and in the Lena River delta: 13–15 August 2022 (c) used for
regression model construction.
3.2. Processing Satellite Images Data
The estimation of suspended sediment concentration (SSC) from remote sensing im-
aging has been investigated since the 1970s [40]. Generally, it has shown both theoretically
Water 2022, 14, 3845 9 of 27
and empirically that optical bands are highly correlated with SSC [41], particularly in the
red wavelength range of 610–700 nm, if the SSC is within the range of 0–50 mg/L, near-
infrared wavelength range of 700–1000 nm, which correlate with an SSC for higher con-
centrations of SSC, but become uncorrelated for lower values and band combinations
(NIR used with a ratio of optical bands: usually the red band but sometimes the green
band) particularly with high SSC values [42]. The type of relationships between SSC and
reflectance can change depending on the presence of other optical materials, such as algae
or dissolved organic matter. There is generally a linear relationship in optical bands (blue,
green and red) and an exponential relationship for higher SSCs [43].
We used high-resolution satellite imaging from the Landsat satellite system to track
variations in SSC. The Landsat satellite system is freely available and provides 30 m spa-
tial resolution with a respective 16-day cycle sufficient to reconstruct surface sediment
concentrations in the main channel and branches of the Lena River during the long-term
period. Utilizing all four Landsat systems allowed for continuous monitoring of the sea-
sonal variations and patterns.
Landsat Collection-1 Level-1 data products were downloaded from the USGS Earth
Explorer website from 1992 to 2018 for the middle of the river and from 1999 to 2022 for
the Lena delta to create the Landsat database. Remotely sensed data used for this study
were Landsat 4 and 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus
(ETM+), and Landsat 8 Operational Land Imager (OLI). The study relied on 150 images:
90 were used to derive long-term SSC in case study 1 and 2 (middle reach of the Lena
River), 50 images for case study 3 (Lena River delta), and 10 Landsat 8 images were used
to calibrate and validate the “reflectance–SSC” model. Landsat images were first radio-
metrically calibrated and atmospherically corrected by the QUAC method (Quick atmos-
pheric correction) [44]. The digital number (DN) recorded by a satellite imaging system
depends on sensors (spectral bands, gains, offsets), sun elevation (time of year, latitude,
typically not time of day because most current satellite images are collected at around the
same time of day—10:40 am local time for Landsat), atmosphere (conditions-scatter-
ing/absorption), topography (slope and aspect). The conversions from DN to absolute
units provide a basis for standardized comparison of data in a single scene or between
images acquired on different dates or by different sensors excluding sun and atmosphere
influence [45]. Calibrated digital numbers (DNs) to absolute units of at-sensor spectral
radiance, top-of-atmosphere (TOA) reflectance were converted using equations and
rescaling factors for the Multispectral Scanner (MSS), Thematic Mapper (TM), Enhanced
Thematic Mapper Plus (ETM+), and Advanced Land Imager (ALI) sensors for the Multi-
spectral Scanner (MSS), Thematic Mapper (TM), Enhanced Thematic Mapper Plus
(ETM+), and Advanced Land Imager (ALI) sensors [46].
The Landsat data used in this study was atmospherically corrected using the DOS
(dark object subtraction) method. DOS models are used to remove the effects of atmos-
pheric scattering from images by subtracting the darkest pixel value that represents a
background signature from each band. This value can be the band minimum, an average
based upon a region of interest (ROI), or a fixed value [44,47]. A major advantage over
more rigorous radiative transfer models is that surface and/or atmospheric field data col-
lected at the time of overflight are not needed. Therefore, entirely image-based models,
such as DOS, can be applied to retrospective images as well as images of areas that are not
easily accessible [47].
We reduced a collection of images “to the common denominator”, which is called the
“reference” image for relative atmospheric correction. The “reference” was an image of
the middle reach of the Lena River from 7 July 2016 (path = 121, raw = 017) with a “dark
object” value of 0.0156. The remaining images of the series were reduced to first reference
values. On the series of images taken on different dates, dark objects have the same bright-
ness and any differences in the images are due to the influence of the atmosphere. In case
of the research reported here, the “dark object” was a clean thermokarst and the floodplain
Water 2022, 14, 3845 10 of 27
lakes were used as a “dark object”. Then, each satellite image calculated the differences
between the reflectance coefficients of “dark objects”.
3.3. Sedimentation Patterns Analyses
The processing of in situ and remote sensing data aimed to understand the erosion
and sedimentation patterns over the case study reaches along the middle and lower course
of the Lena River using three different approaches.
The first approach was based on the measured suspended sediment load WRs (kg s−1)
by ADCP, processed according to [38]. Using ADCP sediment load measurements we es-
timated daily sediment loads as WR = 86,400WRs, which was used to calculate the sediment
budget between adjacent cross-sections:
ΔWR = WRdown − (WRup + WRtrib)
(2)
where WRdown and WRup are the sediment loads at the downstream and upstream profiles
of a considered reach crossing river valley, respectively, and WRtrib is the sediment load
from tributaries along the reach. In the case of the anabranching channel, each WR value
includes the sum of the sediment load at each branch within a particular valley cross-
section. ΔWR > 0 corresponds to the erosion patterns, whereas ΔWR < 0 evidence deposi-
tion–aggradation patterns.
The second approach was based on the analyses of surface sediment concentration
maps generated by applying the retrieved empirical models SSC = f(ρ) from available
Landsat 5, Landsat 7, Landsat 8 images for each of the case studies. We collected 62 cloud-
free and open-water satellite images from all the analyzed sensors for the ice-free period
(June–September) from 1992 to 2018, which covered a high range of water discharge var-
iation (discharges from 4760 m3s−1 to 36,300 m3s−1 by Tabaga gauging station) for the chan-
nel reach between Pokrovsk and the Aldan River (case study 1). For case study 2 of the
Lena River (from the Aldan River to Zhigansk) we collected 30 images from 1999 to 2020.
Water discharges were calculated using data from the Tabaga gauging station on the Lena
River (185 km from the mouth of the Aldan River) and the Verkhoyansky Perevoz gaug-
ing station on the Aldan River (140 km from the mouth of the Aldan River). During this
period water discharges varied at the Tabaga station from 6850 m3s−1 to 31,400 m3s−1, at the
Verkhoyansky station from 3920 m3s−1 to 26,000 m3s−1. A total of 47 images from the 1999–
2021 period were used to determine SSC changes along the Lena River delta. The third
part of the data set covered water discharges from 17,000 m3s−1 to 72,200 m3s−1 at the
Kyusyur gauging station. All images were classified into two classes: water and other (in-
cluding land, cloud cover/shade, sand bar and aquatic vegetation) over a wide area
around each station. SSC maps were generated for the water class. Most of the operations,
such as digital numbers to reflectance calculation and water mask creation, were per-
formed in ArcGIS software. Tools for radiometric and atmospheric correction were per-
formed in the QGIS software using a semi-automatic classification plugin, which pro-
vided calibration coefficients for all the Landsat sensors. All available images were trans-
formed into the generated surface sediment concentration maps.
The third approach was based on surface sediment concentration SSC profiles. For
the case study area 1, we selected 20 cross-sections to calculate the longitudinal change in
SSC (Figure 2a). Similarly, 16 cross-sections were taken along case study 2 between Aldan
and Viluy confluences (Figure 2b). Between each of these cross-sections the SSC change
(ΔS0) was calculated on the basis of the difference between SSC at the upstream (SSC1)
and downstream (SSC2) locations:
ΔS0 = SSC2 − SSC1
(3)
Values for SSC1 and SSC2 were estimated as an average for every cross-section. Rela-
tive sediment changes ΔS (%) were calculated as:
Water 2022, 14, 3845 11 of 27
ΔS = ΔS0/SSC1
(4)
For case study 3 (Figure 2c) each of Equations (3) and (4) were applied for different
distributary river sections: the main channel and four branches: Bykovskaya, Tro-
fimovskaya, Tumatskaya, and Olenekskaya channels. Three possible ratios between the
sediment input and output were analyzed: SSC2 > SSC1, SSC2 < SSC1, and SSC2 ≈ SSC1,
which corresponded to either erosional (ΔS > 0) or deposition (ΔS < 0) patterns. Repre-
sentative sections and profiles were selected from each study area to calculate the surface
SSC changes (Figure 2).
4. Results
4.1. Set of Regional Lena River Models for SSC Determination
Based on in situ measurements and remote sensing data we constructed a set of re-
gional empirical models between the surface sediment concentration SSC and T, BI and ρ
for the Lena River (Table 2). Some of them were similar for the entire middle and lower
reaches of the Lena River (T and BI), while the ρ = f(SSC) generally varied between the
various hydrological seasons but can also be interpolated using a single relationship.
According to the measurements in case study 1, average D50 decreased from 25.6 µm
(flood water, 2022) to 11.3 µm (low water, 2021). In case study 2 measurements were car-
ried out only during flood water (27 June 2020) and low water (3–4 July 2021) and the
average D50s decrease from 34.7 µ m to 14.7 µ m, respectively. The average size of all the
measurements was 19.6 µm. For case study 3 we sampled during August 2022, when the
average D50 was 15.0 µ m. Additionally, we calculated the maximum and minimum of
D50 values. Maximum values were observed in case study 2 (Saham anabranching sys-
tem) during the summer expedition campaign at the end of July and the values increased
to 68.0 µm. Minimum values were observed during the period of low water, e.g., Septem-
ber 2021 measurements, and were 6.33 µm (Table 1).
Table 2. Lena River empirical SSC models.
Parameter
Study Reach,
Hydrological Season
Equation
R2
Turbidity T, NTU
Case study 1, 2, 3
SSC = 0.95T + 5.56
0.81
Backscatter intensity BI,
dB
Case study 1
SSC = 0.59e0.382BI
0.57
Reflectance coefficient ρ
(of the Landsat image
in the red spectral
band)
Case study 1, low water
season
SSC = 540ρ − 2.61
0.91
Case study 1,2 low water
season
SSC = 1046ρ − 13.0
0.94
Case study 1, spring–sum-
mer flood water season
SSC = 891ρ − 7.65
0.82
Case study 3, low water
season
SSC = 1470ρ − 24.7
0.62
Case study 1, 2, 3
SSC = 1.42e86.3ρ
0.66
The SSC = f(T) relationship was based on almost 360 simultaneous measurements
over case studies 1, 2 and 3 at different seasons and yielded a good correlation between
all the data. Consistently, a monotonic increase in SSC with increasing turbidity T was
observed (Figure 4). The SSC = f(BI) relationship was performed based on 18 measure-
ments carried out on case study 1, which led to the following relationship (R2 = 0.57).
Water 2022, 14, 3845 12 of 27
Figure 4. The SSC = f(T) relationship for the middle and low courses of the Lena River. 2016-
model: 82 measurements from 20 to 29 June and from 8 to 10 July 2016 (Lena River Yakutsk ana-
branching system). 2020-model: 58 measurements from 25 June to 10 July (Lena River from
Pokrovsk to Saham anabranching system). 2021-models: 106 measurements 2 expeditions: from 3
to 5 July (Saham branching system) and from September 23 to September 25 (Yakutsk anabranch-
ing system). 2022-model: 38 measurements from 10 to 19 June (Lena River Yakutsk anabranching
system) and 22 measurements from 10 to 16 August (Lena River delta).
Based on Landsat images we generated two models for different sections of the river
and different phases of the water regime using in situ observations and reflectance coeffi-
cients (𝜌) from the Landsat 8 remote sensing system. We selected simultaneous satellite
images for two expedition data sets: according to the data from 2016 (9 July) for case study
1 and data from the period of 20 June to 28 June 2020 for case study 2 (Table 3). The cali-
bration of the SSC = f(
𝜌
) relationship was conducted based on the combination of the SSC
field data with the available Landsat images. Subsequently, “reflectance–SSC” model cal-
ibration was performed as the empirical relationship between these two parameters, plot-
ted on a graph (Figure 5). On the total available data set we carried out statistical analyses
to generate the least regression models to predict the surface sediment concentration as a
function of the surface reflectance for each selected river. The equation of the approximat-
ing line presents the relationships of the image brightness of 𝜌 and SSC. The highest ac-
curacy was achieved using a single technique and preparation of the satellite data on the
image’s brightness to eliminate radiometric and atmospheric influence.
Table 3. Descriptive statistics associated with in situ SSC and Landsat satellite images used in the
study for “Reflectance–SSC” model calibration.
Path/Row
Acquisition Date
Date of Water
Sampling
SSC, mg/L
Sam-
ples (n)
Min
Max
Mean
Median
Std. Devi-
ation
121/017
9 July 2016
9 July 2016
21
7.16
13.0
9.79
10.5
1.89
122/016
25 June 2020
20 – 28 June 2016
18
10.9
20.5
16.2
16.8
2.68
124/015
9 July 2020
7 July 2020
2020/07/07
8
9.91
16.4
12.2
11.5
2.40
129/009
16 August 2022
13-15 August 2022
24
3.55
22.6
13.6
14.7
4.55
Water 2022, 14, 3845 13 of 27
Figure 5. A regression model derived for the Lena River. SSC—suspended sediment concentration
of water, ρ—the reflectance coefficient of the Landsat 8 images in the red band. The lines show the
regression equation lines from Table 3.
The verification of the revealed Landsat models was performed based on the inde-
pendently measured SSC data sets. The accuracy of the models was calculated using these
models to satellite image; the known field observation data were not used in model cali-
bration. For validation of the “2016-model” we used in situ measurements from the expe-
dition in 2020 and vice versa for the “2020-model”. Applying the 2016-model to the satel-
lite image from 25 June 2020 (field measurements on 20, 27 and 28 June 2020), the value of
the mean absolute error of the modeled concentrations was approximately 1.55 mg/L. The
average value of mean absolute percentage error was approximately 10%. The root means
square error (1.83) was low and acceptable. However, there were also large deviations in
the range of small and high values, reaching 26% (Table 4).
Table 4. Statistical comparisons between the modeled SSC retrieved from satellite data by the “2016-
model” and measured in situ SSC.
Date of Water
Sampling
Acquisition Date
SSC, mg/L
MAE,
mg/L
MAPE, %
MSE
RMSE,
mg/L
Observed
Modeled
20 July 2020
25 July 2020
14.6
16.6
2.0
13.7
3.37
1.83
20 July 2020
17.7
16.6
1.1
6.21
20 July 2020
19.1
16.9
2.2
11.5
20 July 2020
10.9
13.8
2.9
26.6
20 July 2020
17.1
16.8
0.3
1.75
20 July 2020
17
16.3
0.7
4.12
20 July 2020
17.2
17
0.2
1.16
20 July 2020
20.3
17.5
2.8
13.8
20 July 2020
20.4
17.6
2.8
13.7
20 July 2020
18.6
16.8
1.8
9.68
28 July 2020
13.4
15.5
2.1
15.7
28 July 2020
13.9
13.8
0.1
0.72
28 July 2020
11.7
14.7
3
25.6
28 July 2020
15.4
14.7
0.7
4.55
28 July 2020
17
16.1
0.9
5.29
28 July 2020
16.5
14.5
2.0
12.1
27 July 2020
15.4
14.7
0.7
4.55
Mean
1.55
10
MAE—mean absolute error, MAPE—mean absolute percentage error, MSE—mean squared error,
RMSE—root mean squared error.
According to the results from applying the “2020-model” to the satellite image of 9
July 2016 the mean absolute percentage error and the mean absolute error of the modeled
Water 2022, 14, 3845 14 of 27
concentrations were 1.26 mg/L and 15%, respectively, while the root means square error
by the “2020-model” was low, reaching 1.96 mg/L. However, there were also high errors
in the range of the maximum and minimum concentrations due to the “2020-model” was
calibrated for values within the range of 10 to 20 mg/L, while the measured values in 2016
were 7.3–13.0 mg/L. The “2020-model” showed low errors in the range of 9–13 mg/L (Ta-
ble 5).
Table 5. Statistical comparisons between modeled SSC retrieved from satellite data by the “2020-
model” and measured in situ SSC.
Date of Water
Sampling
Acquisition Date
SSC, mg/L
MAE,
mg/L
MAPE, %
MSE
RMSE,
mg/L
Observed
Modeled
9 July 2016
9 July 2016
10.7
11.5
0.80
7.48
3.86
1.96
9 July 2016
10.1
10.5
0.40
3.96
9 July 2016
9.1
9.4
0.30
3.30
9 July 2016
12.5
11.5
1.00
8.00
9 July 2016
10.7
10
0.70
6.54
9 July 2016
10.5
9.9
0.60
5.71
9 July 2016
11.5
11.2
0.30
2.61
9 July 2016
10.9
11.3
0.40
3.67
9 July 2016
11.0
11
0.00
0.00
9 July 2016
7.3
4.2
3.10
42.5
9 July 2016
7.3
5.1
2.20
30.1
9 July 2016
7.3
3.5
3.80
52.1
9 July 2016
7.2
3
4.20
58.3
9 July 2016
7.8
5.9
1.90
24.4
9 July 2016
8.3
6.9
1.40
16.9
9 July 2016
7.2
4.3
2.90
40.3
9 July 2016
13
12.3
0.70
5.38
9 July 2016
11.6
11.6
0.00
0.00
9 July 2016
11.6
11.8
0.20
1.72
9 July 2016
10.5
11.7
1.20
11.4
9 July 2016
9.6
9.9
0.30
3.13
Mean
1.26
15.6
MAE—mean absolute error, MAPE—mean absolute percentage error, MSE—mean squared error,
RMSE—root mean squared error.
The difference in accuracy of the two models was acceptable for determining the SSC
from satellite images. Higher statistical values of the “2016-model” were due to the since
the satellite information was collected at the same time as the individual SSC measure-
ments, that is, the change in water discharges did not affect in any way the changes in
SSC. A completely different calibration for the “2020-model”, which was developed from
satellite images from 25 June 2020 and ground-based observations, was used from 20, 27
and 28 June 2020.
4.2. Erosion and Sedimentation Patterns of the Lena River—From Pokrovsk to Aldan (Case
Study 1)
The long-term estimates of the erosion and sedimentation patterns of the Lena
River—from Pokrovsk to Aldan—were revealed from the Landsat images. We analyzed
62 SSC maps from 1992 to 2018 and emphasized two major types of sediment distribu-
tions: longitudinal increase (1) longitudinal decrease (2) in SSC (Figure 6). The suspended
sediment load changes (according to Equations (3) and (4)) revealed from satellite images
demonstrated changes in surface suspended sediment along the 200 km reach from
Pokrovsk to the Aldan River. The SSC maps showed the situation under bar-full and bank-
full discharges (> 20,000 m3s−1, Figure 6a,b) were related to the level that submerged barren
bars and the top of the water level before overbank submersion. These maps highlight the
Water 2022, 14, 3845 15 of 27
decrease of 10–20% SSC (from 26 to 20 mg/L along the main central branch) along the 200
km reach. The picture was relatively unstable due to the impact of water diversions over
the smaller branches and floodplains which can cause a very contrasting effect on the SSC.
Under the lower discharges (Figure 6c,d) a downward increase in suspended sedi-
ment concentrations occurred (Q < 20,000 m3s−1). In some cases, the downward trends
were not possible to detect due to sediment plumes extended under particular banks from
upstream sediment sources (tributaries or eroded banks). The sediment plumes of turbid
waters (SSC > 100 mg/L) are mostly associated with floods. Their length along the ana-
branching channel can be over 100 km. This situation happened in less than 10% of the
observed images, but it was typical for case study 1, as shown below.
(a)
(b)
(c)
(d)
Figure 6. SSC maps for the Lena case study 1 demonstrating examples of longitudinal sediment
concentration increase: (a) 15 June 2007 under a Lena discharge of 36,300 m3s−1; (b) 21 June 2021
under a Lena discharge of 24,200 m3s−1; and a decline: (c) 19 September 2013 under a water discharge
of 8680 m3s−1; (d) 30 July 2021 under a Lena discharge of 9480 m3s−1.
This reach includes four relatively separate anabranching reaches located in the val-
ley widening called Pokrovsk, Yakutsk (located near Yakutsk), Kangalas and Haityaa-
lakhsk (Figure 7). The reaches are divided by short reaches of confined river related to
their parent banks. The average data presented in the box plot demonstrates that each of
these anabranching reaches was associated with a downward decrease in SSC: the higher
SSCs are seen in the upper profiles of each of the reaches shifted by lower values of SSC.
Water 2022, 14, 3845 16 of 27
The rates of longitudinal increase in the suspended sediment runoff were highest (ΔS0 = 6
mg/L) along the upstream part of the middle reaches at the site of the flow distribution
and formation of a complex of floodplain anabranches. The most intensive decrease in
surface sediment concentration from 32–34 mg/L to 26–28 mg/L occurred along the Ya-
kutsk reach, characterized by the highest anabranching intensity (near Ponomarev island).
It was significantly affected the change in sediment concentration throughout the 200 km
reach of the Lena River under low water conditions. Figure 7 demonstrates the ΔWR (kg
s−1) data obtained from two ADCP campaigns in 2020 and 2021. The corresponding blue
dots in Figure 7 show similar trends revealed from satellite image SSC change. According
to these measurements, both at low (8000 m3s−1) and average (15,000 m3s−1) flow conditions
the erosion pattern dominated along the anabranching channel. ADCP data shows that
the average increase in suspended sediment load was 4 kg s−1 (2% of the sediment dis-
charge at the upper reach) and 6.4 kg s−1 (3% from the upper discharge) in the first case.
Figure 7. Longitudinal surface SSC changes (ΔS, %) and sediment load budget (ΔWR, kg s-1) in case
study 1 of the Lena River (non-outlier range, mean, STD and 25–75%-interval) (profiles in Figure 2).
Blue circles represent WR values during the flood-water period, and yellow circles during low-water
period.
We plotted the suspended sediment concentration changes for different intervals of
water discharges: less than 10,000 m3s−1, from 10,000 m3s−1 to 20,000 m3s−1 and more than
20,000 m3s−1 (Figure 8). Three discharges classes were identified based on their morpho-
logical effects. The water discharges (< 10,000 m3s−1) are related to conditions of low water
observed during a significant part of the year and corresponds to the lowest hydraulic
effects over the river channel. Bar-full discharge is observed here to the values higher than
20,000 m3s−1 and relates to the level that submerges barren bars and the top of the water
level before overbank submersion [16,17]. It lasts for about 30 days a year. When this water
level reaches the base of the bank, thermal and fluvial erosion begins [16]. The SSC
changes were calculated for groups of profiles: within the first anabranching channel near
Pokrovsk (between profiles 1 and 3); in the anabranching channel near Yakutsk (between
profiles 6 and 9), and within the Haityaalakhsk anabranching channel (between profiles
16 and 19). Sedimentation patterns dominated during flood water (discharge more than
Water 2022, 14, 3845 17 of 27
20,000 m3 s−1), when ΔS decreased to –4% and varied from +1 to –9% at the confidence
interval of 25–75%. The largest gradients of seasonal SSC variability are typical for the
Yakutsk branching system. The rates of longitudinal increase in suspended sediment run-
off were the highest (ΔS0 = 6 mg/L) in the upstream part of the middle reaches at the site
of the flow distribution and formation of a complex of floodplain anabranches. In contrast,
there was a decrease in the mean SSC in the branching system of the river near Ponomarev
Island (on average for 62 situations) with a gradient between the profiles 8 and 9 (ΔS0 = –
3 mg/L).
Figure 8. The surface SSC changes ΔS along case study 1 of the Lena River (62 situations in the
period from 1992 to 2018).
4.3. Erosion and Sedimentation Patterns at the Lena River from Aldan to Zhigansk (Case Study
2)
Along case study 2, the sediment flow is driven by the inputs from the Aldan and
Viluy Rivers. Here, the suspended sediments entering the merged river form plumes
which can dominate SSC distribution during the peak and lower stages. Among the 30
analyzed Landsat images, 13 images had cross-sectional differences in SSC changes—an
increase on the right bank of the Lena River (the influence of Aldan), 8 out of 30 images
showed cross-sectional differences in SSC associated with a decreased SSC on the right
bank (the influence of Lena). Nine images showed a uniform distribution of SSC in the
studied area. Figure 9 provides two contrasting examples of such plumes related to the
input of more turbid water from the Lena River (left) and the Aldan River (right). Due to
the higher water discharge rate and larger sediment yield, the plumes from the Aldan
dominate throughout the hydrological year.
The greatest differences in the SSC of the water flows in the cross-section of the river
channel were observed at the beginning of the model section (first profile), closest to the
confluence of the two rivers (Figure 9). Additionally, the decrease in the sediment differ-
ences with river cross-section was achieved 130–140 km from the Aldan and Lena conflu-
ences (profiles 11–12) where the river channel forms a relatively straight, narrow reach.
Water 2022, 14, 3845 18 of 27
Downstream, 140 km below the Aldan confluence under the anabranching intensity in-
crease the extended anabranching sector was associated with increased differences in the
sediment concentration under left and right banks. This was also impacted by the Viluy
River and smaller Lapiske River.
(a)
(b)
Figure 9. Examples of SSC maps along case study 2 demonstrating significant domination of the
Lena River water discharge over the Aldan river (a)—28 August 2009, QL/QA = 3.53
(QL = 17,700 m3s−1, QA = 5 100 m3s−1), discharge ratio group—1) and domination of the Aldan River
discharge over the Lena river (b)—31 August 2016, QL/QA = 0.74 (QL = 13,200 m3s−1, QA = 9 730 m3s−1),
discharge ratio group—4).
4.4. Erosion and Sedimentation Patterns at the Lena River Delta (Case Study 3)
We revealed contrasting patterns in the left and right sectors of the delta. The differ-
ences can be explained by the various sizes of the river channel (Table 6, Figure 10). A
downward SSC increase was dominant in the Bykovskaya and Trofimovskaya branches
which distributed over 86% of the total Lena discharge. Here, the downward increase of
the surface sediment concentration (for example, from 45 mg/L in the main channel to 60
mg/L in the Trofimovskaya branch) occurred in 75% of the observed images. A significant
drop in SSC was observed only within smaller distributaries, mostly Tumatskaya, which
distributed less than 10% of the total Lena discharge. The average relative sediment
change ΔS along the Tumatskaya branch was −22.7% and −8.96% along the Olenekskaya
branch (a decrease in SSC from 16 mg/L in the main channel to 6 mg/L in the Olenekskaya
channel). These channels were also characterized by the lowest slopes due to maximal
channel length. Morphological expression of these phenomena is extended in sandy bars
which are a dominant feature of both the Tumatskaya and Olenekskaya river channels. It
is interesting to note that that SSC maps showed contrasting patterns under different hy-
drological conditions.
Water 2022, 14, 3845 19 of 27
Table 6. Surface suspended sediment concentration changes along the main distributaries of the
Lena delta.
Branch
% of The
Total
Discharge *
Average
Channel
Slope,
m/km
ΔSmean
ΔSmax
ΔSmin
Numbers of Analyzed
Situations
ΔS > 0
ΔS < 0
In Total
Bykovskaya
(right)
26.9
0.012
9.87
64.8
−57.6
17
11
28
Trofimovskaya
(right)
59.9
0.011
14.4
72.6
−46.4
18
8
26
Tumatskaya (left)
6.5
0.009
−22.7
−2.65
−46.4
0
23
23
Olenekskaya
(left)
6.7
0.008
−8.96
51.6
−49.7
6
16
22
*—measurements performed between 12–15 August 2022 with a total Lena discharge of
31,300 m3s−1.
Figure 10. The suspended sediment changes ΔS (%) along the middle reach of the Lena River by
relative discharges of the particular distributaries.
5. Discussion
5.1. Regional Empirical Models Validation
The proposed surrogate model enhanced capabilities of in situ monitoring of the
Lena River. The observed models demonstrated specific relationships with sediment pa-
rameters. Regarding the turbidity units T (NTU), it was previously found that the increase
in suspended matter size led to the growth of b coefficient which is the angle of the SSC =
f(T) graph [21]. A comparison of two sets of measurements at the Yakutsk anabranching
system in September 2021 (Q = 25,200 m3s−1) and June 2022 (Q = 12,600 m3s−1) confirmed
grain-size influence. Mean diameter (D50) by September 2021 low-period measurements
was 11.33 µ m (a coefficient is 0.28) and the D50 value grew in the flood June 2022 period
to 23.44 (the coefficient was 1.38). Even though we built a single relationship for different
the hydrological conditions of the Lena River (Figure 4), some particular points on the
relationship indicated that fine particles of weight m had a stronger influence on water
Water 2022, 14, 3845 20 of 27
turbidity T, than the coarser particles of the same weight m. This indicates that the appli-
cation of the model should be examined especially under extreme low-flow and high-wa-
ter conditions. At the same time we noted that the proposed SSC = f(T) relationship is
similar to the coefficients observed in large plain alluvial rivers [48–50].
Among the set of models, the relationship SSC = f(BI) was characterized by lowest R2
values (Table 2). We attribute this to the fact that we used raw BI values, not corrected for
intrinsic and ambient noise [51], as well as the impact of grain sizes on light backscatter-
ing. We compared linear fits between BI and SSC for sand fractions (>50 µm). Full sus-
pended sediment concentration samples were reduced to macro class more than 50 µm in
size, which represents the concentration of sand fraction in the river flow (SSC>50 µm,
mg/L). The results indicated that R2 was increasing after changing SSC to SSC>50µm, from
0.59 to 0.67. These findings generally confirm previous research that a substantial correla-
tion with corrected backscatter and SSC exists, while raw backscatter intensity does not
reasonably predict SSC. Further research is recommended to proceed with corrected BI
values and to estimate the error values for each of the individual ADCP units used in the
study [38].
The obtained remote sensing models (Figure 5) were compared. The combination of
regression model graphs for the Lena River in 2016, 2020 and 2022 reflects the influence
of seasonal factors on the slope of regression equations lines. The type of model depends
on changes in water discharges which explains the suspended sediment concentration,
type, grain size, color and mineralogical composition of the sediments [43]. As the water
discharges increased, the suspended sediment concentration and grain size also increased
which influenced the spectral and optical water properties. As the grain size increased the
radiance values decreased [52]. For example, the average size of sediments in integral
samples according to the results of the 2016 expeditions was 96 µm, and surface samples
according to the results of the 2020 expeditions was 146 µm. These factors led to a change
in the slope of the regression equations lines.
5.2. Drivers of Sediment Budget along the Lena River
The contrasting erosion and sedimentation patterns were observed within the ana-
branching channel of the middle and lower Lena River which indicate that various drivers
affect the sediment transport in different sections of the river. Along case study 1 the SSC
drop along the river channel during high waters is explained by the inundation of vege-
tated islands and floodplains. During the high-water period, the surface sediment flows
from the main channel towards the floodplain was detected through the increased num-
ber of river–floodplain connections as well as an increase in channel (water surface) width.
For the analyzed images we counted the relative area of isolated floodplain lakes Flakes and
the total area of the surface waters along the anabranching channel Fchannel. A positive rela-
tionship (R2 = 0.65) was found between changes in suspended sediment concentrations ΔS
and the ratio Flakes/Fchannel which shows that the conditions of sediment concentration de-
cline were associated with the decrease in river–floodplain connections. In particular, an
increase in the Flakes/Fchannel ratio corresponded to the water level drop over the Lena River
during low-water season and further separation of particular floodplain water bodies (for-
mer branches of the river during high-water conditions) from the main channel. Higher
Flakes/Fchannel was related to a lower storage capacity due to the disconnectivity between the
floodplain and the main channel. The role of the floodplain to attenuate surface runoff
and capture sediment has been observed over largest anabranching rivers in the World
(e.g., Amur, Mekong, Yangtze River) [6,53–55]. This floodplain flux can take a large por-
tion of the total sediment flow of alluvial unconfined rivers during the high-water period,
e.g., sediment from the main channel amounts to nearly 50% of the annual sediment flux
entering the floodplain of the Amazon River [56]. Here, on the example of the Lena River
we found that under low-water conditions the disconnectivity of the floodplain and the
main channel led to transversal variability of SSC.
Water 2022, 14, 3845 21 of 27
During low water, the increase in sediment concentration was associated with the
intensification of riverbank erosion and sediment resuspension from the bottom to the
upper layer along the riffle sections [13]. This was illustrated by the transport capacity
(Rtr) which was counted according to the Rossinsky–Kuzmin equation [57] based on the
ADCP velocity maps constructed for the case study 1 reach [11] as a function of water
velocity (V) and channel depth (h) for each pixel of the Landsat-generated SSC map from
9 June 2016 (Lena discharge 12,400 m3s−1):
Rtr = 0.024 V3/hu
(5)
where u is a settling velocity. Additionally, on the SSC map sediment concentration was
averaged for 16 classes of velocity zones, delineated with an interval of 0.1 m s−1. This
yielded the significant equation (R² = 0.91) (Figure 11a):
SSC = 0.29 ln(Rtr) + 11.5
(6)
Considering the observed relationship between sediment concentration and sus-
pended matter grain size (Figure 11b), we conclude that the hydrodynamics exert control
over sediment sorting within the anabranching reach.
(a)
(b)
Figure 11. Relationship between the suspended sediment concentration (SSC) and (a) velocity
(transport capacity—Rtr) from the measurements from 9 July 2016; (b)– mean sediment diameter
(D50, µm) from the measurement from 23 to 24 September 2021.
Due to the changes in velocity, the most significant influence on surface sediment
concentration was from the small secondary branches which are a typical feature of ana-
branching channels. These branches can be relatively sinuous and extended over 10 km
of the floodplain, separated by large islands (up to a few kilometers in width). Along case
study 1 a significant drop in SSC was observed due to the dramatic increase in water dis-
charge and water velocity. Here, both in the low- and peak-water period, the surface sed-
iment flows from the main channel towards the secondary branches declined up to 50%,
and in some case from 25 mg/L to 10 mg/L within rather short distances (less than 1 km)
(Figure 6). Here, additionally, significant sediment sorting occurred. During the low-wa-
ter period (by the data from 24 September 2021) in the 23 km, small left secondary
branches of the Yakutsk anabranching system a decrease in SSC from 7.67 mg/L to
2.25 mg/L and mean diameter (D50) from 10.9 µm to 6.54 µ m occurred.
The drivers of the anabranching channel within case study 2 were related to the in-
fluence of the merging rivers. For statistical assessment we divided each profile into two
parts, left-bank and right-bank, according to the geometric center of the anabranching
channel (Figure 9) to analyze the effect the water flow on SSC changes. A significant dif-
ference in the SSC flow along the left and right banks of the Lena River was observed as a
Water 2022, 14, 3845 22 of 27
function of the water discharges of the two rivers (Lena and Aldan). To perform this as-
sessment, we compared SSC maps from 25 Landsat images with water discharges for
gauges from two rivers—the Tabaga gauging station of the Lena River (QL) and
Verkhoyansky Perevoz of the Aldan River (QA) (the subscripts L and A represent the Lena
and Aldan Rivers, respectively). Further, the daily ratio QL/QA was calculated which
ranged between 0.88 and 3.53 (Table 7). According to the rank, all cases were classified
into four categories: (1) the Lena River water discharge significantly exceeds the Aldan
River water discharge (QL/QA > 1.75), (2) the Lena River water discharge slightly exceeds
the Aldan River water discharge (1.25 < QL/QA < 1.75), (3) similar water discharges in both
rivers (0.75 < QL/QA < 1.25) and (4) the Aldan River water discharge slightly exceeds the
Lena River water discharge (QL/QA < 0.75). Mean values of SSC were calculated for 16
profiles (Figure 2b) on the left Sl and right Sr segments of the river channel for each avail-
able satellite image. The QL/QA ratio was then compared with the Sl/Sr ratio. Sl/Sr changes
along the Lena River case study 2 from the Lena and Aldan confluence (Figure 12) depicts
that a greater difference in water discharges between the rivers was associated with higher
sediment plumes along particular banks.
Table 7. The ratio of water discharge in the Lena River (QL) to the flow in the Aldan River (QA).
Satellite
Image Date
QL,
m3s−1
QA,
m3s−1
QL/QA
Discharge
Ratio Group
Satellite Im-
age Date
QL,
m3s−1
QA,
m3s−1
QL/QA
Discharge
Ratio Group
28 August
2009
17,700
5010
3.53
1
04 August
2006
6850
5760
1.19
3
30 August
2001
9500
4460
2.13
1
09 July 2014
11,700
10,400
1.13
3
30 Septem-
ber 2015
7620
3920
1.94
1
28 Septem-
ber 2000
9890
9180
1.08
3
03 Septem-
ber 2011
9590
5590
1.72
2
08 Septem-
ber 2007
8740
8220
1.06
3
12 June 2015
17,200
10,100
1.70
2
19 Septem-
ber 2011
10,800
10,300
1.05
3
02 August
2011
7880
4660
1.69
2
15 August
2016
12,900
13,100
0.98
3
14 June 2002
22,800
14,100
1,62
2
04 June 2013
20,700
22,000
0.94
3
01 August
2002
8770
5540
1.58
2
07 August
2013
22,800
26,000
0.88
3
12 June 2010
19,700
14,100
1.40
2
23 August
2013
15,900
18,400
0.86
3
12 June 2016
31,400
23,100
1.36
2
20 June 2013
18,200
21,600
0.84
3
10 Septem-
ber 1999
8650
6430
1.35
2
22 June 2013
14,100
18,800
0.75
4
08 June 2000
26,400
20,400
1.29
2
31 August
2016
9730
13,200
0.74
4
10 June 2000
14,100
21,100
0.67
4
Water 2022, 14, 3845 23 of 27
Figure 12. Changes in Sl/Sr ratio along the Aldan–Viluy reach of the Lena River under various hy-
drological conditions QL/QA.
Along case study 3, a downward increase in surface sediment concentration (ΔS > 0)
was associated with thermal erosion processes. The satellite images revealed a significant
increase in sediment concentrations along the left bank. Despite the fact that bank retreat
is governed by water discharge (duration and season) [16,17], we were unable to capture
a significant relationship between sediment concentration changes and water flow rates.
This may be explained by the non-linear links between water discharge, bank retreat and
sediment concentration rates, e.g., for the middle Lena Gautier et al. [16] found a robust
correlation between island morphological changes and several hydrologic parameters, in-
cluding duration of bankfull discharges in May, the number of flood peaks in August;
number of peaks; and number total peaks. These parameters were not used in our study
which could lead to the conclusions related to the hydrological drivers of sedimentation
and erosion patterns. At the same time, we found the relationships between average daily
air temperature and bank erosion (SSC change). More specifically, average daily air tem-
perature (according to the Tiksi weather station) was linked with the suspended sediment
changes along the Bykovskaya and Trofimovskaya distributaries. This was seen especially
in the range of air temperatures from 5 °C to 15 °C (Figure 13). The influence of a positive
influx in total solar radiation and an increase in air temperatures contributes to the deg-
radation of permafrost rocks and the activation of thermal erosion. Summer thawing and
the associated activation of bank erosion raise sediment delivery into the hydrological
network and further increase the concentration of solids in the river. Such effects are
mostly marked along the left riverbanks of the southern exposure of the sublatitudinal
sections of the Bykovskaya and Trofimovskaya channels.
Water 2022, 14, 3845 24 of 27
Figure 13. Relationship between the surface suspended sediment changes (ΔS) and air temperature
(0C) (Tiksi water station) in the Lena delta.
The obtained results imply a few important topics for further research and practice.
They are useful for recalibrating the total export–discharge sediment models as far as the
relationships between SSC and river discharge change owing to the shifting hydrological
regimes and weather, thawing permafrost [58], and/or increasing terrestrial productivity.
The obtained set of models (Table 2) can be used to develop large-scale and systematic
hydrological monitoring systems in the Lena basin. Along the middle course, the more
pronounced is the sediment exchanges between the channel systems and the floodplains
which are important not just from a hydro-geomorphologic point of view, but also as a
driver of sediment quality and geochemistry [59]. The results are critically important to
our understanding of the Arctic carbon and nutrient cycles. Further, sediment-related re-
search over the largest Arctic rivers can be combined with hydro-geochemical studies,
also based on satellite-derived estimates already existing for Kolyma [60] and other large
rivers [61]. The observed patterns have a few important and contrasting biogeochemical
impacts on the riverine terrestrial flux of both mineral and organic components.
6. Conclusions
An effective combination of Landsat images with in situ measurements by gravimet-
ric, turbidity and ADCP approaches were utilized to monitor the sediment transport along
the anabranching channel along the middle and lower Lena River. Based on the set of
empirical models, we revealed 3 transverse erosion and sedimentation patterns along the
middle and lower Lena River anabranching channels
1. In the middle reach (case study 1) the anabranching channel was characterized by
seasonal effects in erosion and sedimentation. Over 50% of the suspended load is
trapped under peak flow conditions. A 10–20% increase in the suspended load oc-
curred during low-flow conditions due to impacts of thermal erosion. Sediment traps
along the anabranching channel exist within secondary branches (100–200 m3s−1).
Sediment concentration changes due to variability of stream flow parameters (veloc-
ity). We found a significant relationship between the velocity (transport capacity) and
sediment concentration and conclude that hydrodynamic controls sediment sorting
within anabranching reach.
2. Anabranching channel reach which is influenced by the largest tributaries (Aldan
and Viluy) (case study 2) is characterized by the sediment plumes which dominated
spatial and temporal SSC distribution along the river section. A greater difference in
water discharges between merging rivers is associated with higher sediment plumes
along particular banks. Under these conditions, seasonal effects of sedimentation and
erosion patterns cannot be detected.
Water 2022, 14, 3845 25 of 27
3. The Lena delta (case study 3) demonstrates contrasting sediment patterns over its
various parts. The observed changes in the suspended sediment transport in the delta
are mostly related to channel erosion driven by thermal erosional processes. Here,
the weather conditions during the summer affect the sediment budget along Yedoma
banks which was reflected by the positive relationship observed between the daily
air temperature and surface suspended sediment changes.
Author Contributions: Conceptualization, writing—original draft preparation, S.C.; methodology,
validation, visualization, K.P. All authors have read and agreed to the published version of the man-
uscript.
Funding: Field works, study concept and numerical simulations are done within implementation
of Russian Scientific Foundation project 21-17-00181. Additionally, study has been supported by the
Kazan Federal University Strategic Academic Leadership Program ("PRIORITY-2030").
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: The Authors would like to thank the USGS for Landsat data, Russian Federal
Service for Hydrometeorology and Environmental Monitoring (Roshydromet) and the Arctic Great
Rivers Observatory (ArcticGRO) for the valuable measurement data.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Peterson, B.J.; Holmes, R.M.; McClelland, J.W.; Vörösmarty, C.J.; Lammers, R.B.; Shiklomanov, A.I.; Shiklomanov, I.A.;
Rahmstorf, S. Increasing River Discharge to the Arctic Ocean. Science 2002, 298, 2171–2173. https://doi.org/10.1126/sci-
ence.1077445.
2. Alexeevsky, N.I.; Chalov, R.S.; Berkovich, K.M.; Chalov, S.R. Channel changes in largest Russian rivers: Natural and anthropo-
genic effects. Int. J. River Basin Manag. 2013, 11, 175–191.
3. Chalov, S.R.; Liu, S.; Chalov, R.S.; Chalova, E.R.; Chernov, A.V.; Promakhova, E.V.; Berkovitch, K.M.; Chalova, A.S.; Zavadsky,
A.S.; Mikhailova, N. Environmental and human impacts on sediment transport of the largest Asian rivers of Russia and China.
Environ. Earth Sci. 2018, 77, 274. https://doi.org/10.1007/s12665-018-7448-9.
4. Escoube, R.; Rouxel, O.J.; Pokrovsky, O.S.; Schroth, A.; Max Holmes, R.; Donard, O.F.X. Iron isotope systematics in Arctic rivers.
Comptes Rendus -Geosci. 2015, 347, 377–385. https://doi.org/10.1016/j.crte.2015.04.005.
5. Rachold, V.; Alabyan, A.M.; Hubberten, H.W.; Korotaev, V.N.; Zaitsev, A.A. Sediment transport to the Laptev Sea—Hydrology
and geochemistry of the Lena River. Polar Res. 1996, 15, 183–196. https://doi.org/10.3402/polar.v15i2.6646.
6. Pietroń, J.; Nittrouer, J.A.; Chalov, S.R.; Dong, T.Y.; Kasimov, N.; Shinkareva, G.; Jarsjö, J. Sedimentation patterns in the Selenga
River delta under changing hydroclimatic conditions. Hydrol. Process. 2018, 32, 278–292. https://doi.org/10.1002/hyp.11414.
7. Magritsky, D.V. Annual Sediment Yield of Russian Artctic Rivers and Its Anthropogenic Changes; Vestnik Moskovskogo Universiteta
Series 5 Geography; Moskovskij Universitet: Moscow, Russia, 2010; pp. 17–24. (In Russian)
8. Hasholt, B.; Bobrovitskaya, N.; Bogen, J.; McNamara, J.; Mernild, S.H.; Milburn, D.; Walling, D.E. Sediment transport to the
Arctic Ocean and adjoining cold oceans. Hydrol. Res. 2006, 37, 413–432.
9. Syvitski, J.P.M.; Kettner, A. Sediment flux and the anthropocene. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 2011, 369, 957–975.
https://doi.org/10.1098/rsta.2010.0329.
10. Gautier, E.; Costard, F. Anastomosing-fluvial systems in the periglacial zone: The Lena River and its main tributaries (Central
Siberia). Geogr. Phys. Quat. 2000, 54, 327–342.
11. Golovlyov, P.; Kornilova, E.; Krylenko, I.; Belikov, V.; Zavadskii, A.; Fingert, E.; Borisova, N.; Morozova, E. Numerical modeling
and forecast of channel changes on the river Lena near city Yakutsk. Proc. Int. Assoc. Hydrol. Sci. 2019, 381, 65–71.
https://doi.org/10.5194/piahs-381-65-2019. (In Russian)
12. Bolshiyanov, D.Y.; Makarov, A.S.; Schneider, V.; Stoof, G. Origin and Development of the Lena River Delta; AARI: St. Petersburg,
Russia, 2013. (In Russian)
13. Chalov, S.; Prokopeva, K.; Habel, M. North to South Variations in the Suspended Sediment Transport Budget within Large
Siberian River Deltas Revealed by Remote Sensing Data. Remote Sens. 2021, 13, 4549. https://doi.org/10.3390/rs13224549.
14. Fedorova, I.; Chetverova, A.; Bolshiyanov, D.; Makarov, A.; Boike, J.; Heim, B.; Morgenstern, A.; Overduin, P.P.; Wegner, C.;
Kashina, V.; et al. Lena Delta hydrology and geochemistry: Long-term hydrological data and recent field observations. Biogeo-
sciences 2015, 12, 345–363. https://doi.org/10.5194/bg-12-345-2015.
Water 2022, 14, 3845 26 of 27
15. Agafonova, S.A.; Frolova, N.A.; Surkova, G.V.; Koltermann, K.P. Modern characteristics of the ice regime of russian arctic rivers
and their possible changes in the 21st centure. Geogr. Environ. Sustain. 2017, 10, 4–15. https://doi.org/10.24057/2071-9388-2017-
10-4-4-15.
16. Georgiadi, A.G.; Koronkevich, N.I.; Milyukova, I.P.; Barabanova, E.A. Integrated projection for runoff changes in large Russian
river basins in the XXI century. Geogr. Environ. Sustain. 2016, 9, 38–46. https://doi.org/10.24057/2071-9388-2016-9-2-59-65.
17. Lammers, R.B.; Shiklomanov, A.I.; Vörösmarty, C.J.; Fekete, B.M.; Peterson, B.J. Assessment of contemporary Arctic river runoff
based on observational discharge records. J. Geophys. Res. Atmos. 2001, 106, 3321–3334. https://doi.org/10.1029/2000JD900444.
18. Costard, F.; Gautier, E.; Brunstein, D.; Hammadi, J.; Fedorov, A.; Yang, D.; Dupeyrat, L. Impact of the global warming on the
fluvial thermal erosion over the Lena River in Central Siberia. Geophys. Res. Lett. 2007, 34, L14501.
https://doi.org/10.1029/2007GL030212.
19. Gautier, E.; Dépret, T.; Cavero, J.; Costard, F.; Virmoux, C.; Fedorov, A.; Konstantinov, P.; Jammet, M.; Brunstein, D. Fifty-year
dynamics of the Lena River islands (Russia): Spatio-temporal pattern of large periglacial anabranching river and influence of
climate change. Sci. Total Environ. 2021, 783, 147020. https://doi.org/10.1016/j.scitotenv.2021.147020.
20. Tananaev, N.I. Hydrological and sedimentary controls over fluvial thermal erosion, the Lena River, central Yakutia. Geomor-
phology 2016, 253, 524–533. https://doi.org/10.1016/j.geomorph.2015.11.009.
21. Chalov, S.; Golosov, V.; Tsyplenkov, A.; Theuring, P.; Zakerinejad, R.; Märker, M.; Samokhin, M. A toolbox for sediment budget
research in small catchments. Geogr. Environ. Sustain. 2017, 10, 43–68. https://doi.org/10.24057/2071-9388-2017-10-4-43-68.
22. Chalov, S.R.; Bazilova, V.O.; Tarasov, M.K. Suspended sediment balance in Selenga delta at the late XX–early XXI century:
Simulation by LANDSAT satellite images. Water Resour. 2017, 44, 463–470. https://doi.org/10.1134/S0097807817030071.
23. Pavelsky, T.M.; Smith, L.C. Remote sensing of suspended sediment concentration, flow velocity, and lake recharge in the Peace-
Athabasca Delta, Canada. Water Resour. Res. 2009, 45, W11417. https://doi.org/10.1029/2008WR007424.
24. Martinez, J.M.; Guyot, J.L.; Filizola, N.; Sondag, F. Increase in suspended sediment discharge of the Amazon River assessed by
monitoring network and satellite data. Catena 2009, 79, 257–264. https://doi.org/10.1016/j.catena.2009.05.011.
25. Wang, J.; Lu, X.; Zhou, Y. Retrieval of suspended sediment concentrations in the turbid water of the Upper Yangtze River using
Landsat ETM+. Chin. Sci. Bull. 2007, 52, 273–280. https://doi.org/10.1007/s11434-007-7012-6.
26. Park, E.; Latrubesse, E.M. Modeling suspended sediment distribution patterns of the Amazon River using MODIS data. Remote
Sens. Environ. 2014, 147, 232–242. https://doi.org/10.1016/j.rse.2014.03.013.
27. Park, E.; Latrubesse, E.M. A geomorphological assessment of wash-load sediment fluxes and floodplain sediment sinks along
the lower Amazon River. Geology 2019, 47, 403–406. https://doi.org/10.1130/G45769.1.
28. Latrubesse, E.M.; Park, E.; Kastner, K. The Ayeyarwady River (Myanmar): Washload transport and its global role among rivers
in the Anthropocene. PLoS ONE 2021, 16, e0251156. https://doi.org/10.1371/journal.pone.0251156.
29. Mertes, L.A.K.; Smith, M.O.; Adams, J.B. Estimating suspended sediment concentrations in surface waters of the Amazon River
wetlands from Landsat images. Remote Sens. Environ. 1993, 43, 281–301. https://doi.org/10.1016/0034-4257(93)90071-5.
30. Yang, D.; Kane, D.L.; Hinzman, L.D.; Zhang, X.; Zhang, T.; Ye, H. Siberian Lena River hydrologic regime and recent change. J.
Geophys. Res. Atmos. 2002, 107, 4694. https://doi.org/10.1029/2002JD002542.
31. Babin, M.; Stramski, D.; Ferrari, G.M.; Claustre, H.; Bricaud, A.; Obolensky, G.; Hoepffner, N. Variations in the light absorption
coefficients of phytoplankton, nonalgal particles, and dissolved organic matter in coastal waters around Europe. J. Geophys. Res.
Ocean. 2003, 108. https://doi.org/10.1029/2001jc000882.
32. Strauss, J.; Laboor, S.; Schirrmeister, L.; Fedorov, A.N.; Fortier, D.; Froese, D.; Fuchs, M.; Günther, F.; Grigoriev, M.; Harden, J.;
et al. Circum-Arctic Map of the Yedoma Permafrost Domain. Front. Earth Sci. 2021, 9, 758360.
https://doi.org/10.3389/feart.2021.758360.
33. Tananaev, N.I. Seasonal and long-term freezing of the cryolithozone river beds and its influence on channel changes (on the
example of average Lena). Eros. Fluv. Process. Hydroecol. Probl. 2004, 195–201. (In Russian)
34. Chalov, R.S. Fluvial Processes: Theory and Applications; Springer International Publishing: Cham, Switzerland, 2021; ISBN 978-3-
030-66182-3.
35. Schwamborn, G.; Rachold, V.; Grigoriev, M.N. Late Quaternary sedimentation history of the Lena Delta. Quat. Int. 2002, 89,
119–134. https://doi.org/10.1016/S1040-6182(01)00084-2.
36. Latrubesse, E.M. Patterns of anabranching channels: The ultimate end-member adjustment of mega rivers. Geomorphology 2008,
101, 130–145. https://doi.org/10.1016/j.geomorph.2008.05.035.
37. Ivanov, V.V.; Piskun, A.A.; Korabel, R.A. Distribution of Runoff through the Main Channels of the Lena River Delta. Proc. AARI
1983, 378, 59–71. (In Russian)
38. Chalov, S.; Moreido, V.; Ivanov, V.; Chalova, A. Assessing suspended sediment fluxes with acoustic Doppler current profilers:
Case study from large rivers in Russia. Big Earth Data 2022, 1–23. https://doi.org/10.1080/20964471.2022.2116834.
39. Dominguez Ruben, L.G.; Szupiany, R.N.; Latosinski, F.G.; López Weibel, C.; Wood, M.; Boldt, J. Acoustic Sediment Estimation
Toolbox (ASET): A software package for calibrating and processing TRDI ADCP data to compute suspended-sediment
transport in sandy rivers. Comput. Geosci. 2020, 140, 104499. https://doi.org/10.1016/j.cageo.2020.104499.
40. Carlson, P.R.; Conomos, T.J.; Janda, R.J.; Peterson, D.H. Principal Sources and Dispersal Patterns-of Suspended Particulate Matter in
Nearshore Surface Waters of the Northeast Pacific Ocean; Final Report; EROS Center: Sioux Falls, SD, USA, 1974.
41. Curran, P.J.; Novo, E.M.M. The relationship between suspended sediment concentration and remotely sensed spectral radiance.
Rev. J. Coast. Res. 1988, 361–368.
Water 2022, 14, 3845 27 of 27
42. Wackerman, C.; Hayden, A.; Jonik, J. Deriving spatial and temporal context for point measurements of suspended-sediment
concentration using remote-sensing imagery in the Mekong Delta. Cont. Shelf Res. 2017, 147, 231–245.
https://doi.org/10.1016/j.csr.2017.08.007.
43. Novo, E.M.M.; Hansom, J.D.; Curran, P.J. The effect of sediment type on the relationship between reflectance and suspended
sediment concentration. Int. J. Remote Sens. 1989, 10, 1283–1289. https://doi.org/10.1080/01431168908903967.
44. Chavez, P.S. Image-Based Atmospheric Corrections—Revisited and Improved. Photogramm. Eng. Remote Sens. 1996, 62, 1025–
1036.
45. Chavez, P.S. Radiometric calibration of Landsat Thematic Mapper multispectral images. Photogramm. Eng. Remote Sens. 1989,
55, 1285–1294.
46. Chander, G.; Markham, B.L.; Helder, D.L. Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+,
and EO-1 ALI sensors. Remote Sens. Environ. 2009, 113, 893–903. https://doi.org/10.1016/j.rse.2009.01.007.
47. Chavez, P.S. An improved dark-object subtraction technique for atmospheric scattering correction of multispectral data. Remote
Sens. Environ. 1988, 24, 459–479. https://doi.org/10.1016/0034-4257(88)90019-3.
48. Sokolov, D.I.; Erina, O.N.; Tereshina, M.A.; Puklakov, V.V. Impact of Mozhaysk Dam on the Moscow River Sediment Transport.
Geogr. Environ. Sustain. 2020, 13, 24–31. https://doi.org/10.24057/2071-9388-2019-150.
49. Ankcorn, P.D. Clarifying Turbidity—The Potential and Limitations of Turbidity as a Surrogate for Water-Quality Monitoring.
In Proceedings of the 2003 Georgia Water Resources Conference, Tifton, GA, USA, 23–24 April 2003.
50. Belozerova, E. V.; Chalov, S.R. Assessment of River Water Turbidity Using the Optic Methods; Vestnik Moskovskogo Universiteta
Seriya 5 Geografiya; Moskovskij Universitet: Moscow, Russia, 2013; pp. 39–45. (In Russian)
51. Szupiany, R.N.; Lopez Weibel, C.; Guerrero, M.; Latosinski, F.; Wood, M.; Dominguez Ruben, L.; Oberg, K. Estimating sand
concentrations using ADCP-based acoustic inversion in a large fluvial system characterized by bi-modal suspended-sediment
distributions. Earth Surf. Process. Landf. 2019, 44, 1295–1308. https://doi.org/10.1002/esp.4572.
52. Choubey, V.K. Laboratory experiment, field and remotely sensed data analysis for the assessment of suspended solids concen-
tration and secchi depth of the reservoir surface water. Int. J. Remote Sens. 1998, 19, 3349–3360.
https://doi.org/10.1080/014311698214037.
53. Binh, D. Van; Kantoush, S.; Sumi, T. Changes to long-term discharge and sediment loads in the Vietnamese Mekong Delta
caused by upstream dams. Geomorphology 2020, 353, 107011. https://doi.org/10.1016/j.geomorph.2019.107011.
54. Martynov, A.V. Influence Of The Large Flood on the Element Composition of Fluvisols in the Amur River Valley. Geogr. Environ.
Sustain. 2020, 13, 52–64. https://doi.org/10.24057/2071-9388-2019-03.
55. Wang, J.J.; Lu, X.X.; Liew, S.C.; Zhou, Y. Retrieval of suspended sediment concentrations in large turbid rivers using Landsat
ETM+: An example from the Yangtze River, China. Earth Surf. Process. Landf. 2009, 34, 1082–1092.
https://doi.org/10.1002/esp.1795.
56. Bonnet, M.P.; Barroux, G.; Martinez, J.M.; Seyler, F.; Moreira-Turcq, P.; Cochonneau, G.; Melack, J.M.; Boaventura, G.; Maurice-
Bourgoin, L.; León, J.G.; et al. Floodplain hydrology in an Amazon floodplain lake (Lago Grande de Curuaí). J. Hydrol. 2008,
349, 18–30. https://doi.org/10.1016/j.jhydrol.2007.10.055.
57. Chalov, S.R.; Alexeevsky, N.I. Braided rivers: Structure, types and hydrological effects. Hydrol. Res. 2015, 46, 258.
https://doi.org/10.2166/nh.2013.023.
58. Vandenberghe, J. Climate forcing of fluvial system development: An evolution of ideas. Quat. Sci. Rev. 2003, 22, 2053–2060.
https://doi.org/10.1016/S0277-3791(03)00213-0.
59. Gordeev, V.V.; Rachold, V.; Vlasova, I.E. Geochemical behaviour of major and trace elements in suspended particulate material
of the Irtysh river, the main tributary of the Ob river, Siberia. Appl. Geochem. 2004, 19, 593–610. https://doi.org/10.1016/j.apgeo-
chem.2003.08.004.
60. Griffin, C.G.; Frey, K.E.; Rogan, J.; Holmes, R.M. Spatial and interannual variability of dissolved organic matter in the Kolyma
River, East Siberia, observed using satellite imagery. J. Geophys. Res. 2011, 116, G03018. https://doi.org/10.1029/2010JG001634.
61. McClelland, J.W.; Holmes, R.M.; Peterson, B.J.; Raymond, P.A.; Striegl, R.G.; Zhulidov, A.V.; Zimov, S.A.; Zimov, N.; Tank, S.E.;
Spencer, R.G.M.; et al. Particulate organic carbon and nitrogen export from major Arctic rivers. Glob. Biogeochem. Cycles 2016, 30,
629–643. https://doi.org/10.1002/2015GB005351.
