Estimation of Chlorophyll-a, TSM and Salinity in Mangrove Dominated Tropical Estuarine Areas of Hooghly River, North East Coast of Bay of Bengal, India Using Sentinel-3 Data

This study aims to explore the variations in spatio-temporal characteristics of water quality factors of three estuaries in the western portion of the Indian Sundarbans. Reliable retrieval of near surface concentrations of parameters such as Chlorophyll-a, SST & TSM in various aquatic ecosystems with broad ranges of trophic needs has always remained a complex issue. In this study the application of C2RCC processor has been tested for its accuracy across different bio optical regimes in inland & coastal waters. Satellite images for the same period were also collected and analysed using the C2RCC processing sequence to retrieve values of factors like the depth of water, surface re�ectance, water temperature, inherent optical properties (IOPs), chlorophyll-a, salinity and total suspended matter (TSM) using the SNAP software. During the 2017-2020 season, in situ sampling from speci�c locations and laboratory water quality analysis were carried out. The OLCI retrieved results were then trained and corroborated by means of the in situ datasets. It was observed that the highest amount of TSM was recorded in Diamond Harbour during the pre-monsoon, in the year 2018 (301.40 mgL -1 in-situ value, and 308.54 mg L -1 estimated value). Similarly, chlorophyll-a had higher concentrations through the monsoon season (3.03 mg m -3 , in-situ, and 2.96 mg m -3 , estimated) in Fraserganj and Sagar south points. Very good �tted correlation results for all seasons between Chl-a, r = 0.829 and TSM, r = 0.924 remained established throughout the comparisons of OLCI and in situ results. The high level of correlation highlights the importance of both primary as well as secondary information in understanding any dynamic system properly. Finally, the result shows that the water quality model outperforms conventional techniques and OLCI chl-a and TSM products. This paper empirically investigates a reliable remote sensing method for estimating coastal TSM and chl-a concentrations and supports the use of OLCI data in ocean colour remote sensing.


Introduction
The rst satellite operation to assess coastal aquatic quality and ocean e ciency from a remote platform was the Coastal Zone Color Sensor (CZCS), which was propelled in October 1978 (Acker, J. 2013, Mondal et al., 2014;Kyryliuk, et al., 2019).This device has been utilized in approximating global production yield, and has steered our growing understanding of the importance of oceanic and littoral phytoplankton production (Longhurst et al., 1995;Behrenfeld et al., 2006).Chlorophyll-a is one of the typical water quality factors observed in aquatic bodies.It is measured by water sampling and laboratory analysis and through in situ measurements using a water quality checker.However, the CZCS has had di culties in distinguishing between chlorophyll-a (Chl-a) and (TSM).Successive operations by the National Aeronautics and Space Administration (NASA) and the European Space Agency (ESA) have aimed to increase the accuracy of the retrieval of various water ingredients from space technology.The launch of OLCI (Ocean and land colour Instrument) on board the Sentinel 3A by the (ESA), in 2016 has enabled better management of the environment and to appreciate and improve data collection in uenced by the effects of climate change (Donlon et al., 2014;Bonekamp et al., 2016;Mondal et al., 2019;2020).OLCI mission is the follow up of the MERIS mission (2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012) (Pavel et al., 2011) with improved capabilities as its spectral conformation is premeditated for optically intricate coastal and inland aquatic bodies (Mondal et al., 2018;ESA, 2019;Kyryliuk, et al., 2019).Various studies have revealed that the OLCI is presently the most appropriate satellite instrument for water color remote sensing in inland aquatic bodies (Mograne et al., 2019;Xue et al., 2019).
The Hooghly estuary and its tributaries are an extraordinarily complex study object for ocean colour remote sensing.The very high amount of coloured dissolved organic matter (CDOM) received from the Sundarban mangrove forest in the catchment area along with anthropogenic inputs (Thakur et al., 2019) makes the water darker.As a result, the water loss signal is very small, necessitating the use of highly sensitive remote sensing devices as well as extremely precise atmospheric correction.In addition, due to fresh water input in the upper reaches and differential tidal in uences, low salinity (psu) is common in some areas, while it is much higher in others.Similarly, there are stark differences in the TSM values along different stretches of the estuaries.This necessitates the need for a good algorithm to study the entire network of estuaries criss-crossing and crossing through the Indian Sundarbans.Eutrophication and hypoxia are regular phenomena in the Indian Sundarbans and, thus, environmentalists have been working to decrease the environmental pollution of runoff.Monitoring of chlorophyll concentrations is a very imperative contribution to evaluating these endeavours.Similar conditions have been stated in the OLCI images, with very good accuracy.
Most of the studies employing the OLCI images have been taken up in Europe and South East Asia.
Studies along the eastern coast of India are very limited and, although the Sundarbans is one of the most studied areas in the world, no study using OLCI images has been attempted yet.Because of the fragility of water and the need for continuous monitoring, there is a need for real-time, surface-based studies of water quality parameters.However, due to accessibility and other logistics, that target has not been achieved on a regular basis.This study is an attempt to bridge the gap by using OLCI images from the Sentinel-3 to study a select few parameters of the Hooghly, Saptamukhi and Muri Ganga estuaries.
In this research, we presented a novel empirical model for estimating TSM and Chl-a concentrations in the Sundarban coastal waters of the Hooghly Estuary by integrating OLCI data with in situ data.Finally, the model was run using the time-series Sentinel-3 data to chart the spatial dispersal of TSM and Chl-a concentrations, and then the geographical and temporal distribution features of Chl-a concentration in the Sundarban coastal deltaic waters were analyzed.The main objectives of the study are-to estimate water quality factors such as salinity, surface temperature, chlorophyll-a concentration, and TSM concentration through seasonal eld surveys (in-situ data); to use Sentinel-3 OLCI images in estimating salinity, surface temperature, chlorophyll-a concentration, and TSM concentration during 2017-19 by C2RCC processor & to validate the results obtained by the OLCI estimations with the in situ data sets through regression equations.

Study area
The study area is part of the Sundarbans delta extending from 21°20'N to 22°40'N and 87°0'E to 89°0'E (Fig. 1) and is crisscrossed by many estuaries which are mostly distributaries of the Hooghly estuary (Thakur et al, 2020ab;Mondal, et al., 2016;2021b;Bag et al., 2019;Bandyopadhyay et al., 2014).The delta is an ecologically delicate area that is stuck by consistent tidal ebb and ow of 12 durations each (Mondal et al., 2021a).The samples of water were collected at nine widely dispersed sampling stations along the mangrove-dominated banks of the Mooriganga, Saptamukhi, and Hooghly estuaries, three distributaries of the Ganges.The names of the stations are mentioned in table 1.A mapdisplaying the sampling stations along with speci c geographical positions has been shown in gure 1.
The region has a monsoonal climate with three main seasons, viz., pre-monsoon, monsoon and postmonsoon.Seasonal eld observations were conducted at all the nine sampling stations throughout the monsoon season (July-Oct), the post-monsoon season (November-January), and the pre-monsoon seasons (March-June) for a period of two years, from November 2017 to January 2020.& 24.04.2019 for the pre-monsoon.All waterbody samples were collected using clean plastic buckets from the middle portion of the river using mechanized boats.Both eld data and satellite data were collected on the matching day and at the similar time (9.32 a.m.IST ± 30 min), irrespective of the tidal conditions in the estuary.Collected water samples were kept in acid cleaned dry polythene bottles and transported to the laboratory for the measurement of suspended particulate matter and salinity.The temperature of surface water was measured at each station using a hand-held thermometer.The transparency was measured using a metal Secchi disc of 20 cm in diameter.The mean of three Secchi disc depths was chosen as the water transparency at each location (Preisendorfer 1986;Lee et al. 2015).
All samples were delivered to the laboratory as soon as they were collected.Every time, triplicate samples were collected and analysed periodically to check the reproducibility of results and to evaluate the precision of measurements.Salinity was measured by argentometric iteration following the Mohr-Knudsen method (Grasshoff et al. 1983).TSM was separated by ltering an aliquot of water sample (1-2 litters) through a pre-weighed 0.45 (mm) Millipore membrane lter under vacuum, and weighing it with an accuracy of (0 ± 0.1mg) after drying at 60 o C or in desiccators in the presence of Conc.H 2 SO 4 .

Estimation of Salinity and Temperature
Salinity was estimated through the argentometric titration method, following Strickland & Parsons, 1972.
Samples of waste-water (15 ml) were titrated against a standard AgNO 3 solution, using a K 2 CrO 4 solution (3.5 g l -1 ) as an indicator.Silver nitrate solution was previously standardized against standard seawater solution (3.5 g NaCl in 96.5 ml distilled water), having chlorinity 19.375 ppt and salinity 35‰.The salinity and chlorinity were related by Knudsen's equation; S‰ = 0.03 + 1.805 Cl‰.This technique is precise up to 0.05 to 0.1 ‰ of salinity.The water temperature was recorded during each sample collection using a laboratory thermometer.

Chlorophyll measurement
For the validation of water parameters, we collected around 15-20 cm under the river surface using Niskin bottles, and samples have been moved to the laboratory for further analysis.Water samples were clari ed over 0.7 mm Whatman glass ber lter (GF/F) paper under low vacuum and kept in acetone of 90% concentration at 0°C for 24h in the dark for the complete extraction.The extracted solution was centrifuged at 10000 rpm for 10min and the solvent was used to estimate Chl-a concentration using a Shimadzu UV-vis 2450 spectrophotometer after the technique described in Strickland and Parsons (1972).The precision of Chl-a estimation is at the 5 µG level, the correct value lies in the range: The organic and inorganic section of (TSM in mg L -1 ) has been quanti ed using the gravimetric technique meticulously given by Toming (2017) and also the MERIS protocols.Whatman Glass Fiber Filters (GF/F) are washed with ultrapure water to remove any drop lter bits, and they now combust at 4800C with an instruction to burn off any conceivable organic contamination (Doerffer et al., 2002).The weighted clean lters and they are kept in a folded square of aluminum foil paper (0.020 * 100 * 100 mm) through counted numbers up to their percolation.In accumulation, the amount of water samples (1-2 L) has been ltered in triplicates over and done with the pre-weighed and also pre-combusted lter methods (Kratzer et al., 2018).The funnel and lter wash the clean water, then add 50mL of ultrapure water to take out any residual salt.The lters were dried up overnight at 60 0 C and preserved in a desiccator prior to consideration by a microbalance (±1µg).TSM was obtained through the dissimilation of both the tare and the dry weight.Then, the sample data is combusted at 480 0 C in an incinerator, monitored by an alternative weighing stage.The mass of the inorganic TSM was then equal to the mass of the combusted lters (modi ed tare weight), and the organic segment was the variance of the entire and the inorganic TSM.
2.2.3.Sentinel-3 Satellite data processing 2.2.3.1.Image pre-processing The Sentinel-3 OLCI has a speci cation sensor as per its precursor Medium Resolution Imaging Spectrometer (MERIS) on ENVISAT with the capability to achieve an analysis of bio-optical ingredients in global coastal and inland regions (Doerer et al., 1999, Moore et al., 1999;Merheim-Kealy et al., 1999).The Sentinel-3 OLCI bands (Table 2) are an inheritance of MERIS and are supplemented to improve the quality of the measurement of ocean colour remote sensing.The Sentinel 3 OLCI instrument swath area covers around 1270 km and it's tilted across by a track of 2.6 in the opposite track to the sun's angle in direction to minimize the potential effect of sun glint (ESA, 2019).Cloud free images for the period 2017-2020, viz. the monsoon (June), post-monsoon (November), and pre-monsoon (April) were downloaded and used in the study.The Sentinel-3 satellite passed above the selected sampling area around 9.32 am and satellite images with a ± 6-hour time lag from water sampling timings were collected on the following dates: 13.06.2018& 19.06.2019 for the monsoon, 13.12.2017,10.12.2018 & 2.01.2019 for the post-monsoon, and 21.04.2018& 24.04.2019 for the pre-monsoon.Using the SNAP software, the images were corrected for atmospheric effects.The Case-2 Regional/coast colour (C2RCC) processor was used for atmospheric correction (Brockmann et al., 2016).The C2RCC processor is a processing chain that helps in recovering water quality factors such as Chl-a, TSM, SST and salinity.The processor comprises of two neural nets.
One net removes atmospheric and water surface effects (such as glint), while the other net retrieves absorption and scattering coe cients from which optically active substance concentrations (such as Chla) are calculated.The processor is available freely through the SNAP (Sentinel Application Platform) software.Finally, we have used the cloud-free pixel values relating to each sampling station's location were extracted from each thematic layer and veri ed against ground truth data.The extraction multi-layer values to points tool in Arc GIS was used to extract the Chl-a and TSM indices from the images at locations speci ed in a point feature class.For each input raster cluster value, a novel arena comprising the cell values for individual input raster is added to the input particular point feature class, and each pixel that covered the geographic location of the station value is extracted.After that, an attribute table was disseminated to MS Excel to build a connection between both the radiometric values and in-situ Chla and TSM concentration, as well as an analysis of the models produced.Figure 2 shows a summary of the methods used in this study.

Methods
The water pixels are processed exclusively by the Sentinel 3 OLCI data processing system.A part of it is based on atmospheric correction, which produces water-leaving re ectance's and, as a by-product, the aerosol weight overhead it, and the other part is based on ocean colour processing, which is the derivation of the colour of the aquatic body itself from the water-leaving re ectance's of the suite of merchandises telling it.The results of the laboratory analysis were then arranged in a systematic manner.Cloud free OLCI C2RCC processed pixel values conforming to the sampling location points were extracted.These values of data sets giving chl-a, TSM, SST and Salinity were also added to the database (Table 1; Fig. 2).

Statistical analysis
The validation operations took place between 2017 and 2020 in the Hooghly estuary of the Bay of Bengal.In different seasons of the year, marine, coastal, and inland water bodies have been covered, including in situ measurements of parameters such as chlorophyll-a, TSM, salinity, and sea surface temperature.After the datasets were assembled in one database, they were subjected to statistical analysis, and nally, the OLCI retrieved data validated with the in situ results by applying regression and correlation analysis using Matlab software.

Results
The physico-chemical and biological feature of a owing river system largely depend on the tidal impacts, the intensity of contribution from point and non-point sources contributing on both sides of the banks, and the anthropogenic activity in and around the riverine system.Thus, the variations of the studied components are ascribed to tidal variations, spatial variations, and seasonal and annual variations.In this research, we have estimated the water quality factors from Sentinel-3 images along with multi-seasonal in-situ eld data and have tried to validate and analyse the ndings.
3 Summer has dominated the pre-monsoon, resulting in higher mean water temperature values at all sampling stations.In this season it varied between 34.65 ± 0.64°C at Sagar south and 30.05 ± 0.07°C at Beguakhali and Henry's island and 30.05 ± 0.35°C at Bakkhali.In the same season, the mean water temperature at BabuGhat was 32.90 ± 2.69°C and at Diamond Harbour it was 33.90 ± 0.14°C (Fig. 4).A point to be mentioned is that the water temperature variation majorly depends on tidal circulation and the sample collection time.Hence, from its variation, no important conclusion can be drawn directly.

Mapping Chl-a Concentration from the OLCI Images
Chlorophyll content of water is an excellent indicator of biotic production in any aquatic system.In estuaries like Hooghly, chlorophyll-a concentrations signi cantly describe water circulation and dilution patterns.It is sometimes also an important indicator of anthropogenic input of nutrients into the nearshore areas.In our study, the monsoonal concentration of in-situ ( eld survey) Chl-a was lower than in the other two seasons.During the monsoon, the mean Chl-a concentration varied from 1.16 ± 0.10 mg m -3 at Henry's Island to 2.47 ± 0.36 mg m -3 at Diamond Harbour.Among other stations, Namkhana had exhibited 1.42 ± 0.16 mg m -3 , Sagar south had exhibited 1.86 ± 0.31 mg m -3 , Babu Ghat had exhibited 1.72 ± 0.68 mg m -3 , Beguakhali had exhibited 1.73 ± 0.22 mg m -3 and Bakkhali had exhibited 1.40 ± 0.10 mg m -3 (Fig. 5).During the same season, the estimated (Sentinal-3) Chl-a values were lower than the insitu values.The estimated values varied from 0.68 ± 0.03 mg m -3 at Namkhana to 1.47 ± 0.29 mg m -3 at Fraserganj (Fig. 6a).
During post-monsoon the estimated Chl-a values were higher compared to the monsoon, with a highest of 1.96 ± 0.80 mg m -3 at Sagar South and a lowest of 1.05 ± 0.28 mg m -3 at Henry's Island.Other major contributors to the in-situ mean Chl-a concentration in the post-monsoon season were 1.13 ± 0.05 mg m -3 at Babu Ghat, 1.22 ± 0.27 mg m -3 at Diamond Harbour, 1.16 ± 0.83 mg m -3 at Namkhna point, and 1.53 ± 0.40 mg m -3 at Bakkhali point (Fig. 6b).
During the same season, the in-situ Chl-a levels were higher than the estimated levels.The in-situ Chl-a varied from 2.28 ± 0.70 mg m -3 at Sagar South to 1.26 ± 0.04 mg m -3 at Babu Ghat (Fig. 5) When compared to the other two seasons, pre-monsoon concentrations of in-situ Chl-a were highest.During the pre-monsoon, the mean in-situ Chl-a concentration has varied from 2.45 ± 0.50 mg m -3 at Bakkhali to 1.23 ± 0.14 mg m -3 at Namkhana.Higher mean in-situ Chl-a concentrations were also observed on Henry's Island, i.e. of 2.15 ± 0.52 mg m -3 , in Diamond Harbour, i.e. of 1.80 ± 0.31 mg m -3 , in Babu Ghat, i.e.
Again, in the pre-monsoon season, the estimated Chl-a concentration was lower than the in-situ values.
The highest estimated Chl-a concentration was recorded at Bakkhali, i.e. 1.82 ± 0.44 mg m -3 and the lowest was at Namkhana, i.e. 0.80 ± 0.16 mg m -3 in this season during 2018 -19 (Fig. 6c).This trend of mean Chl-a concentration variation has indicated that as the river ows downstream towards the Bay of Bengal, the parameter gets diluted and so the productivity drops lower.

Mapping of Total Suspended Matter (TSM) from the OLCI Images
The weight of TSM is dependent upon a number of factors present in the water column, e.g.turbidity, chlorophyll concentration, waste water dilution, and tidal uctuations.Because of massive anthropogenic in ows, the upstream parts of the Hooghly estuary have shown higher values.TSMs concern their Bay of Bengal ward counterparts, with Beguakhali as an exception during the post-monsoon and monsoon seasons.During the post-monsoon season, the mean weight of in-situ TSM was the highest, 129.24 ± 27.08 mg L -1 at Diamond Harbour, and the lowest, 25.53 ± 9.48 mg L -1 at Babu Ghat.Other major contributors to the post monsoonal mean weight of in-situ TSM were Namkhana with 57.97 ± 23.01 mg L -1 , Kachubaria with 101.71 ± 32 mg L -1 , Henry's Island with 82.65 ± 18.21 mg L -1 and Bakkhali with 69.87 ± 25.44 mg L -1 (Fig. 7).
The estimated TSM was higher compared to the in-situ values for this season.The highest estimated TSM concentration was recorded at Diamond Harbour, i.e. 134.61 ± 30.55 mg L -1 and the lowest was at Babu Ghat, i.e. 41.86 ± 22.24 mg L -1 (Fig. 8a).During the pre-monsoon, the scenario changes.With a mean weight of estimated TSM of about 114.06 ± 30.61 mg L-1, Babu Ghat had exhibited the lowest and, with about 281 ± 26.95 mg L-1, Kachubaria had acquired the highest position.Among other upstream sampling stations, Diamond Harbour had exhibited 180.58 ± 25.83 mg L -1 and Namkhana had exhibited 102.93 ± 84.48 mg L -1 .Among downstream points, Sagar south had exhibited 119.26 ± 11.39 mg L -1 and Henry's Island had exhibited 179.99 ± 18.68 mg L -1 (Fig. 8b).This season, the highest in-situ TSM was measured at Kachubaria (273.05 28.35 mg L-1) and the lowest at Babu Ghat (108.03 27.72 mg L-1) (Fig. 7).
During the monsoon period at Fraserganj, the highest mean weight of estimated TSM, i.e. 150.41 ± 26.17 mg L -1 was observed, and at Babu Ghat, the lowest value of 26.30 ± 2.49 mg L -1 was observed.Similar to this, the weight of in-situ TSM was also highest at Fraserganj, i.e. 148.96 ± 25.36 mg L -1 and was lowest ground data and satellite data is signi cantly viable for larger populations and varied hydrological parameters.This could be bolstered further by a correlation study between the two.A signi cant correlation was observed between eld data and Sentinel-3 data in all seasons along the Hooghly River for both parameters, i.e.Chl-a and TSM (Table 3).All r values for individual seasons were > 0.5, which indicates a signi cant relationship.This may further be supported by p values, i.e. all values ≤ 0.05.This validates the level of accuracy of eld sampling and also states the importance of both primary and secondary data in understanding any dynamic system properly.When studied for all seasons, for Chl-a, r = 0.829 and p = 0.000 and for TSM, r = 0.924 and p = 0.000.Both these were again signi cantly supported by the strength of satellite data and eld data variables.Although there is a continuing argument amongst scholars about the reliability of primary eld data versus secondary satellite data, this type of comparative study might work as a conduit between the two and will have a futuristic application of interpretation in oceanic research.
The study's ndings indicate the need to develop an algorithm speci c to the Hooghly estuary, particularly for the C2RCC processor, because the current ones only provide sensible re ectance outcomes in Chlorophyll and TSM circumstances and correlation with in-situ data.

Discussion Of The Study
In the present scenario, the spectral and temporal resolution of OLCI makes it ideal for mapping and collecting water quality parameter information for large-scale coastal waters.The OLCI data has been acquired nearly daily using the Sentinel-3A and Sentinel-3B satellites' ocean monitoring programme.
However, image quality is constantly affected by the weather, particularly in cloudy and rainy subtropical coastal regions such as the Sundarban delta.This limits the usefulness of the OLCI images for studying ocean colour, yet this climatic factor is unavoidable in optical remote sensing.To acquire additional ocean colour information in the future, we should explore combining multi-source satellite data.
Furthermore, the atmospheric adjustment is important for retrieving ocean colour.The quality of the atmospheric adjustment has a substantial effect on the remote sensing inversion outcome.For ocean colour modelling, a good atmospheric correction technique can provide high-quality re ectance data.
Despite the fact that ocean colour missions have given us with a plethora of ocean colour products, their suitability for use in local environments has not been determined.In this study, we compared the Chl-a and TSM with a correlation matrix-based model on in-situ measurement with the (IOPs) product.While the spatial distribution of TSM and Chl-a concentrations from various techniques is usually comparable, the TSM and Chl-a values from each method have vastly different ranges.The primary reason for this is that eld measurement data for other ocean colour models has been gathered from different ocean areas throughout the years, while in-situ measurement data for the correlation matrix-based model is collected from the Hooghly estuary's coastal waters (Fig. 10).According to our result shows the Sentinel-3 products are suitable for ocean color monitoring in Hooghly estuary of Sundarban coastal waters.Although the correlation matrix-based technique has shown positive results in the coastal waters of the Hooghly estuary, for the spatial application of this method.Finally, our approach, on the other hand, offers signi cant reference value for the study of ocean colour in other coastal regions.

Conclusions And Outlook
The retrieval of the estuarine water quality parameters and its subsequent validation was generally successful with moderately consistent results (comparatively stumpy bias and scatter) for the Chl-a, TSM, salinity, and water temperature.In general, the C2RCC does well in recovering a remote sensing spectrum with a distinctive deltaic shape for estuary waters with a re ectance peak at 560 nm.Results revealed that the maximum concentration of chlorophyll-a was observed during the post-monsoon period (2.96 mg m -3 , estimated and 3.17 mg m -3 , in-situ) on Sagar Island South.The amount of chlorophyll-a belongs moderately in the continental shelf zone than continental slope because of the abundance of phytoplankton.That is why the continental shelf zone is better for shing activities.The in-situ results total amount of TSM has decreased progressively into land towards the sea completely over the year (2017-2020).The amount of TSM is very low at the most upstream sampling point, i.e.Babu Ghat.It has shown a mean of 36.76 mg L -1 in-situ and 58.04 mg L -1 estimated TSM in all seasons of our study period.The study results have also shown the highest amount of in-situ TSM at Fraserganj during the monsoon time in 2019 (174.32 mgL-1), much more in the monsoon period than in the post-monsoon period in the Hooghly estuaries.
The intention of this study was to check the performance of Sentinel-3 OLCI in coastal waters by appraising the results of atmospherically corrected ocean colour products produced by a standard processor (C2RCC), and the results point to very good performance in the test.Very good tted correlation results for all seasons between Chl-a, r = 0.829 and TSM, r = 0.924 were obtained during the comparisons of OLCI with the in situ results.The high level of correlation highlights the importance of both primary and secondary data in understanding any dynamic system properly.This research could pave the way for future research into the estuarine waters of the Indian Sundarbans, which are di cult to access for laboratory-based studies.Real-time monitoring using OLCI images and the application of different algorithms could bring about a very good chance in the environmental monitoring programs of the Indian Sundarbans.Competing Interests:

Declarations
This manuscript has not been published or presented elsewhere in part or in entirety and is not under consideration by another journal.There are no con icts of interest to declare.

Availability of data and materials:
The data that support the ndings of this study are available from the author, [Ismail Mondal, ismailmondal58@gmail.com], upon reasonable request.
Declarations:  In-situ Measurement of Water Quality Estimation in the Hooghly Estuary

Funding
been done under the project entitled "Water quality assessment using AVIRIS-NG satellitederived data along the Hooghly (Ganges) River Estuary, Eastern part of India" [Reference No.: A/F/693/4S-2973/2016] with collaboration between Space Applications Centre, ISRO, Ahmedabad, and the Department of Marine Science, University of Calcutta, Kolkata, India.The nancial support from SAC, ISRO, Ahmedabad, is gratefully acknowledged.The help from the Port Trust Authority of India and Shipping Corporation of India is being duly and gratefully acknowledged for their support during the sampling periods by providing us with means of riverine transportation and personnel of assistance.Finally, the authors are indebted to the University Grants Commission, Government of India; for the award of the DS Kothari Post-Doctoral Fellowship (ES/20-21/0009) to Dr. Ismail Mondal.And also the European Space Agency (ESA) for providing the Sentinel-3 data to further extend our research work. Figures

Table 1 :
Name of sampling stations from North to south

Table 2 :
OLCI bands with MERIS heritage bands and additional highlighted in bold (source: ESA)

Table 3 .
Table showing Seasonal correlation between Sentinel-3 observed data and field