water

Article A Study on the Flux of Total Suspended Matter in the in Based on Remote-Sensing Data

Zhuoqi Zheng 1, Difeng Wang 1,2,*, Fang Gong 1,3, Xianqiang He 1,2 and Yan Bai 1,3

1 State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, MNR, Hangzhou 310012, China; [email protected] (Z.Z.); [email protected] (F.G.); [email protected] (X.H.); [email protected] (Y.B.) 2 Marine Resources Big Data Center of South China Sea, Southern Marine Science and Engineering Guangdong Laboratory, Zhanjiang 524088, China 3 National Earth System Science Data Center, Beijing 100101, China * Correspondence: [email protected]

Abstract: The flux of total suspended matter (TSM), FTSM, output by several large rivers in Asia, has been in decline due to human activities. As the estuary of the –, the Padma River transports a significant amount of suspended matter (SM) to the Bay of

each year. In this study, the TSM concentration (CTSM) and FTSM in the Padma River in the period 1991–2019 were calculated based on the data acquired by the Landsat series satellites and an empirical TSM algorithm model for large, high-turbidity rivers. The results showed that the maximum and

minimum FTSM values (318 ± 62 and 73 ± 29 mt, respectively) in the Padma River occurred in 2011 and 2015, respectively. On average, FTSM in the Padma River decreased at an annual rate of 3.3 mt (p < 0.01). The impact of human activities on C contributed more significantly to the changes in  TSM  FTSM (R = 0.76) than natural factors (R = 0.44). Due to a lack of water conservancy facilities within the river basin, changes in the water and soil retention capacity due to the changes in vegetation Citation: Zheng, Z.; Wang, D.; Gong, R − F.; He, X.; Bai, Y. A Study on the Flux coverage were an important human factor ( = 0.79). of Total Suspended Matter in the Padma River in Bangladesh Based on Keywords: total suspended matter; ; Landsat; flux estimation Remote-Sensing Data. Water 2021, 13, 2373. https://doi.org/10.3390/ w13172373 1. Introduction Academic Editor: Achim A. Beylich Total suspended matter (TSM) is an important index for evaluating the water quality in water bodies. In water-quality management, both the concentration and flux of TSM Received: 14 July 2021 (CTSM and FTSM, respectively) depend on the gross primary productivity, the concentration Accepted: 26 August 2021 of heavy metals, and the flux of substances such as radionuclides and organic micropollu- Published: 29 August 2021 tants [1]. The concentration of total suspended matter, similarly, has a significant impact on the biogeochemical properties of coastal oceanic waters. The presence of TSM reduces Publisher’s Note: MDPI stays neutral the underwater transmittance of a water body [2] and alters its optical properties. For with regard to jurisdictional claims in inland water bodies (e.g., rivers and lakes), TSM also plays a substantial role in maintaining published maps and institutional affil- ecosystem equilibrium and controlling the habitat of aquatic organisms. The C value in iations. TSM a water body can help to maintain its ecosystems only when controlled within a suitable range. For example, an extremely high CTSM may cause wear damage to the eggs of aquatic organisms or directly lead to the death of their larvae [3]. On the other hand, some prey fish can evade their natural enemies by taking advantage of turbid water [4,5]. In addition, TSM Copyright: © 2021 by the authors. and sediment can provide the environments and nutrients needed for the development of Licensee MDPI, Basel, Switzerland. estuarine ecosystems. This article is an open access article Monitoring and studying the changes in the flux of TSM in large rivers is currently distributed under the terms and a focus area for research. Recent studies have shown that the flux of SM transported conditions of the Creative Commons by large deltaic rivers is in decline, especially in Asia [6]. The transfer of both water Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ and sediment to the sea will decrease as more dams and other river projects are built and 4.0/). utilized [7]. In the past 60 years, the flux of sediment transported seaward by the nine major

Water 2021, 13, 2373. https://doi.org/10.3390/w13172373 https://www.mdpi.com/journal/water Water 2021, 13, 2373 2 of 20

rivers in China [8] as well as the Mekong River [9] have shown a statistically significant negative trend due to the construction of reservoirs. At present, the traditional calculation methods of FTSM are mainly based on measured hydrological data [8], hydrological climate model [9], isotope analysis [10], UAV measurement [11], and so on. The above method has the disadvantages of a large amount of calculation, difficult data acquisition, and excessive research cost [12]. In large tropical rivers such as Padma River, the installation and maintenance of in situ stations is a very costly task, limiting the data available for both river discharge and sediment load trend assessment. For these reasons, assessment of FTSM in large rivers by remote sensing monitoring has received increasing interest from the scientific community [13]. As an efficient, low-cost, long-term indirect observation technique, remote sensing plays a vital role in several research fields including hydrology. The Padma River is the estuary of the Ganges–Brahmaputra River. Glacial meltwater from the Tibetan Plateau, known as the “Asian Water Tower”, is a primary source of fresh- water for the Ganges–Brahmaputra River [14]. After converging in the Padma River, the water in the Ganges–Brahmaputra River flows into the Bay of Bengal, significantly affecting the eco-environmental changes in its northern waters. Many researchers have conducted relevant studies on the water constituents in the northern Bay of Bengal, including the measurement and monitoring of parameters such as the concentration of heavy metals [15], nutrients [16], dissolved oxygen [17], chlorophyll [18], and sediment [19]. Other researchers have observed this region over a large area through satellite remote sensing [20]. However, to date, few studies have monitored and estimated the flux of SM at the mouth of the Ganges River based on high-resolution satellite remote-sensing data.

2. Materials and Methods 2.1. Study Area The Padma River is the estuary of the Ganges–Brahmaputra River. The estuaries of three large rivers, namely, the Ganges, the Brahmaputra, and the Meghna, flow into the Bay of Bengal through Padma River (see Figure1). Each year, the Ganges–Brahmaputra river system, which has the world’s largest annual sediment discharge, approximately 100 billion tons [21,22], transports a significant amount of freshwater and SM into the Bay of Bengal [23]. The northern Bay of Bengal is home to the world’s largest mangrove forest—the —which discharges 18 million tons of SM and organic matter into the Bay of Bengal via various small tributaries, resulting in a complex ecosystem in the coastal waters of the northern Bay of Bengal [24]. The Padma River—a main tributary of the Ganges River—converges in the down- stream region with the —a tributary of the Brahmaputra River—and eventu- ally flows into the Bay of Bengal. Since 1966, approximately 660 km2 of land have been eroded by the Padma River [25].

2.2. Hydrological Data The reanalysis data used in this study included discharge and precipitation. The discharge data originated from the global modelled daily river discharge data from the Global Flood Awareness System [26]. Specifically, the monthly mean discharge data for the period 1991–2019 with the resolution 0.1◦ × 0.1◦ were used. There were a total of 348 data records. Water 2021, 13, x FOR PEER REVIEW 3 of 20

Water 2021, 13, 2373 3 of 20

Figure 1. Illustration of the location of the study area. The image on the left-hand side shows a map of the East Asian Figure 1. Illustration of the location of the study area. The image on the left-hand side shows a map of the East Asian continent with digital elevation model (DEM) data downloaded from USGS as the background. The blue lines signify six continent with digital elevation model (DEM) data downloaded from USGS as the background. The blue lines signify six largelarge rivers originatingoriginating fromfrom the the Tibetan Tibetan Plateau. Plateau. The The area area within within the the black black outline outline is the is Ganges–Brahmaputra the Ganges–Brahmaputra River River Basin, Basin,while thewhile area the within area within the red the box red is the box study is the area. study The area. image The on image the right-hand on the right-hand side is an side example is an ofexample a Landsat of a 5 Landsat TM image 5 TMof the image Padma of the River Padma (taken River on 17(taken October on 17 1994). October 1994).

2.2. HydrologicalThe precipitation Data data originated from the Global Precipitation Climatology Project [27], which is a satellite precipitation product (resolution: 0.5◦ × 0.5◦) produced by the Global The reanalysis data used in this study included discharge and precipitation. The dis- Precipitation Climatology Center through combining the infrared (IR) and microwave data charge data originated from the global modelled daily river discharge data from the acquired by dozens of geostationary and polar-orbit satellites and correcting the resulting Global Flood Awareness System [26]. Specifically, the monthly mean discharge data for data against the data acquired by multiple ground stations worldwide. Specifically, the the period 1991–2019 with the resolution 0.1° × 0.1° were used. There were a total of 348 monthly mean precipitation data (data format: netCDF4) for the period 1991–2019 were data records. used in this study. There were a total of 348 data records. TheThe precipitation Bay of Bengal data has aoriginated sub-tropical from the Global climate Precipitation with high Climatology temperature Project all year [27],round which and distinctis a satellite monsoon precipitation and non-monsoon product seasons.(resolution: The 0.5° monsoon × 0.5°) season produced is from by Maythe Globalto October, Precipitation and about Climatology 84% of the rainfallCenter through occurs from combining June to the October. infrared The (IR) precipitation and micro- is waveconcentrated. data acquired In addition, by dozens the terrain of geostationary is low and flat, and and polar-orbit the drainage satellites is not and smooth, correcting which theis easy resulting to lead data to floodagainst disasters. the data The acquired dry season by multiple is from ground November stations to April. worldwide. Affected Spe- by cifically,the northeast the monthly monsoon, mean there precipitation is drought data and (data less precipitation. format: netCDF4) Meanwhile, for the period the water 1991– in 2019non-monsoon were used seasonin this isstudy. not enough There were to meet a total the of requirements 348 data records. of irrigation and shipping, nor toThe maintain Bay of Bengal the minimum has a sub-tropical environmental monsoon flow climate of the river with [ 28high]. temperature all year round and distinct monsoon and non-monsoon seasons. The monsoon season is from May to2.3. October, Remote-Sensing and about Data 84% of the rainfall occurs from June to October. The precipitation is concentrated.Landsat data, In addition, including the Landsat-8 terrain Operationalis low and flat, Land and Imager the drainage (OLI)/Thermal is not smooth, Infrared whichSensor is (TIRS) easy to L1 lead products to flood (resolution: disasters. 30The m) dry and season Landsat-5 is from Thematic November Mapper to April. (TM) Af- L1 fectedproducts by the (resolution: northeast 30 monsoon, m), were usedthereas is experimentaldrought and satelliteless precipitation. data in this Meanwhile, study. Sentinel- the water2 L2A in reflectance non-monsoon products season (resolution: is not enough 10 m) wereto meet used the to requirements perform spectral of irrigation validation and on shipping,the Landsat nor satellite to maintain data. the These minimum satellite en datavironmental were downloaded flow of the from river the [28]. official website of the United States Geological Survey (USGS) (https://earthexplorer.usgs.gov/ (accessed 2.3.on 22Remote-Sensing July 2020)). The Data Landsat products, covering the period from 1991 to 2019, provided imagesLandsat of the data, non-tidal including reaches Landsat-8 of the Padma Operatio Rivernal Land (row andImager column (OLI)/Thermal numbers: 137–044).Infrared SensorAt least (TIRS) one image L1 products was selected (resolution: for each 30 season. m) and After Landsat-5 the unusable Thematic data Mapper covered (TM) by cloud L1 productswere removed, (resolution: a total 30 of m), 85 imageswere used were as obtained. experimental The numbersatellite ofdata effective in this values study. of Senti- each nel-2pixel L2A can bereflectance seen in Figure products2. (resolution: 10 m) were used to perform spectral validation on the Landsat satellite data. These satellite data were downloaded from the official web-

Water 2021, 13, x FOR PEER REVIEW 4 of 20

site of the United States Geological Survey (USGS) (https://earthexplorer.usgs.gov/ (ac- cessed on 22 July 2020)). The Landsat products, covering the period from 1991 to 2019, provided images of the non-tidal reaches of the Padma River (row and column numbers: 137–044). At least one image was selected for each season. After the unusable data covered by cloud were removed, a total of 85 images were obtained. The number of effective val- Water 2021, 13, 2373 ues of each pixel can be seen in Figure 2. 4 of 20

Figure 2. Location of the section selected for calculating FTSM. The color bar shows the number of effective values of each pixel. The section was selected from the branchless part of the river Figure 2.with Location a relatively of the large section amount selected of available for calculating data. North FTSM of. theThe section color arebar mostly shows input the number tributaries, of effective whilevalues south of each of the pixel. section The are section mostly was distributories. selected from The contributionsthe branchless of part the tributaries of the river were with also a relativelyconsidered large amount when of the available flux was calculated.data. North of the section are mostly input tributaries, while south of the section are mostly distributories. The contributions of the tributaries were also consid- ered when theThe flux Landsat was calculated. data used in this study were obtained from the official website of the USGS. Specifically, the L1 standard data (which had been geometrically corrected) were Thedownloaded. Landsat data In addition,used in thethis remote-sensing study were obtained data were from corrected the official based on website the DEM of data. the Further, the following processes were performed to obtain more accurate reflectance data. USGS. Specifically, the L1 standard data (which had been geometrically corrected) were downloaded.2.3.1. Radiometric In addition, Calibration the remote-sensing data were corrected based on the DEM data. Further, the Whenfollowing remote-sensing processes were data performe acquired atd to different obtain timesmore andaccurate locations reflectance or by differ- data. ent sensors are compared, it is necessary to convert the grey values of images to abso- 2.3.1. Radiometriclute radiance Calibration values to facilitate the calculation of the spectral reflectance or radiance Whenof ground remote-sensing features [29 data]. In acquired this study, at radiometric different times calibration and locations was performed or by usingdifferent the sensors Radiometricare compared, Calibration it is necessary function to within conver thet the Radiometric grey values Correction of images tool into theabsolute ENVI 5.3ra- software package. diance values to facilitate the calculation of the spectral reflectance or radiance of ground features2.3.2. [29]. AtmosphericIn this study, Correction radiometric calibration was performed using the Radiometric Calibration Infunction this study, within atmospheric the Radiometric correction was Correction completed tool using in thethe FLAASH ENVI 5.3 Atmospheric software package.Correction Model within the Radiometric Correction tool in the ENVI 5.3 software package. The FLAASH Atmospheric Correction Model, developed through improving the Moderate Resolution Atmospheric Transmission model, has relatively high accuracy in correcting hyperspectral and multispectral data [30]. Specifically, based on the geographical location of the study area, the Tropical Atmospheric Model and Maritime Aerosol Model were selected.

Water 2021, 13, 2373 5 of 20

2.4. CTSM Inversion Model

The CTSM algorithm for the estuary of the Yangtze River developed by Liu et al. (2019) [31] was selected in this study for CTSM estimating based on the following considera- tions (see Table1):

1. The estuarian waters of both the Yangtze and Ganges Rivers are high in CTSM and contain TSM mostly from terrestrial sources [31]. 2. The estuaries of the Yangtze and Ganges Rivers are similar in terms of CTSM and the content ranges of main water constituents (e.g., chlorophyll) [18]. 3. The sensitive bands of the Landsat TM sensor for the TSM and main water constituents (e.g., chlorophyll) in both the estuaries of the Yangtze and Ganges Rivers are identical. 4. The regions where the estuaries of the Yangtze and Ganges Rivers are located share the same rainy season. Therefore, high discharges and high CTSM values occur at the same time in the estuaries of the Yangtze and Ganges Rivers.

Table 1. Comparison of the characteristics of the Yangtze and Padma Rivers.

Average Sensitive Bands Chlorophyll River C in the River Rainy Season Data Sources TSM for TSM Concentration at the Estuary Yangtze River 30.5–735.4 mg/L B4 and B1 May–October 3.60 mg/L Liu et al. (2019) [31] Rahman et al. (2016) [12] Padma River 32.2–821.7 mg/L B4 and B1 May–October 5.54 mg/L and Jutla et al. (2012) [18]

Thus, the algorithm developed by Liu et al. (2019) [31] is also applicable to the estuary of the Ganges River. The CTSM was calculated using the following algorithm:

 log (TSM) = a × Ratiob, 10 (1) Ratio = (band4)/(band1),

where a and b are both fitting coefficients (a = 2.1454 and b = 0.2945 for Landsat-5 TM data; a = 2.1012 and b = 0.2953 for Landsat-7 Enhanced TM Plus (ETM+) data; the absolute relative errors for Landsat-5 TM and Landsat-7 ETM+ data are 13.69% and 13.46%, respec- tively) and band 4 and band 1 are the reflectance values of bands 4 and 1 of the Landsat-5/7 sensors, respectively. To apply the above algorithm to the Landsat-8 OLI, the ETM+/OLI transformation functions established by Roy et al. (2016) [32] (see Table2) were used.

Table 2. Transformation functions between ETM+ and OLI top-of-atmosphere reflectance sensors developed through ordinary-least-squares (OLS) regression (Roy et al. [32]).

Mean Mean Relative Root-Mean-Square Sensor Transformation n 2 DifferenceOLI— Deviation Band Functions R (p Value) ETM+ Difference (Reflectance) OLI—ETM+ (%) (Reflectance) OLI = 0.0173 + 0.8707 ETM+ Blue 29,697,049 0.710 (<0.0001) (~0.48 µm) ETM+ = 0.0219 + 0.8155 OLI 0.0013 0.69 0.0259 OLI = 0.0374 + 0.9281 ETM+ Near infrared 29,767,214 0.711 (<0.0001) (~0.85 µm) ETM+ = 0.0438 + 0.7660 OLI 0.0194 6.45 0.0637 Water 2021, 13, 2373 6 of 20

2.5. M–K Test The Mann–Kendall (M–K) test [33,34] is a climate diagnosis and prediction technique that can be used to determine whether an abrupt climate change occurred from a climate data series and, if so, its time of occurrence:

n i−1  S = ∑ ∑ sign Xi − Xj , i=2 j=1  S−1 Z = q S > 0,  n(n−1)(2n+5) (2)  18 Z = 0 S = 0,  S+1  Z = q S < 0,  n(n−1)(2n+5) 18

where Xn is the value of the variable. A positive Z value suggests an upward trend in the variable, a negative Z value suggests a downward trend in the variable, and absolute Z values greater than 1.28, 1.64, and 2.32 signify the passing of significance tests at confidence levels of 90%, 95%, and 99%, respectively. The test result (i.e., Z value) reflects the overall trend of the variable. The M–K abrupt-change test can be used to determine whether there exists an abrupt change. Similarly, for variable series xn, we define the following:

k  1, x > x , S = r , r = i j (3) k ∑ i i 0, x < x , i=1 i j

k(k − 1) k(k − 1)(2k + 5) E[S ] = , var[S ] = 1 ≤ k ≤ n, (4) k 4 k 72

(Sk − E[Sk]) UFk = p , (5) var[Sk]

UBk = −UFk. (6)

Then, whether there is an abrupt change in variable xn can be analyzed based on the two statistics UFk and UBk. The point of intersection of the UFk and UBk curves is the abrupt-change point. The UFk curve and UBk curve signify the sequential and reversed time series for xn, respectively. The critical values for the significance levels a = 0.05 and 0.01 (U0.05 and U0.01, respectively) are ±1.96 and ±2.32, respectively. If the UFk or UBk value is greater than 0, it means that the series exhibits an upward trend; otherwise, the series exhibits a downward trend. If the curves exceed the critical straight lines, this indicates a significant upward or downward trend. The time period for which the curves exceed the critical lines is determined as the time period of the abrupt change. The point of intersection between the UFk and UFk curves within the critical lines corresponds to the starting time of the abrupt change.

2.6. Validation of Landsat Data with Sentinel Data Sentinel-2 and Landsat-8 satellite data are extensively used in remote sensing of ocean colors. In addition, a large number of algorithms have been developed to integrate the data acquired by the sensors of these two satellites [35]. In this study, two Sentinel-2 L2A products were used to validate the atmospheric correction accuracy of the Landsat-8 data (see Figure3). The two selected Sentinel-2 images were captured at 04:29:19 on 11 February 2020 and at 04:31:09 on 23 November 2019, respectively. The two corresponding Landsat-8 OLI/TIRS images were captured at 04:24:52 on 11 February 2020 and at 04:24:52 on 23 November 2019, respectively. Water 2021, 13, x FOR PEER REVIEW 7 of 20

critical lines is determined as the time period of the abrupt change. The point of intersec- tion between the UFk and UFk curves within the critical lines corresponds to the starting time of the abrupt change.

2.6. Validation of Landsat Data with Sentinel Data Sentinel-2 and Landsat-8 satellite data are extensively used in remote sensing of ocean colors. In addition, a large number of algorithms have been developed to integrate the data acquired by the sensors of these two satellites [35]. In this study, two Sentinel-2 L2A products were used to validate the atmospheric correction accuracy of the Landsat-8 data (see Figure 3). The two selected Sentinel-2 images were captured at 04:29:19 on 11 February 2020 and at 04:31:09 on 23 November 2019, respectively. The two corresponding Landsat-8 OLI/TIRS images were captured at 04:24:52 on 11 February 2020 and at 04:24:52 on 23 November 2019, respectively. Water 2021, 13, 2373 7 of 20

Figure 3. Location of the areas in the Landsat and Sentinel images (the image with a red border is the Sentinel image, and the image with a blue border is the Landsat image). As mentioned above, the Landsat-8 OLI image was radiometrically calibrated and atmospherically corrected using the ENVI 5.3 software package. The image captured by Figure the3. Location Multispectral of the Instrument areas in (MSI) the Landsat onboard theand Sentinel-2 Sentinel satellite images was (the radiometrically image with a red border is the Sentinelcalibrated image, and and atmospherically the image corrected with a blue using border the SNAP is software.the Landsat The differenceimage). between the OLI and MSI was eliminated using the sensor-transformation method developed by AsZhang mentioned et al. (2018) above, [36] (see the Table Landsat-83). OLI image was radiometrically calibrated and Table 3. Transformationatmospherically functions between corrected the Sentinel-2 using MSI andthe Landsat-8 ENVI OLI5.3 developedsoftware through package. OLS regression The image captured by (Zhang et al. [36]).the Multispectral Instrument (MSI) onboard the Sentinel-2 satellite was radiometrically calibrated and atmospherically corrected using theMean SNAP software.Mean Relative The difference be- Band Transformation Function Sample Size R2 (p Value) OLI—MSI OLI—MSI tween the OLI and MSI was eliminated using theDifference sensor-transformationDifference (%) method devel- µ OLI = 0.0003 + 0.9570 MSI − − Blue λ (~0.48 m) opedMSI by = 0.0039Zhang + 0.9383 et OLIal. (2018)65,347,909 [36] (see0.8980 Table (<0.0001) 3). 0.0014 4.64 Near infrared λ OLI = 0.0077 + 0.9644 MSI 65,380,148 −0.0003 −0.22 (~0.85 µm) MSI Band 8A MSI = 0.0147 + 0.9355 OLI 0.9022 (<0.0001)

2.7. Normalized Difference Vegetation Index (NDVI) The NDVI is an important parameter that reflects vegetation coverage and can be calculated for Landsat datasets using the equation [37]:

R − R NDVI = NIR R , (7) R + R NIR R Water 2021, 13, x FOR PEER REVIEW 8 of 20

Table 3. Transformation functions between the Sentinel-2 MSI and Landsat-8 OLI developed through OLS regression (Zhang et al. [36]).

Mean Relative Mean OLI—MSI Band Transformation Function Sample Size R2 (p Value) OLI—MSI Differ- Difference ence (%) OLI = 0.0003 + 0.9570 MSI 0.8980 Blue λ (~0.48 μm) 65 347 909 −0.0014 −4.64 MSI = 0.0039 + 0.9383 OLI (<0.0001) Near infrared λ (~0.85 OLI = 0.0077 + 0.9644 MSI 0.9022 65 380 148 −0.0003 −0.22 μm) MSI Band 8A MSI = 0.0147 + 0.9355 OLI (<0.0001)

2.7. Normalized Difference Vegetation Index (NDVI) The NDVI is an important parameter that reflects vegetation coverage and can be calculated for Landsat datasets using the equation [37]:

𝑅 −𝑅 NDVI = , (7) 𝑅 +𝑅 Water 2021, 13, 2373 8 of 20 where 𝑅 and 𝑅 are the reflectance values in the NIR and red bands, respectively.

3. Results where R and R are the reflectance values in the NIR and red bands, respectively. 3.1. HydrologyNIR in the StudyR Area The3. Results changes of discharge and precipitation in the Padma River Basin during the 3.1. Hydrology in the Study Area study period are shown in Figure 4. The discharge result is calculated from GloFAS data, and the precipitationThe changes of dischargechange result and precipitation is calculated in the from Padma GPCP River Basindata, during which the is study the total pre- period are shown in Figure4. The discharge result is calculated from GloFAS data, and the cipitationprecipitation in the changebasin. resultFrom is May, calculated precipitation from GPCP data,and whichdischarge is the totalbegins precipitation to increase, in and be- gins tothe decrease basin. From after May, October. precipitation From and the discharge change begins of monthly to increase, average and begins results, to decrease the change of dischargeafter October. lags the From precipitation. the change of The monthly time-series average results,data for the the change discharge of discharge and lagsprecipitation in thethe Padma precipitation. River were The time-series analyzed data using for thethe discharge M–K test. and The precipitation Z value and in the its Padma rate of change River were analyzed using the M–K test. The Z value and its rate of change were obtained were respectivelyobtained respectively (discharge: Z = (discharge: 0.5730, slope =Z 1.4077 = 0.5730,× 108, slopep > 0.5; = precipitation: 1.4077 × 108,Z = p− >1.2784, 0.5; precipita- tion: slopeZ = − =1.2784, 0.0375, slopep > 0.5). = The0.0375, results p > show 0.5). that The there results is no obviousshow that trend there in discharge is no obvious and trend in dischargeprecipitation and in precipitation the past 30 years. in the past 30 years.

Figure 4. ChangesFigure 4. ofChanges Padma of River Padma discharge River discharge and precipitation and precipitation in the in basin the basin during during the the study study period. period. The The figures figures on the left and right onshow the leftthe andaverage right showannual the discharge average annual and dischargeprecipitat andion precipitation and the average and the monthly average monthly discharge discharge and precipitation, and respectively.precipitation, respectively.

3.2. Comparison of Atmospheric Correction Results between Landsat and Sentinel 3.2. Comparison of Atmospheric Correction Results between Landsat and Sentinel The NIR-band/blue-band ratio was calculated for points randomly selected within Thethe river NIR-band/blue-band area, and the results were ratio then was compared. calculated Similar for ratios points (R2 =randomly 0.9563, root-mean- selected within the riversquare area, error and (RMSE) the results = 0.0811 were in the then 2020 image.compared.R2 = 0.8963, Similar RMSE ratios = 0.0440 (R2 = in0.9563, the 2019 root-mean- 2 squareimage. errorR (RMSE)= 0.9461, RMSE= 0.0811 = 0.0596 in the in total)2020 wereimage. obtained R2 = based0.8963, on theRMSE data = acquired 0.0440by in the 2019 the two2 satellites (as shown in Figure5), suggesting that the atmospherically corrected data image.were R accurate= 0.9461, and RMSE reliable. = 0.0596 in total) were obtained based on the data acquired by

Water 2021, 13, x FOR PEER REVIEW 9 of 20

Water 2021, 13, 2373 9 of 20 the two satellites (as shown in Figure 5), suggesting that the atmospherically corrected data were accurate and reliable.

FigureFigure 5. Comparison of of the the NIR-band/blue-band NIR-band/blue-band rati ratiosos between between the theLandsat Landsat and andSentinel Sentinel data. data.

3.3.3.3. Distribution of of C CTSMTSM inin the the Padma Padma River River

The arithmetic mean mean CCTSMTSM waswas calculated calculated for for each each season season of each of each year year in the in theperiod period 1991–2019.1991–2019. OnOn thisthis basis,basis,the the seasonal seasonal distribution distribution of ofC TSMCTSMin in the the Padma Padma River River was was obtained, ob- astained, shown as shown in Figure in 6Figure. In terms 6. In terms of the of distribution the distribution trend, trend,CTSM CTSMin in the the Padma Padma River River was highwas high in areas in areas close close to to the the banks banks and and in in thethe downstream region region and and low low in areas in areas away away fromfrom thethe banksbanks and in in the the upstream upstream regi region.on. The The waters waters with with relatively relatively high high CTSMCTSM valuesvalues werewere distributed near near the the estuary estuary in inthe the downstream downstream region, region, which which was primarily was primarily due to due tothe the water water and andsoil loss soil on loss the on islands the islands near the near mouth the of mouth the estuary of the caused estuary by the caused scouring by the scouringof the water of the flow water in the flow river in as the well river as asthe well convergence as the convergence and resuspension and resuspension of the matter of the matter(e.g., sediment) (e.g., sediment) carried carriedhere by herethe water by the flow water in the flow river inthe from river the fromupstream the upstream region. At region. a seasonal scale, CTSM was the lowest in winter (December through February of the follow- At a seasonal scale, CTSM was the lowest in winter (December through February of the ing year). The CTSM remained below 150 mg/L in most parts of the river but reached up to following year). The CTSM remained below 150 mg/L in most parts of the river but reached up180 to mg/L 180 in mg/L some in tributaries some tributaries and 200 mg/L and 200 in the mg/L downstream in the downstream region. There region. was a Therecertain was increase in CTSM in spring (March through May). Compared to winter, CTSM in spring ex- a certain increase in CTSM in spring (March through May). Compared to winter, CTSM in hibited an overall similar distribution trend but was higher in magnitude. The study area spring exhibited an overall similar distribution trend but was higher in magnitude. The has a tropical monsoon climate with abundant precipitation in summer. Due to heavy study area has a tropical monsoon climate with abundant precipitation in summer. Due to precipitation and a large discharge, CTSM displayed a relatively uniform distribution pat- heavy precipitation and a large discharge, C displayed a relatively uniform distribution tern and, overall, remained at a relatively highTSM level in summer (June through August). pattern and, overall, remained at a relatively high level in summer (June through August). As a result of the soil erosion caused by precipitation and high flow rates, high CTSM values As a result of the soil erosion caused by precipitation and high flow rates, high CTSM values were distributed along both banks of the river and in the inland tributaries. The CTSM in were distributed along both banks of the river and in the inland tributaries. The C in the river was almost above 150 mg/L. In addition, due to an increase in the discharge, theTSM theerosion river and was siltation almost zone above at 150the mouth mg/L. of In the addition, estuary dueshifted to anoutward. increase Similar in the to discharge, summer, the erosion and siltation zone at the mouth of the estuary shifted outward. Similar to summer, CTSM in fall (September through November) remained above 150 mg/L in the main parts C in fall (September through November) remained above 150 mg/L in the main parts ofTSM the river, and high-CTSM zones shifted downstream. of the river, and high-CTSM zones shifted downstream.

Water 2021, 13, x FOR PEER REVIEW 10 of 20

Water 2021, 13, 2373 10 of 20

Figure 6. Distribution of CTSM in the Padma River in: (a) winter, (b) spring, (c) summer, and (d) fall.

3.4. Interannual Variations of CTSM in the Padma River Figure 6. Distribution of CTSM in the Padma River in: (a) winter, (b) spring, (c) summer, and (d) fall. After the waters were separated from the land in the images, CTSM in the waters was estimated using the algorithm equation introduced in Section 2.4. The arithmetic means of 3.4. Interannualthe pixels in Variations each image of of C theTSM section in the showed Padma in River Figure 2 was calculated to represent the mean CTSM. After the waters were separated from the land in the images, CTSM in the waters was We also used Sentinel-2 data to calculate CTSM from 2016 to 2019 based on the same estimatedalgorithm. using In the order algorithm to ensure equation the consistency introduced of the algorithm, in Section the 2.4. reflectance The arithmetic data of means of theSentinel-2 pixels in were each converted image toof thethe reflectance section sh bandsowed of Landsat-8in Figure OLI 2 was by using calculated the equation to represent the meanin Table CTSM3,. and then converted to the reflectance bands of Landsat-7 ETM+ according to the formula in Table2. Finally, the CTSM of Sentinel-2 sensor was calculated through We also used Sentinel-2 data to calculate CTSM from 2016 to 2019 based on the same Equation (1). The results calculated by Sentinel-2 data with higher temporal and spatial algorithm.resolution In order are similar to ensure to those the calculated consistency using Landsatof the algorithm, data, which the shows reflectance that the atmo- data of Sen- tinel-2spheric were correctionconverted and toC TSMthe algorithmreflectance have band goods accuracy. of Landsat-8 Here, there OLI is by a little using systematic the equation in Tablebias 3, and between then the converted two satellites, to the which reflectance is probably bands caused byof theLandsat-7 reason that ETM+ the BNIR according/Bblue to the value of Sentinel-2 is slightly higher than that of Landsat (see in Figure5). However, from formula in Table 2. Finally, the CTSM of Sentinel-2 sensor was calculated through Equation the perspective of the overall change trend, we still believe that the results of the two are (1). Thesimilar. results Therefore, calculated due to by the Sentinel-2 advantages data of Landsat with datahigher quantities temporal and theand fact spatial that the resolution are similar to those calculated using Landsat data, which shows that the atmospheric cor- rection and CTSM algorithm have good accuracy. Here, there is a little systematic bias be- tween the two satellites, which is probably caused by the reason that the BNIR/Bblue value of Sentinel-2 is slightly higher than that of Landsat (see in Figure 5). However, from the perspective of the overall change trend, we still believe that the results of the two are sim- ilar. Therefore, due to the advantages of Landsat data quantities and the fact that the bi- ases are within the acceptable range, we believe that Landsat satellites can be used to eval- uate the flux of total suspended matter in the Padma River in the past 30 years.

Water 2021, 13, x FOR PEER REVIEW 11 of 20

Figure 7 shows the changes in the mean CTSM in the Padma River in the period 1991– 2019. During this period, the maximum and minimum CTSM values (194.56 ± 54 and 58.97 ± 16 mg/L, respectively) occurred in 2011 and 2015, respectively. Overall, CTSM displayed Water 2021, 13, 2373 11 of 20 a downward trend in this period. Especially, the strongest global La Niña event in the past six decades occurred in 2011 and 2012. In April of 2011, intense cyclones struck Bangladesh and brought heavy rainfall. biasesIn July are of within2011, continuous the acceptable heavy range, precipitation we believe triggered that Landsat floods satellites and forced can bemore used than to evaluate10,000 people the flux to ofevacuate. total suspended The flood matter caused in by the La Padma Niña lasted River inunti thel the past summer 30 years. of 2012. The Figurefloods 7caused shows a the large changes amount in theof SM mean fromCTSM terrestrialin the Padma sources River to rush in the into period the Padma 1991– 2019River,. Duringresulting this in an period, abnormally the maximum elevated andlevel minimum of turbidityC TSMand anvalues abnormally (194.56 high± 54 mean and 58.97CTSM in± 201116 mg/L and, 2012. respectively) At the end occurred of 2012, in th 2011e impact and caused 2015, respectively. by La Niña weakened, Overall, C TSMand displayedthe CTSM level a downward returned to trend the innormal this period. trend in 2013.

FigureFigure 7.7. ChangesChanges inin CCTSMTSM inin the the Padma Padma River River in the period 1991–2019, andand thethe redred areaarea representsrepresents thethe durationduration ofof LaLa Niña.Niña.

3.5. ChangesEspecially, in F theTSM strongestin the Padma global River La Niña event in the past six decades occurred in 2011 and 2012.Due to In the April high of 2011,flow intenserate and cyclones large discha struckrge Bangladesh in the Padma and River, brought we heavy can approxi- rainfall. Inmately July ofconsider 2011, continuousthat the TSM heavy in the precipitation Padma River triggered was sufficiently floods andmixed. forced In addition, more than the 10,000non-tidal people reaches to evacuate. were selected The flood in this caused study by for La analysis, Niña lasted and, untiltherefore, the summer the tidal of effects 2012. Thedo not floods need caused to be aconsidered. large amount Then, of SMwe fromcan approximately terrestrial sources consider to rush that into the the TSM Padma was River, resulting in an abnormally elevated level of turbidity and an abnormally high mean uniformly distributed in the Padma River. The monthly mean FTSM was calculated through C in 2011 and 2012. At the end of 2012, the impact caused by La Niña weakened, and theTSM multiplication of the arithmetic mean CTSM at the selected section (23°10′44″ N) esti- thematedCTSM fromlevel the returned Landsat to data the by normal the arithmetic trend in 2013. mean discharge at the selected section. On this basis, subsequent calculations were performed. Figure 8 shows the changes in FTSM in 3.5. Changes in FTSM in the Padma River the Padma River in the period 1991–2019. The maximum and minimum FTSM values (318 Due to the high flow rate and large discharge in the Padma River, we can approx- ± 62 and 73 ± 29 mt, respectively) occurred in 2011 and 2015, respectively. Due to the im- imately consider that the TSM in the Padma River was sufficiently mixed. In addition, pact of the 1998 flood, the discharge and CTSM were at a high level, so the FTSM increased the non-tidal reaches were selected in this study for analysis, and, therefore, the tidal significantly and returned to the normal level in 2000. At the same time, the low flow effects do not need to be considered. Then, we can approximately consider that the TSM caused by the drought in 2006 also continued to reduce FTSM and restored in the next year. was uniformly distributed in the Padma River. The monthly mean FTSM was calculated In 2011, due to the flood disaster caused by La Niña, the discharge and CTSM◦ in 0Padma00 through the multiplication of the arithmetic mean CTSM at the selected section (23 10 44 N) River increased significantly, resulting in a significant increase in FTSM level. In 2012, the estimated from the Landsat data by the arithmetic mean discharge at the selected section. impact of La Niña continued. Although the discharge decreased, it remained at a high On this basis, subsequent calculations were performed. Figure8 shows the changes in FTSM level. At the same time, the erosion of flood on land did not reduce CTSM, so FTSM was still in the Padma River in the period 1991–2019. The maximum and minimum FTSM values at a high level. In 2013, the flood disaster was over, and CTSM returned to normal level; FTSM (318 ± 62 and 73 ± 29 mt, respectively) occurred in 2011 and 2015, respectively. Due to the also decreased. The time-series data for the changes in FTSM were analyzed using the M–K impact of the 1998 flood, the discharge and CTSM were at a high level, so the FTSM increased 5 significantlytest. The Z value and and returned its rate to of the change normal were level obtained in 2000. (Z At = − the2.6073, same slope time, = − the33.27 low × 10 flow, p caused by the drought in 2006 also continued to reduce FTSM and restored in the next year. In 2011, due to the flood disaster caused by La Niña, the discharge and CTSM in Padma River increased significantly, resulting in a significant increase in FTSM level. In 2012, the impact of La Niña continued. Although the discharge decreased, it remained at a high level. At the same time, the erosion of flood on land did not reduce CTSM, so FTSM was still at a high level. In 2013, the flood disaster was over, and CTSM returned to normal level; FTSM Water 2021, 13, 2373 12 of 20

Water 2021, 13, x FOR PEER REVIEW 12 of 20

also decreased. The time-series data for the changes in FTSM were analyzed using the M–K test. The Z value and its rate of change were obtained (Z = −2.6073, slope = −33.27 × 105, < 0.01).p < 0.01). In the In past the past30 years, 30 years, FTSMF inTSM thein Padma the Padma River River showed showed a relatively a relatively significant significant down- warddownward trend and, trend on and, average, on average, decreased decreased by 3.3 by mt 3.3 each mt eachyear. year. The Theresult result passed passed the the signifi- cancesignificance test at a testconfidence at a confidence level of level 0.01. of 0.01.

400 mt

350

300

250

200 TSM flux 150

100

50

0 1990 1995 2000 2005 2010 2015 2020 year

FigureFigure 8. Changes 8. Changes in F inTSMFTSM in thein the Padma Padma River River in in the the period period 1991–2019. The The dotted dotted line line is theis the linear linear regression regression of FTSM of .FTSM.

3.6.3.6. Changes Changes of ofNDVI NDVI in in the the Padma Padma River TheThe NDVI NDVI for for the the Padma River River Basin Basin was was calculated calculated based based on the on Moderate the Moderate Resolution Resolu- Imaging Spectroradiometer (MODIS) global monthly mean NDVI products. The resolution tion Imaging Spectroradiometer (MODIS) global monthly mean NDVI products. The res- of MODIS NDVI product is 1 km, and the selected coverage covers the whole Padma River olutionBasin. of The MODIS monthly NDVI mean product NDVI images is 1 km, for theand period the selected 2000–2019 coverage were selected. covers On the the whole Padmaother River hand, Basin. NDVI alongThe monthly the Padma mean river isNDVI calculated images from for Landsat the period images. 2000–2019 As shown were in se- lected.Figures On9 the and other 10, overall, hand, the NDVI NDVI along for the the Padma Padma River river Basin is calculated displayed from an upward Landsat trend, images. As suggestingshown in Figures a notable 9 increaseand 10, inoverall, the vegetation the NDVI coverage for the alongPadma the River banks Basin of the displayed Padma an upwardRiver. trend, This suggests suggesting that the a notable significant increase increase in in the the vegetation coverage coverage in along the basin the of banks of thethe Padma RiverRiver. increased This suggests the water that and the soil significant retention capacityincrease ofin the the land, vegetation reduced coverage the in theextent basin ofsoil of the erosion, Padma and, River to a certainincreased extent, the reduced water and the dischargesoil retention of SM capacity from terrestrial of the land, reducedsources, the thereby extent making of soil someerosion, contribution and, to a to certain the decrease extent, in reduced the content the of discharge TSM in the of SM Padma River. from terrestrial sources, thereby making some contribution to the decrease in the content of TSM in the Padma River.

Water 2021, 13, 2373 13 of 20 Water 2021, 13, x FOR PEER REVIEW 13 of 20

Water 2021, 13, x FOR PEER REVIEW 13 of 20

FigureFigure 9. Mean NDVI NDVI values values from from MODIS MODIS for for th thee Ganges–Brahmaputra Ganges–Brahmaputra River River Basin. Basin. (a) ( aMean) Mean NDVI NDVI Figurevalue for 9. Meanthe river NDVI basin values in 2000. from (b MODIS) Mean NDVIfor the valueGanges for– Brahmaputrathe river basin River in 2019. Basin. (c) (Changesa) Mean NDVIin the value for the river basin in 2000. (b) Mean NDVI value for the river basin in 2019. (c) Changes in the valuemean forNDVI the valueriver basinbetween in 2000. 2000 (andb) Mean 2019. NDVI value for the river basin in 2019. (c) Changes in the meanmean NDVI value between 2000 2000 and and 2019. 2019.

Figure 10. Changes in the average NDVI value for the Landsat image as well as the CTSM at the TSM FigureFiguremouth 10.of10. the Changes river in in inthe the period average average 2000–2019. NDVI NDVI value value for for the the Landsat Landsat image image as as well well as as the the CCTSM at atthe the mouthmouth of the river in the period period 2000 2000–2019.–2019.

WaterWater2021 2021, ,13 13,, 2373 x FOR PEER REVIEW 1414 of 2020

4.4. DiscussionDiscussion 4.1.4.1. NaturalNatural FactorsFactors AffectingAffecting thethe ChangesChanges inin thethe TSMTSM inin thethe PadmaPadma RiverRiver

ChangesChanges in FTSMTSM dependdepend on on changes changes in inthe the discharge discharge and and CTSMC. TSMThe. discharge The discharge in the inPadma the Padma River depends River depends primarily primarily on natural on naturalfactors factorssuch as suchthe precipitation as the precipitation and ground- and groundwaterwater resources resources within withinthe river the basin river and basin water and watersources sources (e.g., (e.g.,freshwater freshwater released released from fromthe melting the melting of ice of on ice the on Tibetan the Tibetan Plateau). Plateau). The Thetime-series time-series data data for the for thedischarge discharge in the in thePadma Padma River River were were analyzed analyzed using using the M–K the M–K test. test.The TheZ valueZ value and its and rate its of rate change of change were 8 wereobtained obtained (Z = 0.5730, (Z = 0.5730, slope = slope 1.4077 = 1.4077× 108, p× > 100.5)., pThe>0.5). results The showed results that showed the discharge that the dischargeitself displayed itself displayed no notable no trend notable in the trend past in 30 the years. past 30In years.addition, In addition,the correlation the correlation between betweenthe interannual the interannual changes in changes the discharge in the discharge and FTSM andwas Fanalyzed.TSM was analyzed.According According to the results to theshown results in Figure shown 11, in Figurethere was 11, little there significant was little significant correlation correlation between the between changes the in changes the dis- incharge the discharge and FTSM and(R = F0.44,TSM p(R < =0.05). 0.44, Hence,p < 0.05). the Hence,changes the in changesthe discharge in the might discharge have might had a havelimited had impact a limited on F impactTSM. on FTSM.

Figure 11. Correlation between the changes in the discharge and FTSM in the Padma River. Figure 11. Correlation between the changes in the discharge and FTSM in the Padma River.

AtAt thethe samesame time,time, thethe discharge maymay affect the degree of bank erosion of Padma river. BankBank erosionerosion isis partpart ofof thethe riverriver system’ssystem’s naturalnatural disruptiondisruption [[38].38]. TheThe erosionerosion ofof PadmaPadma RiverRiver isis highlyhighly irregularirregular [[39].39]. AnupomAnupom HalderHalder etet al.al. (2021)(2021) studiedstudied thethe changechange ofof PadmaPadma riverriver bankbank inin thethe pastpast 4040 yearsyears byby usingusing RSRS andand GISGIS techniques.techniques. TheThe resultsresults showedshowed aa linearlinear relationshiprelationship betweenbetween thethe erosionerosion and and discharge discharge ( R(R22= = 0.9989)0.9989) [[40].40]. FromFrom 19891989 toto 1999,1999, charchar landland ofof thethe PadmaPadma RiverRiver continuouslycontinuously increasedincreased (see(see FigureFigure 12 12).). TheThe islandisland barsbars werewere mostmost vulnerablevulnerable duringduring flooding.flooding. CombinedCombined withwith thethe changechange analysisanalysis ofofC CTSMTSM, itit cancan bebe seenseen thatthat charchar land in in Padma Padma River River is is in in a acontinuous continuous growth growth state state before before 2010, 2010, and and CTSMC TSMis alsois alsoat a stable at a stable high high concentration concentration level. level. Char Char land landin Padma in Padma stops stops growing growing from from2010 2010to 2019, to 2019,and C andTSM beginsCTSM begins to gradually to gradually decrease. decrease. Unstable Unstable char land char bar land is bara major is a major source source of TSM, of TSM,and the and area the of area unstable of unstable riparian riparian lines determines lines determines the change the change of CTSM of. Therefore,CTSM. Therefore, during duringthe study the period, study period, although although the discharge the discharge itself has itself not has changed not changed significantly, significantly, the bank the bankerosion erosion of Padma of Padma River River caused caused by large by large discharg dischargee and the and change the change of fragile of fragile char charland land area areacaused caused by the by change the change of bank of bank line lineindirectly indirectly affect affect the change the change of C ofTSMC ofTSM Padmaof Padma River, River, thus thusaffecting affecting the transportation the transportation of FTSM of .F TSM.

Water 2021, 13, 2373 15 of 20 Water 2021, 13, x FOR PEER REVIEW 15 of 20

FigureFigure 12.12. CharChar areaarea changechange ofof thethe PadmaPadma RiverRiver [[40].40].

4.2.4.2. HumanHuman FactorsFactors AffectingAffecting ChangesChanges inin thethe TSMTSM inin thethe PadmaPadma RiverRiver WhileWhile neglectingneglecting thethe changeschanges inin thethe discharge,discharge, wewe performedperformed aa correlationcorrelation analysisanalysis on the annual mean changes in C and F (see Figure 13). A significant correlation on the annual mean changes in CTSM and FTSMTSM (see Figure 13). A significant correlation (R (R = 0.76, p < 0.01) was found between the changes in C and F , suggesting that the = 0.76, p < 0.01) was found between the changes in CTSMTSM and FTSMTSM, suggesting that the changes in F in the Padma River depended heavily on the changes in C . changes in FTSMTSM in the Padma River depended heavily on the changes in CTSMTSM.

TSM TSM FigureFigure 13.13. CorrelationCorrelation betweenbetween thethe changeschanges inin CCTSMTSM andand FFTSMTSM inin the the Padma Padma River. River.

In the Padma River Basin, the changes in CTSM were primarily a result of the changes in the discharge of TSM from terrestrial sources, including the discharge due to soil

Water 2021, 13, x FOR PEER REVIEW 16 of 20

Water 2021, 13, 2373 16 of 20

In the Padma River Basin, the changes in CTSM were primarily a result of the changes in the discharge of TSM from terrestrial sources, including the discharge due to soil ero- sionerosion [41]. [41 As]. no As large no large reservoirs reservoirs were were cons constructedtructed in the in Ganges–Brahmaputra the Ganges–Brahmaputra river river sys- TSM tem,system, the theimpact impact of water of water conservancy conservancy facilities facilities on F on F canTSM becan excluded. be excluded. In addition, In addition, rela- TSM tivelyrelatively large large discharges discharges may maylead leadto soil to erosion soil erosion and have and a have greater a greater impact impact on C on. ThisCTSM is. particularlyThis is particularly true for true the Padma for the PadmaRiver—a River—a river with river highly with unstable highly unstable shorelines. shorelines. The Padma The RiverPadma is Riverrelatively is relatively wide, with wide, severe with bank severe erosion bank erosion near Harirampur near Harirampur in the indistrict the district of Man- of ikganjManikganj on its on left its leftbank bank and and Naria Upazila in inthe the district district of of Shariatpur Shariatpur on on its its right right bank. bank. Erosion on the left and right banks of the Padma River occurs at annual rates of 4.82 and 8.19 km 2,, respectively respectively [42]. [42]. Substitution of of FFTSMTSM seriesseries into into the the M–K M–K abrupt-change abrupt-change test test equation equation yielded yielded UFk UFandk UBandk UBvalues,k values, as shown as shown in Figure in Figure 14. Th 14e. point The point of intersection of intersection between between the UF thek UFandk andUBk curvesUBk curves correspond correspond to the to abrupt-change the abrupt-change point point of the of time the time series. series. The Theabrupt abrupt change change oc- curredoccurred in 2014. in 2014. The TheUFk valueUFk value fluctuates fluctuates near 0 near before 0 beforethe point the of point intersection of intersection and remains and withinremains the within a = 0.05 the straighta = 0.05 line, straight suggesting line, suggesting that before that 2014, before the 2014,Padma the River Padma was River rela- tivelywas relatively stable and stable its F andTSM exhibited its FTSM exhibited no notable no trend notable of change. trend of An change. abrupt An change abrupt occurred change inoccurred 2014, when in 2014, FTSM when beganFTSM to begandecrease. todecrease. This is becaus This ise, because,around 2014, around the 2014, government the govern- of Bangladeshment of Bangladesh began to beganimplement to implement a series of a seriesenvironmental of environmental protection protection measures, measures, adopted aadopted sustainable a sustainable development development strategy, strategy,and partic andipated participated in the Blue in theEconomic Blue Economic Zone Cooper- Zone ationCooperation in the Bay in theof Bengal Bay of [43]. Bengal In addition [43]. In, addition, the government the government of Bangladesh of Bangladesh and non-gov- and ernmentalnon-governmental organizations organizations vigorously vigorously carried out carried afforestation out afforestation projects along projects the alongcoast, there- sultingcoast, resulting in a significant in a significant increase increasein the green in the land green area. land Vegetation area. Vegetation planting, planting, to a certain to a extent,certain reduced extent, reduced soil erosion soil in erosion the coastal in the zone coastals, thereby zones, reducing thereby reducingthe content the of content TSM [44]. of ToTSM validate [44]. To this validate hypothesis, this hypothesis, the NDVI thewas NDVI analyzed. was analyzed.

Figure 14. M–K abrupt-change test results for FTSM in the Padma River. The blue UFk curve and red UBk curve signify the Figure 14. M–K abrupt-change test results for FTSM in the Padma River. The blue UFk curve and red UBk curve signify the sequential and reversed time series for FTSM, respectively. The critical values for the significance levels a = 0.05 and 0.01 sequential and reversed time series for FTSM, respectively. The critical values for the significance levels a = 0.05 and 0.01 (U0.05 and U0.01, respectively) are ±1.96 and ± 2.32, respectively. (U0.05 and U0.01, respectively) are ±1.96 and ±2.32, respectively.

Water 2021, 13, x FOR PEER REVIEW 17 of 20

Water 2021, 13, 2373 17 of 20 4.3. Effects of the NDVI on the Changes in CTSM Mithun Kumar et al. (2021), in their monitoring and mapping of forest, cover changes in eastern Sundarban based on satellite remote sensing technology using the maximum 4.3. Effects of the NDVI on the Changes in CTSM likelihood classification approach [45]. The classification map shows that, from 1989 to Mithun Kumar et al. (2021), in their monitoring and mapping of forest, cover changes in2014, eastern the Sundarbanvegetation based coverage on satellite in the remotestudy sensingarea decreased technology by using2.66%, the while, maximum from 2014 to likelihood2019, the classificationvegetation coverage approach increased [45]. The classificationby 2.22%. Water map shows body that,and frommudfat, 1989 low to laying 2014,area, theas well vegetation as intertidal coverage zone in thehave study been area increased decreased with by 2.66%,the decrease while, fromin vegetation 2014 to cover. 2019,Using the Landsat vegetation series coverage satellite increased images by separat 2.22%.ed Water by water body and and mudfat, land, the low NDVI laying mean area, of land asin wellthe asimage intertidal frame zone is calculated have been increased to obtain with the the change decrease of average in vegetation NDVI. cover. The Using trend of the Landsataverage series NDVI satellite in image images was separated opposite by to water that andof C land,TSM (correlation the NDVI mean coefficient of land inR the= −0.79, p < image0.01) (see frame Figure is calculated 15), indicating to obtain a the significan change oftly average negative NDVI. correlation The trend between of the average the NDVI and NDVI in image was opposite to that of CTSM (correlation coefficient R = −0.79, p < 0.01) CTSM. The results show that, from 2000 to 2019, under the background of the overall NDVI (see Figure 15), indicating a significantly negative correlation between the NDVI and CTSM. Thevalue results rising show in that,the basin, from 2000 the toNDVI 2019, level under along the background the Padma of theRiver overall also NDVI shows value an upward risingtrend. in At the the basin, same the time, NDVI the level downward along the trend Padma before River 2014 also and shows the anupward upward trend trend. after 2014 Atcan the also same be time,clearly the seen. downward The research trend before area of 2014 this and study the upwardoverlaps trend with after that 2014 of Mithun can et al. also(2021) be To clearly a certain seen. Theextent. research Therefore, area of the this si studymilar overlapsconclusions with of that the of two Mithun can etalso al. support (2021)the results To a certain of this extent. study. Therefore, It should thebe noted similar that conclusions the vegetation of the two coverage can also is support not completely therelated results to ofthe this NDVI study. index, It should because be noted in thatthe st theudy vegetation range, more coverage vegetation is not completely in Padma River relatedbasin is to transformed the NDVI index, into because water inrather the study than range,bare land more due vegetation to land in erosion. Padma River At the same basintime, isas transformed mentioned intoabove, water in the rather last than decade, bare due land to due environmental to land erosion. protection At the same policies and time, as mentioned above, in the last decade, due to environmental protection policies policies to encourage planting, the level of NDVI has increased and the land retention has and policies to encourage planting, the level of NDVI has increased and the land retention hasimproved. improved. The The improvement improvement of of land land conservation conservation also also slows slows down down the the generation generation of of frag- fragileile bar bar lines, lines, so so it it also also slowsslows down down the the growth growth of char of char land area.land Thus,area. theThus, land-based the land-based emissionemission of of TSM TSM and andCTSM CTSMin in Padma Padma River River are reduced.are reduced. In conclusion, In conclusion, the vegetation the vegetation cover cover levellevel and and NDVI NDVI of of the the Padma Padma River River Basin, Basin, especially especially the coastal the coastal land area, land will area, affect will the affect the landland conservation conservation capacity, capacity, thus thus indirectly indirectly affecting affecting the FTSM thetransport FTSM transport of the Padma of the River.Padma River.

FigureFigure 15. 15.Correlation Correlation between between the meanthe mean NDVI NDVI value value for the for Landsat the Landsat image and imageCTSM andat the CTSM mouth at the mouth ofof thethe river river in in the the period period 2000–2019. 2000–2019.

5. Conclusions 5. Conclusions In this study, CTSM in the Padma River Basin was estimated from the Landsat-8 OLI/TIRSIn this and study, Landsat-5 CTSM TM in L1 the reflectance Padma products River Basin for the was period estimated 1991–2019. from In addition, the Landsat-8 OLI/TIRS and Landsat-5 TM L1 reflectance products for the period 1991–2019. In addition, the causes of the changes in CTSM were analyzed based on the discharge, precipitation, the causes of the changes in CTSM were analyzed based on the discharge, precipitation, and

Water 2021, 13, 2373 18 of 20

and NDVI data for the same time period. The main conclusions of this study can be summarized as follows: (1) The CTSM values in the Padma River were high in areas near the banks and in the downstream region and low in areas away from the banks and in the upstream region. The waters with relatively high CTSM values were distributed in the downstream region near the mouth of the river. Across each year, CTSM was the lowest in winter and spring and the highest in summer and fall. (2) In the period 1991–2019, the maximum and minimum mean FTSM values in the Padma River (318 ± 62 and 73 ± 29 mt, respectively) occurred in 2011 and 2015, respectively. Abnormally high FTSM values occurred in the year with a strong La Niña event. Overall, FTSM displayed a downward trend and decreased at an annual average rate of 3.3 mt. (3) The FTSM values in the Padma River depended collectively on the discharge and CTSM at the mouth. Compared with natural factors (e.g., discharge) (R = 0.44, p < 0.05), the changes in CTSM contributed more significantly to the changes in FTSM in the Padma River (R = 0.76, p < 0.01). A significantly negative correlation was found between the NDVI and CTSM (R = −0.79, p < 0.01), suggesting that the increase in the NDVI within the Ganges–Brahmaputra River Basin and the coastal bar line in the last decades led to a decrease in the discharge of TSM from terrestrial sources and CTSM, and that changes in human factors (e.g., vegetation changes) negatively contributed to FTSM in the Padma River. Hence, human activities were the dominant factor causing the changes in FTSM in the Padma River (Ganges–Brahmaputra River system). This study ensured the accuracy of atmospheric correction of satellite data and the accuracy of various algorithms. However, due to the climatic characteristics of the ge- ographical location of the study area, it is difficult to obtain cloudless and high-quality remote sensing images, so the sufficiency of the amount of remote sensing data cannot be guaranteed. There may not be data available for the entire quarter in summer and autumn in some years. In this case, after calculating the TSM concentration of the seasons with data, we use the following method to make up for the gap in the cloudy season data: Take the multi-year arithmetic average of all TSM data in each season to obtain the CTSM of the four seasons during the 30 years. Then the climatic distribution of each grid point is calculated, and the ratio of the four seasonal climatic CTSM at each grid point is calculated. On this basis, the CTSM of the missing season is calculated based on this ratio. Due to the numerical averaging of the climatic data, although the absolute value of the data may not be so accurate, it should be of good reference value when estimating the trend of the FTSM changes. Because there are few hydrological data in the study area, and because they are difficult to obtain, this study can provide ideas for hydrological research in similar areas.

Author Contributions: Conceptualization, X.H. and D.W.; methodology, Z.Z.; software, validation, Z.Z.; formal analysis, Z.Z. and F.G.; resources, F.G. and D.W.; writing—original draft preparation, Z.Z.; writing—review and editing, D.W., X.H. and Y.B.; supervision, project administration and funding acquisition, D.W. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the National Key R&D Program of China under Grant No. 2018YFB0505005 and 2017YFC1405300, the Key Research and Development Plan of Zhejiang Province under contract No. 2017C03037, the National Natural Science Foundation of China under contract No. 41476157, and the Marine Science and Technology Cooperation Project between Maritime Silk Route and island countries based on marine sustainability. Acknowledgments: The authors would like to thank the USGS for excellent Landsat and Sentinel data. We also thank the satellite ground station and the satellite data processing and sharing center of SOED/SIO for help with the data processing. Our deepest gratitude goes to the editors and reviewers for their careful work and thoughtful suggestions. Conflicts of Interest: The authors declare no conflict of interest. Water 2021, 13, 2373 19 of 20

References 1. Ouillon, S.; Douillet, P.; Andréfouët, S. Coupling satellite data with in situ measurements and numerical modeling to study fine suspended-sediment transport: A study for the lagoon of New Caledonia. Coral Reefs 2004, 23, 109–122. 2. Chacko, N.; Jayaram, C. Variability of total suspended matter in the northern coastal Bay of Bengal as observed from satellite data. J. Indian Soc. Remote Sens. 2017, 45, 1077–1083. [CrossRef] 3. Wilber, D.H.; Clarke, D.G. Biological effects of suspended sediments: A review of suspended sediment impacts on fish and shellfish with relation to dredging activities in estuaries. N. Am. J. Fish. Manag. 2001, 21, 855–875. [CrossRef] 4. Cyrus, D.P.; Blaber, S.J.M. The influence of turbidity on juvenile marine fishes in estuaries. Part 1. Field studies at Lake St. Lucia on the southeastern coast of Africa. J. Exp. Mar. Biol. Ecol. 1987, 109, 53–70. [CrossRef] 5. Cyrus, D.P.; Blaber, S.J.M. The influence of turbidity on juvenile marine fishes in estuaries. Part 2. Laboratory studies, comparisons with field data and conclusions. J. Exp. Mar. Biol. Ecol. 1987, 109, 71–91. [CrossRef] 6. Li, L.; Ni, J.; Chang, F.; Yue, Y.; Frolova, N.; Magritsky, D.; Borthwick, A.G.L.; Ciais, P.; Wang, Y.; Zheng, C.; et al. Global trends in water and sediment fluxes of the world’s large rivers. Sci. Bull. 2020, 65, 62–69. [CrossRef] 7. Milliman, J.D.; Mei-e, R. River flux to the sea: Impact of human intervention on river systems and adjacent coastal areas. In Climate Change; CRC Press: Boca Raton, FL, USA, 2021; pp. 57–83. 8. Li, T.; Wang, S.; Liu, Y.; Fu, B.; Zhao, W. Driving forces and their contribution to the recent decrease in sediment flux to ocean of major rivers in China. Sci. Total Environ. 2018, 634, 534–541. [CrossRef][PubMed] 9. Darby, S.E.; Hackney, C.R.; Leyland, J.; Kummu, M.; Lauri, H.; Parsons, D.R. Fluvial sediment supply to a mega-delta reduced by shifting tropical-cyclone activity. Nature 2016, 539, 276–279. [CrossRef] 10. Portenga, E.W.; Bierman, P.R.; Trodick Jr, C.D.; Greene, S.E.; DeJong, B.D.; Rood, D.H.; Pavich, M.J. Erosion rates and sediment flux within the Potomac River basin quantified over millennial timescales using beryllium isotopes. Bulletin 2019, 131, 1295–1311. [CrossRef] 11. Strick, R.J.; Ashworth, P.J.; Smith, G.H.S.; Nicholas, A.P.; Best, J.L.; Lane, S.N.; Parsons, D.R.; Simpson, C.J.; Unsworth, C.A.; Dale, J. Quantification of bedform dynamics and bedload sediment flux in sandy braided rivers from airborne and satellite imagery. Earth Surf. Process. Landf. 2019, 44, 953–972. [CrossRef] 12. Rahman, M.M.; Goel, N.K.; Arya, D.S. Development of the jamuneswari flood forecasting system: Case study in bangladesh. J. Hydrol. Eng. 2012, 17, 1123–1140. [CrossRef] 13. Gallay, M.; Martinez, J.M.; Mora, A.; Castellano, B.; Yépez, S.; Cochonneau, G.; Laraque, A. Assessing Orinoco river sediment discharge trend using MODIS satellite images. J. S. Am. Earth Sci. 2019, 91, 320–331. [CrossRef] 14. Immerzeel, W.W.; van Beek, L.P.H.; Bierkens, M.F.P. Climate change will affect the Asian water towers. Science 2010, 328, 1382–1385. [CrossRef] 15. Khan, M.Z.H.; Hasan, M.R.; Khan, Z.H.; Aktar, S.; Fatema, K. Distribution of heavy metals in surface sediments of the Bay of Bengal coast. J. Toxicol. 2017, 2017, 9235764. [CrossRef][PubMed] 16. Sarin, M.M.; Krishnaswami, S.; Dilli, K.; Somayajulu, B.L.K.; Moore, W.S. Major ion chemistry of the Ganga–Brahmaputra river system: Weathering processes and fluxes to the Bay of Bengal. Geochim. Cosmochim. Acta 1989, 53, 997–1009. [CrossRef] 17. Mitra, A.; Gangopadhyay, A.; Dube, A.; Schmidt, A.; Banerjee, K. Observed changes in water mass properties in the Indian Sundarbans (Northwestern Bay of Bengal) during 1980–2007. Curr. Sci. 2009, 97, 1445–1452. 18. Jutla, A.S.; Akanda, A.S.; Islam, S. Satellite remote sensing of space-time plankton variability in the Bay of Bengal: Connections to cholera outbreaks. Remote Sens. Environ. 2012, 123, 196–206. [CrossRef] 19. Sridhar, P.N.; Ali, M.M.; Vethamony, P.; Babu, M.T.; Ramana, I.V.; Jayakumar, S. Seasonal occurrence of unique sediment plume in the Bay of Bengal. EOS Trans. 2008, 89, 22–23. [CrossRef] 20. Sasamal, S.K.; Panigrahy, R.C.; Misra, S. Asterionella blooms in the northwestern Bay of Bengal during 2004. Int. J. Remote Sens. 2005, 26, 3853–3858. [CrossRef] 21. Milliman, J.D.; Meade, R.H. World-wide delivery of river sediment to the oceans. J. Geol. 1983, 91, 1–21. [CrossRef] 22. Milliman, J.D.; Syvitski, J.P.M. Geomorphic/tectonic control of sediment discharge to the ocean: The importance of small mountainous rivers. J. Geol. 1992, 100, 525–544. [CrossRef] 23. Burton, J.D. River inputs into ocean systems: Status and recommendations for research. Final report of SCOR Working Group 46. In UNESCO Technical Papers in Marine Science; No. 55; UNESCO: Paris, France, 1988; p. 25. 24. Aziz, A.; Paul, A.R. Bangladesh Sundarbans: Present status of the environment and biota. Diversity 2015, 7, 242–269. [CrossRef] 25. The Daily Star. Over 66,000 Hectares Lost to Padma Since 1967: NASA Report; The Daily Star: , Bangladesh, 2018. 26. Alfieri, L.; Burek, P.; Dutra, E.; Krzeminski, B.; Muraro, D.; Thielen, J. GloFAS—Global ensemble streamflow forecasting and flood early warning. Hydrol. Earth Syst. Sci. 2013, 17, 1161–1175. [CrossRef] 27. Adler, R.; Wang, J.J.; Sapiano, M.; Huffman, G.; Chiu, L.; Xie, P.P.; NOAA CDR Program. Global Precipitation Climatology Project (GPCP) Climate Data Record (CDR), Version 2.3 (Monthly); National Centers for Environmental Information: Washington, DC, USA, 2016. [CrossRef] 28. Hossain, S.; Cloke, H.L.; Ficchì, A.; Turner, A.G.; Stephens, E. Hydrometeorological drivers of the 2017 flood in the brahmaputra basin in bangladesh. Hydrol. Earth Syst. Sci. Discuss. 2019, 1–33. [CrossRef] 29. Schroeder, T.A.; Cohen, W.B.; Song, C.; Canty, M.J.; Yang, Z. Radiometric correction of multi-temporal Landsat data for characterization of early successional forest patterns in western Oregon. Remote Sens. Environ. 2006, 103, 16–26. [CrossRef] Water 2021, 13, 2373 20 of 20

30. Felde, G.W.; Anderson, G.P.; Cooley, T.W.; Matthew, M.W.; Adler-Golden, S.M.; Berk, A.; Lee, J. Analysis of Hyperion Data with the FLAASH Atmospheric Correction Algorithm. In Proceedings of the 2003 IEEE International Geoscience and Remote Sensing Symposium, Toulouse, France, 21–25 July 2003. 31. Liu, D.; Bai, Y.; He, X.; Pan, D.; Chen, C.-T.A.; Li, T.; Xu, Y.; Gong, C.; Zhang, L. Satellite-derived particulate organic carbon flux in the Changjiang River through different stages of the Three Gorges Dam. Remote Sens. Environ. 2019, 223, 154–165. [CrossRef] 32. Roy, D.P.; Kovalskyy, V.; Zhang, H.K.; Vermote, E.F.; Yan, L.; Kumar, S.S.; Egorov, A. Characterization of Landsat-7 to Landsat-8 reflective wavelength and normalized difference vegetation index continuity. Remote Sens. Environ. 2016, 185, 57–70. [CrossRef] 33. Kendall, M.G. Rank correlation methods. Br. J. Psychol. 1990, 25, 86–91. [CrossRef] 34. Mann, H.B. Nonparametric tests against trend. Econometrica 1945, 13, 245–259. [CrossRef] 35. Shao, Z.; Cai, J.; Fu, P.; Hu, L.; Liu, T. Deep learning-based fusion of Landsat-8 and Sentinel-2 images for a harmonized surface reflectance product. Remote Sens. Environ. 2019, 235, 111425. [CrossRef] 36. Zhang, H.K.; Roy, D.P.; Yan, L.; Li, Z.; Huang, H.; Vermote, E.; Skakun, S.; Roger, J.-C. Characterization of Sentinel-2A and Landsat-8 top of atmosphere, surface, and nadir BRDF adjusted reflectance and NDVI differences. Remote Sens. Environ. 2018, 215, 482–494. [CrossRef] 37. Tucker, C.J. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 1979, 8, 127–150. [CrossRef] 38. Mount, J.F.; Florsheim, J.L.; Chin, A. Bank erosion as a desirable attribute of rivers. BioScience 2008, 58, 519–529. 39. Dewan, A.; Corner, R.; Saleem, A.; Rahman, M.M.; Haider, M.R.; Rahman, M.M.; Sarker, M.H. Assessing channel changes of the ganges-padma river system in bangladesh using landsat and hydrological data. Geomorphology 2017, 276, 257–279. [CrossRef] 40. Halder, A.; Chowdhury, R.M. Evaluation of the river Padma morphological transition in the central Bangladesh using GIS and remote sensing techniques. Int. J. River Basin Manag. 2021, 1–15. [CrossRef] 41. Sarker, M.H.; Thorne, C.R. Morphological response of the Brahmaputra–padma–lower Meghna river system to the earthquake of 1950. Braided Rivers Process. Depos. Ecol. Manag. 2006, 21, 289–310. 42. Rashid, M.B. Channel bar development and bankline migration of the Lower Padma River of Bangladesh. Arab. J. Geosci. 2020, 13, 1–16. [CrossRef] 43. Hussain, M.G.; Failler, P.; Karim, A.A.; Alam, M.K. Major opportunities of blue economy development in Bangladesh. J. Reg. 2018, 14, 88–99. [CrossRef] 44. Zheng, F.-L. Effect of vegetation changes on soil erosion on the Loess Plateau. Pedosphere 2006, 16, 420–427. [CrossRef] 45. Kumar, M.; Mondal, I.; Pham, Q.B. Monitoring forest landcover changes in the Eastern Sundarban of Bangladesh from 1989 to 2019. Acta Geophys. 2021, 69, 561–577. [CrossRef]