Chlorophyll-a Estimation Research Articles

Unveiling primary production dynamics in the southern Bay of Bengal: Comparative analysis of in-situ measured data with multiple primary production models

In this study, marine water samples were collected at 18 sampling stations. The collected marine water samples were analysed using a well-calibrated UV spectrophotometer for chlorophyll-a estimation, and using a hyperspectral underwater radiometer, both buoy mode and profile mode in-situ data were collected. The study evaluated four different primary production models to study the dynamics of the southern Bay of Bengal. The simple Eppley model used in the study resulted in huge variations in the nearshore and frontal regions; the dark and light bottle methods showed enormous variations along all 18 sampling stations and underestimated the productivity of the southern Bay of Bengal. However, the VGPM model and its variants showed similar trends in near shore, frontal, and off-shore waters for all three transects, and the Kameda model closely matched the available literature for the primary production estimates of the southern Bay of Bengal. From the detailed analysis of the Kameda model, it is found that the average primary production for the Southern Bay of Bengal for the post monsoon season (February 2023) is estimated as 339.08 mgC/m2/d.

Temporal variability in the relationship between line height absorption and chlorophyll concentration: a case study from the Northern Gulf of Alaska

The Line Height Absorption (LHA) method uses absorption of light to estimate chlorophyll-a. While most users consider regional variability and apply corrections, the effect of temporal variability is typically not explored. The Northern Gulf of Alaska (NGA) was selected for this study because there was no published regional value and its large swings in temporal productivity would make it a good candidate to evaluate the effect of temporal variability on the relationship. The mean NGA value of 0.0114 obtained here should be treated with caution, as variation in the slope of the relationship (aLH*), and thus chlorophyll-a estimates, in the NGA region varied by ∼25% between spring (aLH* = 0.0109) and summer (aLH* = 0.0137). Results suggest that this change is driven by a shift in pigment packaging and cell size associated with changes in mixed layer depth and stratification. Consideration of how temporal variability may affect the accuracy of the LHA method in other regions is thus recommended.

Atlas of phytoplankton phenology indices in selected Eastern Mediterranean marine ecosystems.

Phytoplankton is a fundamental component of marine food webs and play a crucial role in marine ecosystem functioning. The phenology (timing of growth) of these microscopic algae is an important ecological indicator that can be utilized to observe its seasonal dynamics, and assess its response to environmental perturbations. Ocean colour remote sensing is currently the only means of obtaining synoptic estimates of chlorophyll-a (a proxy of phytoplankton biomass) at high temporal and spatial resolution, enabling the calculation of phenology metrics. However, ocean colour observations have acknowledged weaknesses compromising its reliability, while the scarcity of long-term in situ data has impeded the validation of satellite-derived phenology estimates. To address this issue, we compared one of the longest available in situ time series (20 years) of chlorophyll-a concentrations in the Eastern Mediterranean Sea (EMS), along with concurrent remotely-sensed observations. The comparison revealed a marked coherence between the two datasets, indicating the capability of satellite-based measurements in accurately capturing the phytoplankton seasonality and phenology metrics (i.e., timing of initiation, duration, peak and termination) in the studied area. Furthermore, by studying and validating these metrics we constructed a satellite-derived phytoplankton phenology atlas, reporting in detail the seasonal patterns in several sub-regions in coastal and open seas over the EMS. The open waters host higher concentrations from late October to April, with maximum levels recorded during February and lowest during the summer period. The phytoplankton growth over the Northern Aegean Sea appeared to initiate at least a month later than the rest of the EMS (initiating in late November and terminating in late May). The coastal waters and enclosed gulfs (such as Amvrakikos and Maliakos), exhibit a distinct seasonal pattern with consistently higher levels of chlorophyll-a and prolonged growth period compared to the open seas. The proposed phenology atlas represents a useful resource for monitoring phytoplankton growth periods in the EMS, supporting water quality management practices, while enhancing our current comprehension on the relationships between phytoplankton biomass and higher trophic levels (as a food source).

Response of mineral particles in inland lakes to water optical properties and its influence on chlorophyll-a estimation.

Many chlorophyll-a (Chl-a) remote sensing estimation algorithms have been developed for inland water, and they are proposed always based on some ideal assumptions, which are difficult to meet in complex inland waters. Based on MIE scattering theory, this study calculated the optical properties of mineral particles under different size distribution and refractive index conditions, and the Hydrolight software was employed to simulate remote sensing reflectance in the presence of different mineral particles. The findings indicated that the reflectance is significantly influenced by the slope (j) of particle size distribution function and the imaginary part (n') of the refractive index, with the real part (n) having a comparatively minor impact. Through both a simulated dataset containing 18,000 entries and an in situ measured dataset encompassing 2183 data from hundreds of lakes worldwide, the sensitivities of band ratio (BR), fluorescence baseline height (FLH), and three-band algorithms (TBA) to mineral particles were explored. It can be found that BR showed the best tolerance to mineral particles, followed by TBA. However, when the ISM concentration is less than 30 g m-3, the influence of CDOM cannot be ignored. Additionally, a dataset of over 400 entries is necessary for developing the BR algorithm to mitigate the incidental errors arising from differences in data magnitude. And if the amount of developing datasets is less than 400 but greater than 200, the TBA algorithm is more likely to obtain more stable accuracy.

Primary production and microbial heterotrophy in the Siberian arctic seas, Bering Strait, and Gulf of Anadyr, Bering Sea

The data obtained during the recent decades indicate large-scale climatic changes in the Arctic Ocean, especially in the Siberian shelf seas. The impact of shrinking of the Arctic ice cover on marine net primary production (PP) was mostly studied using satellite-derived estimates of chlorophyll-a (Chl a), sea surface temperature, and sea ice concentrations, as well as various modeling approaches. However, no data are available on the quantitative relations between phytoplankton production and heterotrophic bacterioplankton activity in the broadest and shallowest shelf of the World Ocean - the East Siberian Arctic seas (ESAS). In this paper we present a full set of in situ measurements of main PP and microbial heterotrophy characteristics required for better understanding and further validation of the satellite and modeling approaches. In September–October 2011, during the 57th cruise of RV Akademik M.A. Lavrent'ev, PP, bacterioplankton abundance (BA) and bacterioplankton production (BP), as well the rates of dark CO2 assimilation (DCA) were determined in the Chukchi, East Siberian, and Laptev seas and in the Gulf of Anadyr of the Bering Sea (a total of 26 stations) by in situ measurements using radiolabeled bicarbonate. All three important production indicators exhibited significant variance within the studied water areas of the Arctic (РР from 0.98 to 54.7 μg C L−1 day−1), (ВР from 1.1 to 17.1 μg C L−1 day−1), and (DCA from 0.23 to 8.7 μg C L−1 day−1). The state of phyto- and bacterioplankton was determined by analysis of the BP/PP and DCA/PP ratios. Active photosynthesis with efficient incorporation of inorganic carbon into the biomass was observed in the Chukchi Sea coastal eastern waters and in the Gulf of Anadyr. On the contrary, the phytoplankton primary production was extremely low in oligotrophic basins of the East Siberian and Laptev seas, while the activity of heterotrophic bacterioplankton consuming organic matter of phytoplankton origin was relatively high. The data of direct measurements can be used for further adjustment of the indirect assessments of production obtained by satellite observations. Glossary East Siberian Arctic seas (ESAS), net primary production (PP), bacterioplankton abundance (BA), bacterial biomass (BB), bacterial production (BP), Arctic Ocean PP (AO-PP), organic matter (OM), dissolved organic matter (DOM), colored dissolved organic matter (CDOM), light CO2 assimilation (LCA), dark (heterotrophic + chemotrophic) CO2 assimilation (DCA), production indicators of the surface water layer (PPsurf, DCAsurf, BPsurf), estimated production indicators for the entire depth of the sea (PPint, DCAint, BPint; mg C m−2 day−1).

Automated spectral transfer learning strategy for semi-supervised regression on Chlorophyll-a retrievals with Sentinel-2 imagery

ABSTRACT Multispectral images make it possible to retrieve water quality parameters over a wide range and long time series by remote sensing. The robustness of traditional supervised machine learning models, which have been widely used, is affected in various regions, and these models are developed by in-situ measurements and limited satellite images. This study combined a spectral transfer learning strategy with a semi-supervised regression model to conduct data augmentation and developed a robust model for two inland lakes in Jilin Province, Northeast China, based on Sentinel-2 Multispectral Instruments images. Specifically, this study integrated an automated module for Balance Distribution Adaptation and a spectral features transfer method, and then applied it to Co-training Regressors for Chlorophyll-a retrievals to achieve the highest accuracy, which solved the selection of parameters and models. The Automated model was developed and achieved the best performance (R 2 is 0.91, the root mean square error is 2.62 μg L−1 and the mean absolute percentage error is 22.88%) compared with other models, and showed robustness in various lakes and months. Our results offer a reliable approach to provide an accurate Chlorophyll-a estimation of spatial–temporal variations.

Evaluation of Sentinel-2 Based Chlorophyll-a Estimation in a Small-Scale Reservoir: Assessing Accuracy and Availability

Small-scale reservoirs located in river estuaries are a significant water resource supporting agricultural and industrial activities; however, they face annual challenges of eutrophication and algal bloom occurrences due to excessive nutrient accumulation and watershed characteristics. Efficient management of algal blooms necessitates a comprehensive analysis of their spatiotemporal distribution characteristics. Therefore, this study aims to develop a chlorophyll-a (Chl-a) estimation model based on high-resolution satellite remote sensing data from Sentinel-2 multispectral sensors and multiple linear regression. The multiple linear regression (MLR) models were constructed using multiple reflectance-based variables that were collected over 2 years (2021–2022) in an estuarine reservoir. A total of 21 significant input variables were selected by backward elimination from the 2–4 band algorithms as employed in previous Chl-a estimation studies, along with the Sentinel-2 B1-B8A wavelength ratio. The developed algorithm exhibited a coefficient of determination of 0.65. Spatiotemporal variations in Chl-a concentration generated by the algorithm reflected the movement of high Chl-a concentration zones within the body of water. Through this analysis, it turned out that Sentinel-2-based spectral images were applicable to a small-scale reservoir which is relatively long and narrow, and the algorithm estimated changes in concentration levels over the seasons, revealing the dynamic nature of Chl-a distributions. The model developed in this study is expected to support effective algal bloom management and water quality improvement in a small-scale reservoir or similar complex water quality water bodies.

Fluorescence in the estimation of chlorophyll-a in public water reservoirs in the Brazilian cerrado

Context The usual strategy for monitoring of eutrophication process is the use of traditional limnological methods, based on laboratory analysis. These procedures involve costly and time-consuming analyses, usually with in vitro methodologies, which can still have limitations in terms of sensitivity and reliability, if poorly managed. Phytoplankton pigments, such as chlorophyll-a (Chl-a), are highly fluorescent and can provide the environmental status of water bodies. Aims This study aims to analyse, compare and evaluate an estimation of Chl-a through fluorescence in public water sources in the Brazilian cerrado. Exploratory statistical analyses were conducted by using absolute fluorescence units (AFU) and relative fluorescence units (RFU) compared with traditional laboratory data (standard procedure for the determination of Chl-a by spectroscopic methods) to evaluate the significance of differences in estimating Chl-a concentration. Subsequently, empirical models, based on spectral band combinations, were generated to convert fluorescence measurement in Chl-a concentration, by linear regression. Key results The generated model found a strong correlation and coefficient of determination (r = 0.88; R2 = 0.78). The efficiency of the model was also confirmed by statistical indicators (RMSE = 1.27, MAPE = 26.72 and BIAS = −6.32). Conclusions We concluded that the estimate of Chl-a through RFU was better than through AFU. Implications Therefore, based on the results of this study, it is recommended that RFU be used to obtain more precise and accurate estimates of Chl-a concentration through empirical models based on linear regression.

Dynamic monitoring and analysis of chlorophyll-a concentrations in global lakes using Sentinel-2 images in Google Earth Engine

Remote estimation of Chlorophyll-a (Chl-a) has long been used to investigate the responses of aquatic ecosystems to global climate change. High-spatiotemporal-resolution Sentinel-2 satellite images make it possible to routinely monitor and trace the spatial distributions of lake Chl-a if reliable retrieval algorithms are available. In this study, Sentinel-2 images and in-situ measured data were used to develop a Chl-a retrieval algorithm based on 13 optical water types (OWTs) with a satisfying performance (R2 = 0.74, RMSE = 0.42 mg/m3, MAE = 0.33 mg/m3, and MAPE = 55.56 %). After removing the disturbance of algal blooms and other factors, the distribution of Chl-a in 3067 of the largest global lakes (≥50 km2) was mapped using the Google Earth Engine (GEE). From 2019 to 2021, the average Chl-a concentration was 16.95 ± 5.95 mg/m3 for the largest global lakes. During the COVID-19 pandemic, global lake-averaged Chl-a concentration reached its lowest value in 2020. From the perspective of spatial distribution, lakes with low Chl-a concentrations were mainly distributed in high-latitude, high-elevation, or economically underdeveloped areas. Among all the potential influencing factors, lake surface temperature had the largest contribution to Chl-a and showed a positive correlation with Chl-a in approximately 92.39 % of the lakes. Conversely, factors such as precipitation and tree cover area around the lake were negatively correlated with Chl-a concentration in nearly 61.44 % of the lakes.

“Extreme Highest” and “Extreme Anomalous”: Proposed indices for chlorophyll-a extreme events in European seas between 2003 and 2021

This study addresses the occurrence of extreme values of chlorophyll-a concentrations in European seas, using the satellite's OBGP MODIS AQUA v2018 dataset, for the 2003–2021 period. Two novel and complementary statistical indices based on the combination of the overall period's and the monthly 90th percentiles (P90 and mP90, respectively) are proposed: (i) the Extreme Highest exceedances (EH) include the observations with highest chlorophyll-a magnitudes in each location (larger than P90 and mP90); (ii) the Extreme Anomalous exceedances (EA) address the observations larger than mP90 but lower than P90, thus representing relevant anomalies during low phytoplankton growing seasons. Given the important differences of available observations per pixel, the EH and EA exceedances are normalized and also presented in percentage units.Although the occurrence of these indices greatly varies in time and space, the aggregated statistics and maps reveal some clear patterns: EH and EA have very distinct (almost complementary) seasonal and spatial distribution, EH prevail in mesotrophic and euphotic waters during the main interannual bloom season whilst EA are more abundant in oligotrophic waters out of the main seasonal bloom in each area. Both significant (p < 0.05) increasing and decreasing annual trends have been found in different European Seas. Overall, these EH and EA trends are reflecting the climate-driven physical and ecological changes in European Seas. Although these results and the conceptual and computational simplicity of these indices are encouraging, further ground truth testing is required to account for their uncertainties, mostly related to data representativeness and the performance of the chlorophyll-a estimation algorithms.

An application of C2-Net atmospheric corrections for chlorophyll-a estimation in small reservoirs

This study assesses Sentinel 2 remote sensing data for estimating chlorophyll-a concentrations in 32 small reservoirs located in the Spanish region of Extremadura. The atmospheric correction algorithms C2RCC, C2X, and C2XC were used to obtain chlorophyll-a estimations. In-situ measurements of chlorophyll-a were collected from ninety-four water samples in the study area, and their values were used to validate the satellite-derived estimates. The results showed that the C2RCC atmospheric correction algorithm outperformed the other two algorithms, with a Mean Absolute Average Error (MAAE) of 4.79 mg/m3 for non-hypertrophic reservoirs. C2X and C2XC algorithm performed inaccurately, with extreme RMSE and MAAE values. Spearman correlation coefficients are good for the three methods applied. The findings could be of interest to regional water management authorities and researchers in monitoring water quality in small reservoirs.

Remote sensing water quality inversion using sparse representation: Chlorophyll-a retrieval from Sentinel-2 MSI data

Remote sensing has been widely applied to estimate the water quality parameters using empirical approaches. Numerous algorithms have been developed for water quality retrieval. However, the identification of key features for inversion and the reduction of uncertainty with limited field measurements are crucial aspects. In this study, satellite-based sparse representation is developed as an optimization model to simultaneously determine the major spectral features, and estimate water quality maps under noisy environment. Result shows that the blue–green ratio is the important feature for estimation of chlorophyll-a (Chl-a) concentration, whereas the NIR-red algorithm is the better one with in retrieving Chl-a in a high concentration case. The Chl-a map is estimated by using main spectral features of Sentinel 2 MSI data constrained with observations (correlations between observations and estimations: over 0.9). The rapid mapping of Chl-a in inland water allows us to assess spatial distribution of the water quality. This study provides reliable and interpretable information for policymakers to implement effective water quality management practices.

Incorporating marine particulate carbon into machine learning for accurate estimation of coastal chlorophyll-a

Accurate predictions of coastal ocean chlorophyll-a (Chl-a) concentrations are necessary for dynamic water quality monitoring, with eutrophication as a critical factor. Prior studies that used the driven-data method have typically overlooked the relationship between Chl-a and marine particulate carbon. To address this gap, marine particulate carbon was incorporated into machine learning (ML) and deep learning (DL) models to estimate Chl-a concentrations in the Yang Jiang coastal ocean of China. Incorporating particulate organic carbon (POC) and particulate inorganic carbon (PIC) as predictors can lead to successful Chl-a estimation. The Gaussian process regression (GPR) model significantly outperforming the DL model in terms of stability and robustness. A lower POC/Chl-a ratio was observed in coastal areas, in contrast to the higher ratios detected in the southern regions of the study area. This study highlights the efficacy of the GPR model for estimating Chl-a and the importance of considering POC in modeling Chl-a concentrations.

Recovery of Water Quality and Detection of Algal Blooms in Lake Villarrica through Landsat Satellite Images and Monitoring Data

Phytoplankton is considered a strong predictor of the environmental quality of lakes, while Chlorophyll-a is an indicator of primary productivity. In this study, 25 LANDSAT images covering the 2014–2021 period were used to predict Chlorophyll-a in the Villarrica lacustrine system. A Chlorophyll-a recovery algorithm was calculated using two spectral indices (FAI and SABI). The indices that presented the best statistical indicators were the floating algal index (R2 = 0.87) and surface algal bloom index (R2 = 0.59). A multiparametric linear model for Chlorophyll-a estimation was constructed with the indices. Statistical indicators were used to validate the multiple linear regression model used to predict Chlorophyll-a by means of spectral indices, with the following results: a MBE of −0.136 μ, RMSE of 0.055 μ, and NRMSE of 0.019%. All results revealed the strength of the model. It is necessary to raise awareness among the population that carries out activities around the lake in order for them to take policy actions related to water resources in this Chilean lake. Furthermore, it is important to note that this study is the first to address the detection of algal blooms in this Chilean lake through remote sensing.

A preliminary assessment of Sentinel-3 ocean and land color instrument data for the estimation of chlorophyll-a concentration using bio-optical methods in Annaba Bay and El Kala’s coast (Algerian Basin)

The use of bio-optical methods and high spatial and spectral resolution satellite sensors has enabled the estimation of chlorophyll-a (Chl-a) on a large spatiotemporal scale. However, their application on a regional scale necessitates validation and, in certain cases, calibration, especially when it comes to coastal waters. This study was carried out in the eastern Algerian coastal waters to evaluate the relevance of Sentinel-3 satellite data in Annaba Bay and El Kala’s coast. Sentinel-3 OLCI (Ocean and Land Color Instrument) Level-2 products (OC4Me and Chl-a NN) and the Chl-a generated by applying the C2RCC processor on Level-1B (Chl-a C2RCC) were compared to the measured Chl-a at fixed locations. In addition, nine other Chl-a algorithms have been tested. Finally, six of those later were locally calibrated. The results demonstrate the good performance of the initially available products (Chl-a C2RCC and Chl-a NN), followed by the blue–green algorithms (OC4, OC5, OC6, MedOC4, and ADOC4), whereas the near-infrared (NIR) algorithms (2B-OLCI, 3B-OLCI, and G2B) show underwhelming performance and very poor prediction of the Chl-a concentration. Even though the OC4Me has the best correlation (r = 0.9820), it also has the highest MAE, RMSD, and BIAS (0.628, 1.264, and 0.5696, respectively). The fit of the NIR methods has been significantly refined. Nevertheless, their performance is still inadequate for the estimation of the Chl-a in our region. The performance of the calibrated OCx was improved, with the OC6-AW at the top of the list (r = 0.9881, MAE = 0.08). The latter outperforms the existing ones, which improves the estimation of Chl-a based on the Sentinel-3 OLCI data in our study area.

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

The study aims to assess variations in spatio-temporal characteristics of water quality parameters from three tropical estuaries, namely Muri-Ganga, Saptamukhi, and Hooghly, in the western portion of the Indian Sundarbans. Reliable retrieval of near-surface concentration of water quality parameters such as Chlorophyll-a, SST & TSM from diverse aquatic ecosystems with broad ranges of tropical requirements has always remained a complex issue. In this study, application of Case 2 Regional Colour Correction (C2RCC) processor has been tested for its accuracy across different bio-optical regimes in both inland and coastal waters. Satellite images for the same period were also collected and analysed using the C2RCC processing sequence to retrieve parameters like the depth of water, surface reflectance, water temperature, inherent optical properties (IOPs), chlorophyll-a, salinity, total suspended matter (TSM), etc., using the SNAP software. In situ sampling from specific locations within these estuaries and water quality analysis were conducted for the period 2017–2019. The OLCI retrieved datasets were compared and corroborated with field survey datasets. It was observed that the highest amount of TSM was recorded at Diamond Harbour during the 2018 pre-monsoon season (301.40 mg/L field-based value and 308.54 mg/L estimated value). Similarly, chlorophyll-a had higher concentrations throughout the monsoon season (3.03 mg m−3, (field survey), and 2.96 mg m−3, (estimated) at Fraserganj and Sagar south points. A very good correlation was observed for all seasons for Chl-a (r = 0.829) and TSM (r = 0.924) between the OLCI data and in situ measurements. Higher correlation and significant ‘r’ values highlight the importance of having both field-based as well as remotely-sensed information in understanding any dynamic system in a sustained manner. Results also confirm that the water quality model using OLCI Chl-a and TSM products outperforms conventional techniques. The study demonstrates the efficacy of using Sentinel 3 OCLI data for shallow marine and estuarine remote sensing applications, especially for monitoring TSM and Chl-a concentrations.

Increase in chlorophyll-a concentration in Lake Taihu from 1984 to 2021 based on Landsat observations

Lake Taihu, located in a densely populated and highly industrialized area in eastern China, has experienced dramatic changes in water quality since the reform and opening-up in the 1980s. Landsat data can be used to trace water quality changes over approximately 40 years. However, chlorophyll-a (Chla) estimation, which characterizes the trophic status, has not been thoroughly explored (especially in turbid water using wide bandwidth Landsat) due to the interference of suspended particulate matter (SPM) to Chla. In this study, we used Landsat TM/OLI for turbid water Chla inversion and to analyze the spatiotemporal variation of Chla in Lake Taihu for 38 years and its influencing factors. An optical classification algorithm based on Rrs(green)/Rrs(red) was used to exclude highly turbid waters dominated by SPM; Chla was estimated only in waters with low SPM. We constructed an exponential estimation model based on Rrs(NIR)/Rrs(red), and verified the accuracy of the model using the measured Chla synchronized with satellite data. The model was applied to Landsat images to calculate the Chla concentration in Lake Taihu during 1984–2021, and its spatiotemporal distribution patterns were further analyzed. Spatially, the Chla concentrations in the western and northern regions of Lake Taihu were higher than those in other regions, probably because these areas are estuaries with large exogenous pollutant discharge and more nutrients are imported from exogenous sources. Chla showed an overall significant upward trend from 1984 to 2021 probably because of temperature rise, wind speed reduction, and nutrient increase. The results of the spatial and temporal variation of Chla and the influencing factors in this study provide supporting data for eutrophication monitoring and management in Lake Taihu. The proposed Chla estimation method can be extended to assess the spatial and temporal distribution of eutrophication in other inland waters with similar optical properties.

TRANSFORMATION OF THE NORMALIZED DIFFERENCE CHLOROPHYLL INDEX TO RETRIEVE CHLOROPHYLL-A CONCENTRATIONS IN MANILA BAY

Abstract. Manila Bay is one of the most significant harbors in the region, enabling commerce and trade between the Philippines and other countries. Its abundant natural resources have provided for generations of its inhabitants and have driven socio-economic development for centuries. Like other water bodies adjacent to highly urbanized cities, the increased organic and nutrient loading from untreated domestic, industrial, and agricultural wastes resulted to further degradation of its water quality. While frequent water quality monitoring is ideal, data from traditional field sampling methods might not be sufficient to assess the spatial and temporal variations of water quality in Manila Bay. Remote sensing fills the need for a frequent and full overview of the bay’s water quality. Sentinel-3 images were initially processed through the Case 2 Regional CoastColour (C2RCC) model to retrieve the remote sensing reflectance. The Normalized Difference Chlorophyll Index (NDCI) was computed for the images and were modeled against the C2RCC-derived chlorophyll-a estimates. From this, we were able to get a general equation (TNDCI – transformed NDCI) to retrieve chlorophyll-a concentrations from two reflectance bands (Oa8 and Oa11). TNDCI gave an R2 of 0.9 and RMSE = 5.12 µg/L as compared to C2RCC values, and an R2 = 0.85 and RMSE = 2.44 µg/L with field data.

Development of chlorophyll-a concentration estimation algorithm for turbid productive inland waters in India

Estimation of Chlorophyll-a (Chl-a) concentration from remote-sensing reflectance derived from satellite imagery is highly effective in monitoring the spatial and temporal optical variations in water bodies. In turbid, productive and inland waters, implementation of traditional Chl-a estimation algorithms using the blue and green wavelength bands results in higher error in estimated Chl-a. To minimize the absorption and scattering effects of other water constituents like colored dissolved organic matter and particulate matter, the wavelengths bands in red and Near Infrared (NIR) are generally used for estimating Chl-a concentration in productive waters. Owing to a widely observed optical variability in natural waters, a single universal algorithm is not applicable to all waterbodies, hence, regional tuning of algorithms is necessary for estimating Chl-a in turbid and inland productive waters. Here, a 2-band based algorithm of the form and a 3-band algorithm of form were tuned for Chl-a estimation based on field measurements carried out in three water bodies, Chilika Lake of Orissa, Ujjani Reservoir of Maharashtra, Vallabh Sagar Reservoir of Gujarat, India, wherein the observed Chl-a concentration range is 5–33 mg/m3. Together with measured in situ reflectance spectra and concurrent Chl-a concentrations, tuning of red-NIR algorithms is performed. The optimal wavelengths to estimate Chl-a found by the proximal sensing method are = 691 nm and = 667 nm in case of 2-band red-NIR type algorithm. For 3-band red-NIR algorithm, the optimal wavelengths are = 670 nm, = 696 nm and = 740 nm. The mean absolute percentage error (MAPE) and root-mean-square error (RMSE) values obtained for the estimated Chl-a using the tuned 2-band red-NIR algorithm are 29.9% and 5.35 respectively. Similarly, RMSE and MAPE values for Chl-a concentration estimated using the 3-band tuned red-NIR algorithm are 4.24 and 22.5%, respectively. The tuned 2-band and 3-band algorithms resulted in better Chl-a estimates than the fourteen blue-green and red-NIR algorithms from previous studies. The 2-band and 3-band algorithms were re-calibrated to wavelengths present in the ocean land color imager (OLCI) sensor on Sentinel-3 satellite to monitor these water bodies. Spatio-temporal analysis of OLCI imagery derived Chl-a concentrations over Chilika lake (2019 and 2020) indicated seasonal variations. The tuned 3-band red-NIR algorithm exhibited robust performance in all seasons despite wide variations in colored dissolved organic matter and suspended sediments. The tuned red-NIR algorithms are useful in monitoring inland turbid productive waters with upcoming missions such as Plankton, Aerosol, Cloud Ecosystem (PACE) and Oceansat-3.

Per-Pixel Uncertainty Quantification and Reporting for Satellite-Derived Chlorophyll-a Estimates via Mixture Density Networks

Mixture density networks (MDNs) have emerged as a powerful tool for estimating water-quality indicators, such as chlorophyll-a (Chl <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$a$ </tex-math></inline-formula> ) from multispectral imagery. This study validates the use of an uncertainty metric calculated directly from Chl <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$a$ </tex-math></inline-formula> estimates of the MDNs. We consider multispectral remote sensing reflectance spectra ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\text {rs}}$ </tex-math></inline-formula> ) for three satellite sensors commonly used in aquatic remote sensing, namely, the ocean and land colour instrument (OLCI), multispectral instrument (MSI), and operational land imager (OLI). First, a study on a labeled database of colocated in situ Chl <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$a$ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\text {rs}}$ </tex-math></inline-formula> measurements clearly illustrates that the suggested uncertainty metric accurately captures the reduced confidence associated with test data, which is drawn for a different distribution than the training data. This change in distribution maybe due to: 1) random noise; 2) uncertainties in the atmospheric correction; and 3) novel (unseen) data. The experiments on the labeled in situ dataset show that the estimated uncertainty has a correlation with the expected predictive error and can be used as a bound on the predictive error for most samples. To illustrate the ability of the MDNs in generating consistent products from multiple sensors, per-pixel uncertainty maps for three near-coincident images of OLCI, MSI, and OLI are produced. The study also examines temporal trends in OLCI-derived Chl <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$a$ </tex-math></inline-formula> and the associated uncertainties at selected locations over a calendar year. Future work will include uncertainty estimation from MDNs with a multiparameter retrieval capability for hyperspectral and multispectral imagery.