Conference PaperPDF Available

Development of Remote Sensing Based Index for Estimating/Mapping Suspended Sediment Concentration in River and Lake Environments

Authors:

Abstract and Figures

This research explored the potential of remote sensing to develop an index and estimate co- efficients that can be used in riverine/lake environm ents, especially during ex treme events when routine in-situ measurements are not available. Normalized Difference Suspended Sediment Index (NDSSI) was calculated using the Landsat data and was correlated to the near real-time in-situ measurements of suspended sediment concentrations using a power equation for quantitative estimation of SS concentration in the Mississippi River. This technique, using the obtained coefficients was applied to estimate/map the SS concentration in the Mississippi River during the Mid West USA 2008 flood and in Lake Pontchartrain during (1) Bonnet Carre Spill Way opening event and (2) before and after Hurricane Katrina. The results were compared by the simula tion results of CCHE2D (a numerical model developed at NCCHE) and found in a good general agreement qualitatively and quantitatively. The preliminary results indicate that (1) NDSSI has the potential to estimate (relative variation) and map the spatial distribution of SS concentration in both river and lake environments, (2) NDSSI can be used for quantitative estimation of SS concentration in these e nvironments when coupled with two co-efficients in a power equation, and (3) the same approach can be used to estimate SS concentration in both river and lake water within reasonable error limits using NDSS I. Acquisition of more in-situ measurements of SS concentrations are on going to derive more general co -efficients and achieve more validation results.
Content may be subject to copyright.
Development of Remote Sensing Based Index for Estimating/Mapping
Suspended Sediment Concentration in River and Lake Environments
A. K. M. Azad Hossain
Post Doctoral Research Associate
National Center for Computational Hydroscience and Engineering
The University of Mississippi, 102 Carrier Hall, University, MS, 38677, USA
Xiaobo Chao
Research Scientist
National Center for Computational Hydroscience and Engineering
The University of Mississippi, 102 Carrier Hall, University, MS, 38677, USA
Yafei Jia
Research Professor
National Center for Computational Hydroscience and Engineering
The University of Mississippi, 102 Carrier Hall, University, MS, 38677, USA
Abstract: This research explored the potential of remote sensing to develop an index and estimate co-
efficients that can be used in riverine/lake environments, especially during extreme events when routine
in-situ measurements are not available. Normalized Difference Suspended Sediment Index (NDSSI) was
calculated using the Landsat data and was correlated to the near real-time in-situ measurements of
suspended sediment concentrations using a power equation for quantitative estimation of SS
concentration in the Mississippi River. This technique, using the obtained coefficients was applied to
estimate/map the SS concentration in the Mississippi River during the Mid West USA 2008 flood and in
Lake Pontchartrain during (1) Bonnet Carre Spill Way opening event and (2) before and after Hurricane
Katrina. The results were compared by the simulation results of CCHE2D (a numerical model developed
at NCCHE) and found in a good general agreement qualitatively and quantitatively. The preliminary
results indicate that (1) NDSSI has the potential to estimate (relative variation) and map the spatial
distribution of SS concentration in both river and lake environments, (2) NDSSI can be used for
quantitative estimation of SS concentration in these environments when coupled with two co-efficients in
a power equation, and (3) the same approach can be used to estimate SS concentration in both river and
lake water within reasonable error limits using NDSSI. Acquisition of more in-situ measurements of SS
concentrations are on going to derive more general co-efficients and achieve more validation results.
Keywords: Remote Sensing, NDSSI, Suspended Sediment Concentration, CCHE2D, Mississippi River,
Lake Pontchartrain, Hurricane Katrina.
Introduction
Water quality is one of the most important factors for lake and riverine ecosystems. Sediment
concentration in the water is considered to be a critical water quality parameter that affects the lake and
river habitats negatively. Typically suspended sediments (SS) are the non-dissolved matters in the water
that reflects the physical and chemical property of water. They Influence the total primary productivity as
the quantity of SS affects the transmission of light in water, as well as the transition of heavy metal and
the micro-pollutant. Suspended sediment concentration is a spatially inhomogeneous parameter and its
spatial distribution is difficult to measure with the routine in-situ monitoring method. Numerical
modeling has been used to estimate SS concentration in river channels, over their flood plains and other
surface waters. Remote sensing is an efficient method, which can provide realistic water quality data
with large spatial distributions for water resource study. In this study, remote sensing techniques are used
for mapping SS concentration in the Mississippi River and the Lake Pontchartrain, LA. A normalized
difference suspended sediment index (NDSSI) was developed for this purpose. The study finds this as an
effective approach, the estimated SSC has reasonable values and special distributions. The estimated
results and the numerical model prediction agreed very well.
Numerical Model vs. Remote Sensing
Numerical models are based on hydrodynamic principles, methodologies and algorithms. The results are
predicted continuous data with short time steps usually validated by physical experiments and/or limited
in-situ measurements. Remote sensing techniques are based on correlation between near real-time
observations of spectral reflectance and discrete in-situ measurements. The results are compared with
observed continuous data with large temporal gap also validated by physical experiments and/or limited
in-situ measurements. During extreme events like hurricane and flood usually no ground measurements of
sediment concentration are done in the waters of inundated areas. Therefore, satellite data derived SSC
data serves as the only means to provide observed sediment concentration information during natural
disasters.
Estimation of SS Concentration Using Remote Sensing
Suspended sediment concentration has been estimated and mapped successfully using remote sensing for
the last three decades. Different approaches and algorithms had been developed over time for SS
concentration estimation/mapping using optical satellite data. The available techniques can be categorized
in four general groups: (1) simple regression (correlation between single band and in-situ
measurements)[e.g., Williams and Grabau (1973) – Chesapeake Bay early in 1973], (2) spectral unmixing
techniques [e.g., Gomez, et al., 1997], (3) Band ratio technique using two and more bands [e.g.,
Lathrop ,1992; Populus et al., 1995; Wang et al., 2000], and (4) multiple regressions (using multiple
bands and in-situ measurements)[e.g., Binding et al., 2005].
Usually when suspended sediment concentrations are high, the backscatter/ reflectivity of water is high.
There are three matters dominate the reflectance of inland water, which are yellow substance, suspended
sediment and phytoplankton. Yellow substance is a soluble matter, which has no scatter capability, but it
has a strong absorption effect on short-wave bands that highly reduce the underwater downwelling
irradiance. Therefore, when the absorption of water itself and yellow substance is small, actual suspended
sediment information could be obtained (Wang et al., 2003).
Goal and Objectives
It has been observed that although remote sensing has been considered as a proven technique for SS
concentration estimation all the developed models and algorithms are applicable for specific areas and
environments. Due to this reason it is not possible to use any existing remote sensing based technique to
estimate and map SS concentration during extreme events like hurricane, flush flood etc. To address this
issue in this research we attempted to explore the potential of remote sensing to develop an index and
estimate co-efficients that can be used in different riverine/lake environments, especially during extreme
events when routine in-situ measurements are not available.
Study Site
This research involves three different sites for model development, calibration, validation and
applications. The sites are located in the Midwest and Southeast USA. Selected parts of the Mississippi
River along the western part of Missouri were used for the model development and calibration (Figure 1).
Mississippi River and the adjacent areas between Alexandria, MO and Warsaw, IL areas and Lake
Pontchartrain in Louisiana were used for model application and validation.
A
B
C
A
B
C
A: Model development/calibration
B & C: Model application / validation
Figure 1. Location and extent of study sites (A and C shown on Landsat 5 TM 4,3,2 false color composite
imagery; and B shown on ALOS AVNIR 2 4,3,2 false color composite imagery)
Materials and Methods
Inspired by the concept of Normalized Difference Vegetation Index (NDVI) [Rouse, et al., 1976] we
calculated Normalized Difference Suspended Sediment Index (NDSSI) using Landsat 7 ETM+ imagery to
determine spatial distribution of the relative variation of SS concentration in the river/lake water. We
correlated the NDSSI values to near real time in-situ measurements of SS concentration to estimate the
SS concentration quantitatively.
Normalized Difference Suspended Sediment Index (NDSSI)
It has been observed for Landsat TM/ETM imagery that Band 1 (Blue band/~ 0.450-0.515 microns) and
Band 4 (Near-infrared/~0.750-0.900 microns) are most sensitive to water and water transparency
(turbidity). Band 1and Band 4 usually gives the highest and lowest reflectance values respectively for
water. These characteristics have been observed for water with different levels of turbidity (Figure 2).
NDVI was calculated on the basis of the responses of Band 3 (Red band) and Band 4 (Near infra-red
band) on green vegetation. According to Rouse et al., (1976) for any vegetation Band 4 and band 3 of
Landsat 5/7 TM/ETM+ data always gives the highest and lowest reflectance respectively. Equation 1
shows the calculation of NDVI. The values of NDVI range from -1 to +1 where higher values indicate the
presence of more green vegetation and lower values indicate stressed vegetation or bare soil/concrete etc.
To achieve the capability of similar data interpretability we calculated NDSSI as shown in Equation 2.
Like NDVI the values of NDSSI also range from -1 to +1 where higher values indicate the presence of
more clear water and lower values indicate the presence of more turbid water or land.
0
10
20
30
40
50
60
70
80
90
1234
DNValues(Scaledreflectance%)
Landsat5TMVNIRBands
p ()
Very highconcentrationofSS
HighconcentrationofSS
ModerateconcentrationofSS
LowconcentrationofSS
Very lowconcentration ofSS
Locationsofspectralprofiles
Figure 2. Spectral profiles of water at different levels of suspended sediment (SS) concentrations
observed in Landsat 5 TM VNIR imagery
RNIR
RNIR
NDVI
ρρ
ρ
ρ
+
=
(1)
NIRB
NIRB
NDSSI
ρρ
ρρ
+
=
(2)
Where, ρ
B,
ρ
R,
and
ρ
NIR,
are the reflectance values of Landsat 5/7 TM/ETM+ Band 1, Band 3 and
Band 4 respectively.
Figure 3 shows the nature of NDSSI imagery prepared from Landsat 5 TM VNIR imagery acquired over
Lake Pontchartrain at different periods. These imagery captured the snapshots of different levels of SS
concentrations in the water.
Data Used
Landsat 7 ETM+ VNIR imagery acquired over the Site ‘A (Fig. 1) for 16 dates were obtained from
USGS/NASA (from GLOVIS ) to calculate and calibrate the NDSSI imagery. Total 16 Landsat 7 ETM+
imagery were used in this purpose. Each image dates include two scenes. Table 1 shows the image
acquisition dates. Near real time (of the corresponding image acquisition dates) in-situ measurements of
SS concentrations in the Mississippi River were obtained from 3 USGS stations. TheIDs of these stations
are 7010000, 7020500 and 7022000. Table 1 shows the measurements of SS concentrations for each date.
One scene of ALOS AVNIR2 imagery acquired over the Site ‘B was obtained from ASF to estimate/map
SS concentration in the flood water (during Midwest Flood event in June 20080 using NDSSI. Three
scenes of Landsat 5 TM VNIR imagery acquired over the Site ‘C’ were obtained from USGS/ NASA
(from GLOVIS ) to estimate/map SS concentrations (1) in the water of Lake Pontchartrain during the
Bonnet Care Spillway flooding event, and (2) study the changes occurred in the SS concentrations in the
water of lake Pontchartrain due to Hurricane Katrina. The imagery used in this purpose were acquired on
April 10, 1997; August 22, 2005 and September 07, 2005.
April 10, 1997 August 22, 2005
April 10, 1997 August 22, 2005
September 07, 2005
September 07, 2005
Landsat 5 TM VNIR/True Color
Corresponding NDSSI Imagery
Figure 3. NDSSI imagery for different levels of suspended sediment (SS) concentrations in Lake
Pontchartrain as observed in Landsat 5 TM imagery
Table 1. Image acquisition dates, in-situ measurements of SS concentrations and corresponding values of
NDSSI obtained from the acquired Landsat 7 ETM+ imagery.
USGS Stations
Landsat 7 ETM +
Data Acquisition
Sediment Concentration
(mg/lit)
NDSSI
7010000 5/12/2003 888 0.35
4/26/2003 132 0.58
4/10/2003 115 0.64
10/16/2002 114 0.61
7020500 5/12/2003 765 0.37
4/26/2003 227 0.52
7022000 5/12/2003 591 0.38
4/26/2003 337 0.49
10/16/2002 96 0.62
9/30/2002 97 0.60
7/28/2002 78 0.55
10/29/2001 156 0.44
9/27/2001 317 0.45
9/11/2001 191 0.58
10/10/2000 151 0.47
10/24/1999 119 0.47
Calibration of NDSSI
NDSSI was calculated for each Landsat data and correlated with the acquired USGS SSC measurements.
It was noticed that all three USGS stations are located on top of a bridge over the Mississippi River.
Therefore, it was not possible to get the NDSSI value exactly at the same location of the in-situ SS
concentration measurements. NDSSI values were obtained from the closest 25 pixels downstream the in-
situ measurement stations and an average value was calculated to represent the NDSSI value for the
corresponding measuring station. The obtained NDSSI value for each station is shown in Table 1. Figure
4 shows the locations of the USGS stations and the corresponding 25 pixels of NDSSI calculation for
each station.
(a)
(b)
(c)
(d)
Figure 4. NDSSI calculation for each in-situ SS concentration measuring stations
To determine the most suitable coefficients to estimate the SS concentrations in the river/lake water using
NDSSI, the obtained NDSSI values were plotted against the corresponding near real time in-situ
measurements. The relationship between the correlated NDSSI and in-situ measurements were interpreted
using different numerical equations including linear, exponential, logarithmic, polynomial and power
function. Figure 5 shows the comparisons among the plots.
y = -2196.9x + 1386
R² = 0.6662
0
100
200
300
400
500
600
700
800
900
1000
0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
Sediment concentration (mg/lit)
NDSSI
y = 6386.2e
-6.839x
R² = 0.6994
0
100
200
300
400
500
600
700
800
900
1000
0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
Sediment concentration (mg/lit)
NDSSI
y = -1112ln(x) - 502.36
R² = 0.7296
0
100
200
300
400
500
600
700
800
900
1000
0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
Sediment concentration (mg/lit)
NDSSI
y = 16064x
2
- 18126x + 5200
R² = 0.8896
0
100
200
300
400
500
600
700
800
900
1000
0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
Sediment concentration (mg/lit)
NDSSI
y = 18.69x
-3.399
R² = 0.738
0
100
200
300
400
500
600
700
800
900
1000
0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
Sediment concentration (mg/lit)
NDSSI
A
B C
D E
A: Linear relationship
B: Exponential relationship
C: Logarithmic relationship
D: Polynomial relationship
E: Power relationship
Figure 5. Relationship between NDSSI and in-situ measurements of SS concentrations in water
The coefficients associated with the equation that achieved highest correlation coefficient (R
2
) value were
considered suitable to use for SS concentration estimation in the lake/river water. Accordingly the
polynomial (2
nd
order) equation was found to have a R
2
value of about 0.89 and considered the most
suitable equation initially. After a careful observation it was noticed that the polynomial equation has a
potential to provide accurate estimation of SS concentration in low turbid and high turbid water because it
has the best fit. However, the curve is very flat in the low turbid region and seems not to be capable of
detect variation in the estimation. More importantly, the trend of increase in concentration at high NDSSI
values may result in unphysical predictions. The power equation, although posses lower R
2
value than the
polynomial equation it shows the potential to detect the variability in SS concentration in both low and
high turbid water. Therefore, the power equation was considered to be the most suitable approach to
estimate the SS concentration in the water.
b
NDSSIaSSC
×=
(3)
399.3
69.18
×= NDSSISSC
(4)
Where, SSC = Suspended sediment concentration, a and b are the coefficients
Results and Discussion
The concept of NDSSI was applied using the obtained coefficients (from Equation 4) on (1) the obtained
ALOS AVNIR2 VNIR imagery to estimate and map suspended sediment concentration in the Midwest
USA 2008 flood water (Figure 6), (2) the obtained Landsat 5 TM VNIR imagery to estimate and map the
suspended sediment concentrations in the Lake Pontchartrain during the Bonnet Care Spillway flooding
event (Figure 7a), and before and after the Hurricane Katrina (Figure 7b and 7c).
Figure 6. Suspended sediment concentration estimation in the Mississippi River flood water within
Alexandria, MO and Warsaw, IL areas during the Mid West USA 2008 Flood.
A
B
C
Legend
SSC
A: April 10, 1997 (BCS)
B: August 22, 2005
(Pre-Katrina)
C: September 07, 2005
(Post-Katrina)
Figure 7. Suspended sediment concentration estimation in Lake Pontchartrain in different times
CCHE2D, a two- dimensional depth-averaged model, developed at the National Center for Computational
Hydroscience and Engineering (NCCHE), was applied to simulate the sediment transport in the Midwest
USA 2008 flood water and Lake Pontchartrain during the Bonnet Carre Spillway flooding event in April
1997 . CCHE2D is a 2D hydrodynamic model that can be used to simulate unsteady turbulent flows with
irregular boundaries and free surfaces (Jia et al., 1999, 2002). The simulation results were compared with
suspended sediment concentration estimation by NDSSI and were found in good general agreement
(Figure 8 and Figure 9).
Legend
SSC
Figure 8. Qualitative comparison (by visual inspection) between simulated SS concentration (by
CCHE2D) and remote sensing derived SS concentration estimation (By NDSSI).
Figure 9. Quantitative comparison between simulated SS concentration (by CCHE2D) and remote sensing
derived SS concentration estimation (By NDSSI).
Conclusion
Normalized Difference Suspended Sediment Index (NDSSI) was calculated using the Landsat 7 ETM+
VNIR imagery to map the spatial distribution (relative) of suspended sediment concentration in the water
of Mississippi River at different levels of turbidity. NDSSI index values were correlated with the near
real-time in-situ measurements of suspended sediment concentrations using different numerical
approaches. The relationship between NDSSI and in-situ measurements of SS concentration expressed by
power equation was found most suitable for quantitative estimation of SS concentration in the Mississippi
River. This technique, using the obtained coefficients was applied to estimate/map the SS concentration in
the Mississippi River during the Midwest USA 2008 flood and in Lake Pontchartrain during (1) Bonnet
Carre Spill Way flooding event and (2) before and after Hurricane Katrina. The results were compared by
the near real time simulated SS concentration data generated by CCHE2D numerical model developed at
NCCHE and found in good general agreement qualitatively and quantitatively. This research is still
evolving and this initial results indicate that (1) NDSSI has the potential to estimate (relative variation)
0
50
100
150
200
250
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
SS concentrations (mg/lit)
Number of SS measuring stations
Remote sensing derived SS concentration vs. simulated SS Concentration
Remote sensing SS Simulated SS
and map the spatial distribution of SS concentration in both river and lake environments, (2) NDSSI can
be used for quantitative estimation of SS concentration in these environments when coupled with two co-
efficients in a power equation, and (3) the fact that the same coefficients obtain from the data of the
Mississippi River can be used to estimate SS concentration in the Lake Pontchartrain indicates the
usefulness of the developed NDSSI. Acquisition of more in-situ measurements of SS concentrations are
on going to derive more general co-efficients and achieve more validation results.
Acknowledgments
This research was funded by the US Department of Homeland Security and was sponsored by the
Southeast Region Research Initiative (SERRI) at the Department of Energy's Oak Ridge National
Laboratory.
References
Binding, C. E., Bowers, E.G., and Mitchelson-Jacob. 2005. Estimating suspended sediment concentrations
from ocean color measurements in moderately turbid waters; the impact of variable particle
scattering properties. Remote Sensing of Environment. 94(2005):373-383.
Jia, Y., and Wang, S.S.Y. 1999. Numerical model for channel flow and morphological change studies,
Journal of Hydraulic Engineering. 125(9): 924-933.
Jia, Y., Wang, S.Y.Y., and Xu, Yichen. 2002. Validation and application of a 2D model to channels with
complex geometry. International Journal of Computational Engineering Science. 3(1):57-71.
Lathrop R G. 1992. Landsat Thematic Mapper monitoring of turbid inland water quality. Photogramm.
Eng. Remote Sens. 58(4): 465-470.
Populus, J., Hastuti, W., and Martin, J. L. 1995. Remote sensing as a tool for diagnosis of water quality in
Indonesian Seas. Ocean & Coastal Management. 27(3): 197-215.
Wang, X., and Ma, Y. 2000. The application of remote sensing technology in monitoring the water
quality of Tai Lake. Environmental Science. 11(21): 65-68.
Wang, X., Wang, Q., and Wu, Q. 2003. Estimating suspended sediment concentration in coastal water of
Minjiang River using remote sensing images. Journal of Remote Sensing. 7(1): 54-57.
Williams, A. N., and Grabau, W. E. 1973. Sediment concentration mapping in tidal estuaries. In Third
Earth Resources Technology Satellite-1 Sym. NASA SP-351, 1347-1386.
... Satellite remote sensing technology has been providing an opportunity for synoptic and multitemporal viewing of water quality for more than 25 years [1][2][3][4]. Sensors aboard satellites can measure the amount of solar radiation at various wavelengths reflected by surface water which can be correlated to different water quality parameters [5][6][7]. This constitutes an alternative means of estimating water quality, offering three significant advantages over ground sampling. ...
... It is one of the most studied remotely sensed water quality parameters. The majority of studies have found strongly correlated values detecting remotely sensed Chl-a in lakes, reservoirs and coastal areas using data from Landsat 1-3, Landsat 5, Landsat 8, RapidEye, Sentinel-2, and MODIS satellites [1,5,6,20,21,[32][33][34][35][36][37][38]. Successful correlations have also been found in past studies that utilized airborne remote sensing [30,39]. ...
... When compared to SSC, quantifying TSS has proven to be unreliable for samples measured from natural waters as methods were originally designed for analyzing wastewater [52]. Regardless, TSS have been successfully estimated in lacustrine and estuarine environments using regression analysis techniques based on in situ measurements and reflectance data obtained from spaceborne optical sensors such as Landsat 5 and Landsat 8 [5,[19][20][21]. Airborne remote sensing has been successful in assessing TSS of reservoirs and rivers, noting specific advantages over spaceborne satellites including less atmospheric interference, adaptable spatial resolution, and fewer temporal restraints [30]. ...
Article
Full-text available
The Tennessee River in the United States is one of the most ecologically distinct rivers in the world and serves as a great resource for local residents. However, it is also one of the most polluted rivers in the world, and a leading cause of this pollution is storm water runoff. Satellite remote sensing technology, which has been used successfully to study surface water quality parameters for many years, could be very useful to study and monitor the quality of water in the Tennessee River. This study developed a numerical turbidity estimation model for the Tennessee River and its tributaries in Southeast Tennessee using Landsat 8 satellite imagery coupled with near real-time in situ measurements. The obtained results suggest that a nonlinear regression-based numerical model can be developed using Band 4 (red) surface reflectance values of the Landsat 8 OLI sensor to estimate turbidity in these water bodies with the potential of high accuracy. The accuracy assessment of the estimated turbidity achieved a coefficient of determination (R2) value and root mean square error (RMSE) as high as 0.97 and 1.41 NTU, respectively. The model was also tested on imagery acquired on a different date to assess its potential for routine remote estimation of turbidity and produced encouraging results with R2 value of 0.94 and relatively high RMSE.
... But from the early twenty-first century, the geospatial technology has opened a new horizon for better apprehension of the contemporary water health status of waterbodies. Notably, remote sensing emerged as an indispensable tool and proved to be quite convenient not only in the mapping of the water quality but likely it offers resource managers the mechanism to monitor and maintain waterbodies in a well-timed and cost-effective manner (Biswas et al., 2020;Hossain et al., 2010). The water quality parameters e.g., sediments, nutrients etc. can alter reflected or emitted electromagnetic radiation from lake surface water, and such changes could be detected by remote sensing (El-Zeiny & El-Kafrawy, 2017;Nayek et al., 2017). ...
... The yellow substance is a soluble matter with no scatter capability, but it has a strong absorption effect on short-wave bands that highly reduce the underwater downwelling irradiance. Therefore, when the absorption of water and yellow substances is negligible, actual suspended sediment information could be obtained (Hossain et al., 2010;Wang et al., 2003). For Landsat TM and OLI imagery, it has been observed that the Blue band and NIR are most sensitive to water and water transparency (turbidity) (Wang & Ma, 2001). ...
Article
Full-text available
The Mirik lake is a well-known tourist destination in the Darjeeling Himalayan region of West Bengal. In the last few decades, the lake area experienced a surge in demand for the hospitality industry with the consequences of unrestricted municipal town growth. On this perspective, this research work addresses the contemporary problems of the Mirk municipality area involving the lake system using technologies of remote sensing and GIS. In accordance with the primary objective, the detailed land use/land cover change analysis of the past twenty years shows a significant escalating trend in built-up area as it has increased from 149.85 ha in 2000 to 178.25 ha in 2020, where the lake expresses a steady reduction of its perennial part. The gradual shrinkage in the lake area may become of grave concern and therefore, this particular work tries to examine the water quality parameters of Mirik lake. Analyses have shown that the lake shrinkage may be attributed to waste disposal, landfill pollution etc., substantially contributed to the raised level of suspended sediment concentration (3724.26 mg/lit in 2020) near the boundaries. Moreover, the outcome of the developed water pollution model identifies that the western and south-western parts of the lake are the most polluted regions and confirms numerous inlets of the western side push up the level of phosphorus, nitrogen and biological oxygen demand in lake water. Such lake degradation is a real threat to the mountainous lacustrine environment, and it requires comprehensive plans for better management practices at some regular interregnum.
... The obtained index cannot be attributed to an absolute value of SSC, but multi-temporal comparison is possible. A similar operation is performed, for example, when calculating the vegetation index NDVI [47]. ...
... where, ρ NIR (near-infrared) and ρ BLUE (blue) are the reflectance values of Sentinel Bands 2 and 8 (equivalent to Landsat bands 2 and 5). Hossain et al. [47] introduced the index with the use of data from the Landsat 7 ETM+ satellite. In this case, the blue channel corresponded to a frequency of 0.450-0.515 ...
Article
Full-text available
Glacier-fed hydrological systems in high latitude regions experience high seasonal variation in meltwater runoff. The peak in runoff usually coincides with the highest air temperatures which drive meltwater production. This process is often accompanied by the release of sediments from within the glacier system that are transported and suspended in high concentrations as they reach the proglacial realm. Sediment-laden meltwater is later transported to the marine environment and is expressed on the surface of fjords and coastal waters as sediment plumes. Direct monitoring of these processes requires complex and time-intensive fieldwork, meaning studies of these processes are rare. This paper demonstrates the seasonal dynamics of the Trebrevatnet lake complex and evolution of suspended sediment in the lake and sediment plumes in the adjacent Ekmanfjorden. We use the Normalized Difference Suspended Sediment Index (NDSSI) derived from multi-temporalSentinel-2 images for the period between 2016–2021. We propose a new SSL index combining the areal extent of the sediment plume with the NDSSI for quantification of the sediment influx to the marine environment. The largest observed sediment plume was recorded on 30 July 2018 and extended to more than 40 km2 and a SSL index of 10.4. We identified the greatest sediment concentrations in the lake in the beginning of August, whereas the highest activity of the sediment plumes is concentrated at the end of July. The temporal pattern of these processes stays relatively stable throughout all ablation seasons studied. Sediment plumes observed with the use of optical satellite remote sensing data may be used as a proxy for meltwater runoff from the glacier-fed Trebrevatnet system. We have shown that remote-sensing-derived suspended sediment indexes can (after proper in situ calibration)serve for large scale quantification of sediment flux to fjord and coastal environments.
... 4). This index is also between − 1 and + 1, with higher values indicating cleaner water (Hossain et al. 2010;Arisanty and Saputra 2017). ...
Article
Full-text available
Urmia Lake is the largest inland water body in Iran and the second hyper-saline lake in the world. It has been designated as a “conserved region of biosphere” by UNESCO due to its environmental importance and unique aquatic ecosystem. The lake has faced a variety of natural and anthropogenic hazards in recent years and has encountered dramatic changes in its natural hydrodynamic condition. There are a number of reasons for these changes, where primary causes are increased water consumption, especially in the agricultural sector, development of water storage structures in the lake basin, construction of the causeway, climate change, global warming, and droughts. Due to changes in the lake's natural condition, the study of its hydrodynamic pattern is inevitable. In this regard, the current study aims to simulate changes in the water temperature of Lake Urmia in order to study its hydrodynamic. MIKE hydrodynamic models are developed by DHI Water and Environment for simulation of flows in estuaries, bays, coastal areas, lakes, and oceans. To perform the simulations in this study, MIKE3 model was used. In the modeling process, the flow simulation is carried out simultaneously with the heat transfer model, covering all the hydrodynamic conditions of the lake. In order to simulate water temperature and its effects on the lake hydrodynamics in the MIKE model, density is considered as a function of temperature, then the heat transfer equations are solved at each time step. The forces governing the hydrodynamic equations include wind, air pressure, tide, wave, and Coriolis forces. However, the forces governing heat transfer/diffusion equations are of a different nature. Air temperature, relative humidity, and clearness coefficient are important inputs for the simulation of changes in the water temperature in the lake. Due to the high accuracy of the obtained values, ECMWF model data were used in the model. The validity of the numerical model was also assessed by comparing the simulated results against satellite data. The information is provided by the Group for High Resolution Sea Surface Temperature (GHRSST). In this group, global sea surface temperature data are generated using a multi-scale two-dimensional variational (MS-2DVAR) blending algorithm. These sea surface temperature data are obtained from various satellites with multiple sensors (such as AVHRR, AATSR, SEVIRI, AMSRE, TMI, MODIS, GOES, MTSAT-1R, etc.). In this study, changes in the water temperature of Lake Urmia were simulated in order to study its hydrodynamic condition. Initially, the data used to simulate water temperature changes are presented. These data included air temperature, relative humidity, and clearness coefficient. Due to the high accuracy and generalizability of the ECMWF model output to the entire computational domain, the output of this model was used to obtain the above data. MIKE3 hydrodynamic model was used to perform the simulations. In order to investigate the effects of precipitation, evaporation, and rivers discharge on water temperature changes, two models, one with and the other without considering these factors were implemented. The water temperatures were compared in these two models. The results showed that water temperature values were approximately the same for the two cases. Also, a comparison between the water temperature output results at different depths revealed that due to the low depth of the lake, the temperature difference between the surface layer and the near-bed layer was low and reached a maximum of 0.2 °C. In addition, GHRSST satellite data was used to validate the model results. Evaluations indicated that the model results were in good agreement with the measured data, and seasonal variations in lake surface temperature were also well simulated. Moreover, the effect of causeway on the spatiotemporal distribution of lake water temperature has been investigated. For this purpose, simulation of temperature changes was considered over a one-year period. The results demonstrated that the water temperature of the lake did not change significantly in both with and without causeway, and the temperature exchange between northern and southern parts of the lake occurred in both conditions. Hence, this model can be used as an efficient tool to assess the effect of causeway on the flow pattern, salinity distribution and sedimentation process in both parts of the lake.
... Delineation of settlements, agriculture and horticulture in the immediate vicinity of Wular Lake (2008-2019) using LISS IV satellite data Satellite remote sensing technology and environmetric techniques have been providing an opportunity for the interpretation, evaluation and reliable characterization of surface water quality. These powerful tools have been used to enhance understanding of the spatio-temporal variation in water quality resulting from anthropogenic and natural processes (Hossain et al. 2010(Hossain et al. , 2014Hossain, Mathias, and Blanton 2021;Dar, Hamid, et al. 2021). In the present study, we attempt to map and quantify agriculture, horticulture, water body, and settlement land use categories. ...
Article
Full-text available
During the last one and half decade, Kashmir valley's water bodies have been facing widespread water quality deterioration due to anthropogenic interferences' enhancement. In this context, the present study aims to monitor the water quality of Wular Lake by subjecting the acquired datasets to statistical approaches to infer status about distinctions and similarities among different sampling sites under investigation. The study integrated multiple data sets pertaining to analysis and assessment of water quality parameters, physicochemical concentration values, land use change systems, and water quality index through remote sensing and GIS environments. For limnological characterization of the lake, surface water quality data have been collected from seven randomly selected site pertaining to 22 physicochemical parameters, which were further analyzed. Settlements, agriculture, and horticulture in the lake's immediate vicinity were mapped using LISS IV satellite data of the year 2008–2019. The water quality index under this study comprises 12 measurable physicochemical parameters–pH, electrical conductivity, dissolved oxygen, total hardness, calcium content, magnesium content, nitrate nitrogen, chloride, turbidity, potassium, sodium and total alkalinity. Based on WQI values, the whole lake was divided into five zones, comprising least polluted zone (<50), moderately polluted zone (50–100), highly polluted zone (100–200), very highly polluted zone (200–300) and extremely polluted zone (>300). From this study, it has been found that WQI values for site S1, S2, S3, S4, S5, S6 and S7 are 179.62, 149.8, 160.61, 205.21, 200.5, 164.95 and 183.74, respectively. Thus, revealing that site S4 and S5 fall under the 'very highly polluted zone' and the remaining sites S1, S2, S3, S6 and S7 under 'highly polluted zone'. Land use category settlement record a growth rate of +50.79%, horticulture +25.96%, agriculture −11.8% and water body (open waters) −24.3% from the year 2008–2019. The reckless use of fertilizers, insecticides, pesticides in orchards and agricultural farms reinforced by concrete jungles in the immediate vicinity of Wular Lake have also enhanced the nutrient enrichment and deterioration of the water quality. Thus, escalating its scope towards the process of eutrophication.
... 4). This index is also between − 1 and + 1, with higher values indicating cleaner water (Hossain et al. 2010;Arisanty and Saputra 2017). ...
Article
Full-text available
Different sensing methods provide valuable information for comprehensive monitoring strategies, which are crucial for the ecological management of lakes and watersheds. Subsequently, the resulting spatio-temporal information can be considered the fundamental knowledge for the water resources management of watersheds. Lake Urmia is deemed one of the most important aquatic habitats in Iran. It has been experiencing significant changes during recent years due to climate change, anthropogenic activities, and a lack of coherent management approaches. Hence, awareness of the hydro-ecological factors during the last few decades is critical for identifying the problems. In this research, the impacts of changes in key parameters such as precipitation, evapotranspiration, water surface temperatures, suspended sediment concentration, saline features, and vegetation are explored using satellite imagery. The primary purpose of this study is to evaluate the Lake Urmia crisis concerning human-involved and climate factors such as the agriculture sector and construction of the causeway. In this regard, a limbic-based Emotional Artificial Neural Network (EANN) is developed as a non-linear universal mapping and implemented for the first time to demonstrate the interactions between the considered hydro-ecological factors and the sensitivity of the two indicators the lake health. Providing a comprehensive spatio-temporal analysis is another objective of this study to detect the onset of deterioration in the parameters. The values of the efficiency criteria were measured to evaluate the sensitivity of the EANN models to the related inputs. The results of the model in scenario 4 with evapotranspiration, precipitation, runoff and vegetation as input variables led to higher performance with the best efficiency criteria, including DC = 0.868 and RMSE = 0.096. The quantitative results confirm that the combination of both climate and anthropogenic factors, including the agricultural sector's overdraft, leads to the most efficient EANN model and, consequently, is considered the leading cause of the crisis.
... Various approaches used in mapping by utilizing remote sensing technology have been developed over time to estimate TSS concentrations using satellite imagery. Some of these approaches can be categorized into 4 general groups (Hossain et al., 2007): (1) Simple regression (correlation between single band and in-situ data. (2) Spectral unmixing technique (3) Band ratio technique by using two or more bands, and (4) Multiple regressions (using multiple bands and in-situ data). ...
Article
Full-text available
TSS (Total Suspended Solid) is one of the optical parameters that can be used for turbidity key indicator to assess water quality. The rapid development of remote sensing technology in the field of mapping has resulted in various methods for estimating TSS concentrations. The spatial, spectral, and temporal characteristics of PlanetScope data have the potential to estimate TSS concentrations. This study aims to determine the best method for estimating TSS concentrations and mapping the spatial distribution of TSS at a depth of 0 – 0.2 m using PlanetScope data. There are 4 single bands, 12 band ratio combinations, and 4 PC-bands in TSS mapping. Single bands, band ratio combinations, and PC-band which able to pass the significance limit of r value on the number of samples (n) are used in empirical modeling of PlanetScope data with field data using regression tests. The results show that: 1) 4 band ratio combinations (B1/B4, B2/B4, B3/B4, B4/B3) and one PC-band (PC-2) significantly correlated with TSS (mg/l), 2) PC-2 is the best spectral transformation in estimating TSS concentrations in Menjer Lake, indicated by the SE value of 3.47 mg/l with maximum accuracy produced at 78.62%, 3) all models that significantly correlated are over-estimated, indicated by the variations in model plots are below the 1:1 plot line, 4) high TSS concentrations are in the north, west, and south around the edge of the lake because of the inlets and the floating net cages, while the low concentration is in the middle of the lake.
Article
Ice plays key roles in regulating hydrological, ecological, biogeochemical, and socioeconomic functions of lakes. Long-term in situ lake ice phenological records indicate that lake ice is trending toward later freeze-up, earlier breakup, and a shorter ice duration. Parallel to study of lake ice using in situ records and process-based models, satellite remote sensing can expand our understanding of lake ice change over large spatial scales. However, most remote sensing studies have focused on large lakes or short periods of time, which may not robustly represent changes over multidecadal time periods or in the much more numerous small lakes. Here, we present a random forest model, Sensitive Lake Ice Detection (SLIDE), to accurately extract ice conditions from Landsat TM, ETM+, and OLI images. We trained the model using a manually labeled lake ice dataset (1089 labeled areas over 995 lakes globally). Our results show that our model achieves accurate classification between ice/snow and water (accuracy: 97.8%, kappa coefficient: 95.5%). Comparing Landsat-derived ice cover with in situ ice conditions, we show that our model produces less bias, lower RMSE, and higher kappa than does the Landsat snow/ice flag from the quality assessment band. This is especially true during the transitional period surrounding the ice on and off dates reported from in situ (mean bias -7.3% from our model, -17.3% from the Landsat quality band). Our results demonstrate the feasibility of mining the rich Landsat archive to study lake ice dynamics and of better flagging ice-affected lake observations.
Article
This study reports on an investigation of water quality calibration algorithms under turbid inland water conditions using Landsat Thematic Mapper (TM) multispectral digital data. TM data and water quality observations (total suspended solids and Secchi disk depth) were obtained near-simultaneously and related using linear regression techniques. The relationships between reflectance and water quality for Green Bay and Lake Michigan were compared with results for Yellowstone and Jackson Lakes, Wyoming. Results show similarities in the water quality-reflectance relationships, however, the algorithms derived for Green Bay - Lake Michigan cannot be extrapolated to Yellowstone and Jackson Lake conditions. 17 refs.
Article
In this paper a depth-integrated 2D hydrodynamic and sediment transport model, CCHE2D, is presented. It can be used to study steady and unsteady free surface flow, sediment transport, and morphological processes in natural rivers. The efficient element method is applied to discretize the governing equations, and the time marching technique is used for temporal variations. The moving boundaries were treated by locating the wet and dry nodes automatically in the cases of simulating unsteady flows with changing free surface elevation in channels with irregular bed and bank topography. Two eddy viscosity models, a depth-averaged parabolic model and a depth-averaged mixing length model, are used as turbulent closures. Channel morphological changes are computed with considerations of the effects of bed slope and the secondary flow in curved channels. Physical model data have been used to verify this model with satisfactory results. The feasibility studies of simulating morphological formation in meandering channels and flows in natural streams with instream structures have been conducted to demonstrate its applicability to hydraulic engineering research/design studies of stream stabilization and ecological quality among other problems.
Article
Shrimp development in Indonesia, after a decade of strong expansion and intensification, began to be faced with critical problems of mortality. General degradation of coastal waters is likely to be the major reason, and this mainly occurs on sites in the Java Sea which appear as confined seas subjected to high telluric effluent levels and anthropic pressure. Remote sensing provides synoptic vision over large areas. Several scenes were processed for land-use mapping using conventional classification techniques. In this study, this was specifically done to determine water quality. A general empirical relation for the Java Sea, independent of time and place, was found between image and field data, namely pigments and suspended matter. Applying this relation to a new site provides an initial approach to water type and, together with land use mapping, a preliminary assessment of the suitability of the area for shrimp aquaculture development. It also provides guidance for efficient subsequent field sampling. Recommendations are given for use of remote sensing to improve site assessment and monitoring.
Article
This study investigates the use of single-band reflectance at visible wavelengths for the derivation of suspended sediment concentrations in the Irish Sea. A reasonably strong relationship was observed between irradiance reflectance at 665 nm (R665) and mineral suspended sediment (MSS) concentrations. Variability in the Reflectance–MSS relationship was found to be the consequence of changes in the mass-specific scattering coefficient (bMSS*) brought about by differences in particle properties such as grain size and composition. A systematic increase in the slope of the Reflectance–MSS relationship was observed with increasing bMSS*. A reflectance model is presented that highlights the dependence of reflectance on bMSS* and suggests that the errors in predicted MSS concentrations can be reduced from 56% to as little as 12% with prior knowledge of the scattering properties of the sediments under study. This paper highlights the need for a complete understanding of the scattering properties of particles in order to accurately estimate MSS concentrations from reflectance measurements. It is suggested that in order to obtain quantitative estimates of MSS in moderately turbid waters from space, it may be necessary to pre-determine scattering efficiencies, bMSS*, for the area of interest.
Article
A depth-integrated two-dimensional hydrodynamic and sediment transport model, CCHE2D, has been developed for analyzing free surface flows for general applications to hydraulic engineering problems. Mixed schemes of finite element method (Efficient Element Method) and finite volume method are adopted for solving momentum and continuity equations. A unique velocity correction method was developed to couple the governing equations. Pressure (free surface) field is discretized on a staggered grid. Two zero equation turbulence closure schemes: a depth averaged parabolic, a depth averaged mixing length eddy viscosity model, and a two-equation k-e model for depth-averaged flows are included for different application needs. Algebraic systems of equations are solved implicitly by using SIP method. This model can be used for simulating steady and unsteady near field and natural river flows with relatively shallow water depth. Validation tests for this model show good agreements between the model prediction and data of physical models and field measurements.
Estimating suspended sediment concentration in coastal water of Minjiang River using remote sensing images
  • X Wang
  • Q Wang
  • Q Wu
Wang, X., Wang, Q., and Wu, Q. 2003. Estimating suspended sediment concentration in coastal water of Minjiang River using remote sensing images. Journal of Remote Sensing. 7(1): 54-57.