Globally validated non-unique inversion framework to estimate optically active water quality indicators using in situ and space-borne hyperspectral data sets

Abstract

As water quality is a combination of multiple optically active parameters, there is a growing interest in probabilistic models to predict water quality. This study aims to add to the water quality prediction studies by introducing ensemble learning with deep learning-based mixture density networks with multiple probabilistic Gaussian distributions. We named the approach as Ensembled Gaussian Mixture Density Network (GMDN). Many existing water quality algorithms rely on localized data sets, which limits their applicability. This research addresses this by developing and evaluating the proposed model using the global in situ water quality data set GLORIA (Global Reflectance community data set for Imaging and optical sensing of Aquatic environments). We focused on estimating two key biogeochemical components (BPs): Total Suspended Solids (TSS) and Chlorophyll-a(Chla), along with one inherent optical property (IoP), the absorption coefficient of colored dissolved organic matter (α_CDOM). The proposed approach performs quite reliably when evaluated on the data samples of individual countries. The GMDN algorithm has been fine-tuned on the satellite-matchup for the river Ganga near Varanasi city. The fine-tuning was implemented using the remote sensing reflectance (R_rs) of the spaceborne hyperspectral data set PRISMA (PRecursore IperSpettrale della Missione Applicativa). The contribution of the riverbed floor to the Rrs of PRISMA has been computed using physics-based simulations in the Water Color Simulator (WASI). Overall, the simultaneous use of multiple probabilistic distributions and ensembled architectures improves the predictive accuracy of WQ parameters compared to the existing operational algorithms.

Graphical abstract

Keywords

Biogeochemical components (BP) / Inherent Optical Properties (IoP) / GLORIA / PRISMA / Water quality

Highlight

	● Three WQ parameters have been retrieved simultaneously.
	● The ensemble approach outperforms other algorithms.
	● Accuracy of the architecture has been validated across nations.
	● Bottom reflectance has been estimated using WASI.
	● Satellite matchup confirms the robustness of fine-tuned model.

References

[1] Albert A, Mobley C D. (2003). An analytical model for subsurface irradiance and remote sensing reflectance in deep and shallow case-2 waters. Optics Express, 11(22): 2873–2890

[2] Arabi B, Salama M S, van der Wal D, Pitarch J, Verhoef W. (2020). The impact of sea bottom effects on the retrieval of water constituent concentrations from MERIS and OLCI images in shallow tidal waters supported by radiative transfer modeling. Remote Sensing of Environment, 237: 111596

[3] Balasubramanian S V, Pahlevan N, Smith B, Binding C, Schalles J, Loisel H, Gurlin D, Greb S, Alikas K, Randla M. . (2020). Robust algorithm for estimating total suspended solids (TSS) in inland and nearshore coastal waters. Remote Sensing of Environment, 246: 111768

[4] Balch W, Huntington T, Aiken G, Drapeau D, Bowler B, Lubelczyk L, Butler K. (2016). Toward a quantitative and empirical dissolved organic carbon budget for the gulf of maine, a semienclosed shelf sea. Global Biogeochemical Cycles, 30(2): 268–292

[5] Brando V E, Dekker A G. (2003). Satellite hyperspectral remote sensing for estimating estuarine and coastal water quality. IEEE Transactions on Geoscience and Remote Sensing, 41(6): 1378–1387

[6] Burton C, Stubbs S, Onyisi P. (2021). Mixture density network estimation of continuous variable maximum likelihood using discrete training samples. European Physical Journal C, 81(7): 662

[7] Cao X, Zhang J, Meng H, Lai Y, Xu M. (2023). Remote sensing inversion of water quality parameters in the Yellow River Delta. Ecological Indicators, 155: 110914

[8] Chang N B, Xuan Z, Yang Y J. (2013). Exploring spatiotemporal patterns of phosphorus concentrations in a coastal bay with MODIS images and machine learning models. Remote Sensing of Environment, 134: 100–110

[9] Chauhan M S, Dikshit P K S, Dwivedi S B. (2015). Modeling of discharge distribution in bend of Ganga river at Varanasi. Computational Water, Energy, and Environmental Engineering, 4(3): 25–37

[10] Chen J, Chen S, Fu R, Li D, Jiang H, Wang C, Peng Y, Jia K, Hicks B J. (2022). Remote sensing big data for water environment monitoring: current status, challenges, and future prospects. Earth’s Future, 10(2): e2021EF002289

[11] Chen Y, Zheng G, Wang X, Chen X. (2013). Retrieval of chlorophyll-a concentration with multi-sensor data by GSM01 merging algorithm. Journal of Geo-Information Science, 15(6): 911–917

[12] DekkerAPinnel NGegePBriottetXPetersS TurpieKSterckx SCostaMGiardinoCBrandoV, . (2018). Feasibility study of an aquatic ecosystem Earth Observing System. Canberra, Australia: CSIRO

[13] Doelle O, Klinkenberg N, Amthor A, Ament C. (2023). Probabilistic intraday PV, power forecast using ensembles of deep Gaussian mixture density networks. Energies, 16(2): 646

[14] Du Y, Song K, Wang Q, Li S, Wen Z, Liu G, Tao H, Shang Y, Hou J, Lyu L, Zhang B. (2022). Total suspended solids characterization and management implications for lakes in East China. Science of the Total Environment, 806: 151374

[15] Fan Y, Li W, Chen N, Ahn J H, Park Y J, Kratzer S, Schroeder T, Ishizaka J, Chang R, Stamnes K. (2021). OC-SMART: a machine learning based data analysis platform for satellite ocean color sensors. Remote Sensing of Environment, 253: 112236

[16] Fei Z, Huang Z, Zhang X. (2023). Voltage and temperature information ensembled probabilistic battery health evaluation via deep Gaussian mixture density network. Journal of Energy Storage, 73: 108587

[17] Ficek D, Zapadka T, Dera J. (2011). Remote sensing reflectance of Pomeranian lakes and the Baltic. Oceanologia, 53(4): 959–970

[18] Gaur S, Das N, Bhattacharjee R, Ohri A, Patra D. (2023). A novel band selection architecture to propose a built-up index for hyperspectral sensor PRISMA. Earth Science Informatics, 16(1): 887–898

[19] Gege P. (2004). The water color simulator WASI: an integrating software tool for analysis and simulation of optical in situ spectra. Computers & Geosciences, 30(5): 523–532

[20] Gordon H R. (1978). Removal of atmospheric effects from satellite imagery of the oceans. Applied Optics, 17(10): 1631–1636

[21] Gordon H R, Brown O B, Jacobs M M. (1975). Computed relationships between the inherent and apparent optical properties of a flat homogeneous ocean. Applied Optics, 14(2): 417–427

[22] Gurlin D, Gitelson A A, Moses W J (2011). Remote estimation of chl-a concentration in turbid productive waters—return to a simple two-band NIR-red model? Remote Sensing of Environment,115(12): 3479–3490

[23] He B, Zhang W, Qiao X, Su Z. (2015). A study on remote sensing retrieval of suspended sediment concentration in Middle Yangtze River Based on A, FOASVM Method. Resources and Environment in the Yangtze Basin, 24(4): 647–652

[24] Khan S, Sinha R, Whitehead P, Sarkar S, Jin L, Futter M N. (2018). Flows and sediment dynamics in the Ganga River under present and future climate scenarios. Hydrological Sciences Journal, 63(5): 763–782

[25] LeeZ P (2006). Remote Sensing of Inherent Optical Properties: Fundamentals Tests of Algorithms, and Applications. Plymouth: International Ocean Colour Coordinating Group (IOCCG)

[26] Lehmann M K, Gurlin D, Pahlevan N, Alikas K, Conroy T, Anstee J, Balasubramanian S V, Barbosa C C, Binding C, Bracher A. . (2023). GLORIA-A globally representative hyperspectral in situ dataset for optical sensing of water quality. Scientific Data, 10(1): 100

[27] Li W. (2009). Method of water quality remote sensing and its application. Energy & Environment, 5(5): 62–64

[28] Lima T M A D, Giardino C, Bresciani M, Barbosa C C F, Fabbretto A, Pellegrino A, Begliomini F N. (2023). Assessment of estimated phycocyanin and chlorophyll-a concentration from prisma and olci in brazilian inland waters: a comparison between semi-analytical and machine learning algorithms. Remote Sensing, 15(5): 1299

[29] Liu Y, Huang H, Yan L, Yang X, Bi H, Zhang Z. (2020). Particle size parameters of particulate matter suspended in coastal waters and their use as indicators of typhoon influence. Remote Sensing, 12(16): 2581

[30] Jeba Dev P, Anna Geevarghese G, Purvaja R, Ramesh R. (2022). Measurement of in-vivo spectral reflectance of bottom types: implications for remote sensing of shallow waters. Advances in Space Research, 69(12): 4240–4251

[31] Ma R, Duan H, Liu Q, Loiselle S A. (2011). Approximate bottom contribution to remote sensing reflectance in Taihu Lake China. Journal of Great Lakes Research, 37(1): 18–25

[32] Ma Y, Song K, Wen Z, Liu G, Shang Y, Lyu L, Du J, Yang Q, Li S, Tao H, Hou J. (2021). Remote sensing of turbidity for lakes in northeast China using Sentinel-2 images with machine learning algorithms. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14: 9132–9146

[33] McCluskey E, Brewin R J, Vanhellemont Q, Jones O, Cummings D, Tilstone G, Jackson T, Widdicombe C, Woodward E M, Harris C. . (2022). On the seasonal dynamics of phytoplankton chlorophyll-a concentration in nearshore and offshore waters of Plymouth, in the english channel: Enlisting the help of a surfer. Oceans, 3(2): 125–146

[34] Meng W, Zhang N, Zhang Y, Zheng B. (2009). Integrated assessment of river health based on water quality, aquatic life and physical habitat. Journal of Environmental Sciences, 21(8): 1017–1027

[35] MobleyC D (1995). Hydrolight 3.0 Users’ Guide. Menlo Park: SRI International

[36] Mohseni F, Saba F, Mirmazloumi S M, Amani M, Mokhtarzade M, Jamali S, Mahdavi S. (2022). Ocean water quality monitoring using remote sensing techniques: a review. Marine Environmental Research, 180: 105701

[37] Morley S K, Brito T V, Welling D T. (2018). Measures of model performance based on the log accuracy ratio. Space Weather, 16(1): 69–88

[38] Najah A, Al-Shehhi M R. (2021). Performance of the ocean color algorithms: QAA GSM, and GIOP in inland and coastal waters. Remote Sensing in Earth Systems Sciences, 4(4): 235–248

[39] Nechad B, Ruddick K G, Park Y. (2010). Calibration and validation of a generic multisensor algorithm for mapping of total suspended matter in turbid waters. Remote Sensing of Environment, 114(4): 854–866

[40] Neil C, Spyrakos E, Hunter P D, Tyler A N. (2019). Corrigendum to “A global approach for chlorophyll-a retrieval across optically complex inland waters based on optical water types”. Remote Sensing of Environment, 229: 159–178

[41] Nima C, Frette Ø, Hamre B, Stamnes J J, Chen Y C, Sørensen K, Norli M, Lu D, Xing Q, Muyimbwa D. . (2019). CDOM absorption properties of natural water bodies along extreme environmental gradients. Water, 11(10): 1988

[42] Novoa S, Doxaran D, Ody A, Vanhellemont Q, Lafon V, Lubac B, Gernez P. (2017). Atmospheric corrections and multi-conditional algorithm for multi-sensor remote sensing of suspended particulate matter in low-to-high turbidity levels coastal waters. Remote Sensing, 9(1): 61

[43] O’Reilly J E, Werdell P J. (2019). Chlorophyll algorithms for ocean color Sensors-OC4 OC5 & OC6. Remote Sensing of Environment, 229: 32–47

[44] O’Shea R E, Pahlevan N, Smith B, Boss E, Gurlin D, Alikas K, Kangro K, Kudela R M, Vaičiūtė D. (2023). A hyperspectral inversion framework for estimating absorbing inherent optical properties and biogeochemical parameters in inland and coastal waters. Remote Sensing of Environment, 295: 113706

[45] O’Shea R E, Pahlevan N, Smith B, Bresciani M, Egerton T, Giardino C, Li L, Moore T, Ruiz-Verdu A, Ruberg S. . (2021). Advancing cyanobacteria biomass estimation from hyperspectral observations: demonstrations with HICO and PRISMA imagery. Remote Sensing of Environment, 266: 112693

[46] Odermatt D, Gitelson A, Brando V E, Schaepman M. (2012). Review of constituent retrieval in optically deep and complex waters from satellite imagery. Remote Sensing of Environment, 118: 116–126

[47] Oubelkheir K, Clementson L A, Webster I T, Ford P W, Dekker A G, Radke L C, Daniel P. (2006). Using inherent optical properties to investigate biogeochemical dynamics in a tropical macrotidal coastal system. Journal of Geophysical Research. Oceans, 111(C7): C07021

[48] Pahlevan N, Smith B, Alikas K, Anstee J, Barbosa C, Binding C, Bresciani M, Cremella B, Giardino C, Gurlin D. . (2022). Simultaneous retrieval of selected optical water quality indicators from Landsat-8 Sentinel-2, and Sentinel-3. Remote Sensing of Environment, 270: 112860

[49] Pahlevan N, Smith B, Schalles J, Binding C, Cao Z, Ma R, Alikas K, Kangro K, Gurlin D, Hà N. . (2020). Seamless retrievals of chlorophyll-a from sentinel-2 (MSI) and sentinel-3 (OLCI) in inland and coastal waters: a machine-learning approach. Remote Sensing of Environment, 240: 111604

[50] Palmer S C, Kutser T, Hunter P D. (2015). Remote sensing of inland waters: challenges, progress and future directions. Remote Sensing of Environment, 157: 1–8

[51] Pellegrino A, Fabbretto A, Bresciani M, de Lima T M A, Braga F, Pahlevan N, Brando V E, Kratzer S, Gianinetto M, Giardino C. (2023). Assessing the accuracy of prisma standard reflectance products in globally distributed aquatic sites. Remote Sensing, 15(8): 2163

[52] Qi L, Hu C, Duan H, Barnes B B, Ma R. (2014). An EOF-based algorithm to estimate chlorophyll a concentrations in Taihu Lake from MODIS land-band measurements: implications for near real-time applications and forecasting models. Remote Sensing, 6(11): 10694–10715

[53] Sahoo D P, Sahoo B, Tiwari M K. (2022). MODIS-Landsat fusion-based single-band algorithms for TSS and turbidity estimation in an urban-waste-dominated river reach. Water Research, 224: 119082

[54] Saikia S K, Das D N. (2011). Diversity and productivity (chlorophyll-a and biomass) of periphyton on natural and artificial substrates from wetland ecosystem. Journal of Wetlands Ecology, 5: 1–9

[55] Salem S I, Higa H, Kim H, Kobayashi H, Oki K, Oki T. (2017). Assessment of chlorophyll-a algorithms considering different trophic statuses and optimal bands. Sensors, 17(8): 1746

[56] ShuklaU K (2013). Varanasi and the Ganga River: a geological perspective. In: Jayaswal V, ed. Varanasi: Myths and Scientific Studies. New Delhi: Aryan Books International, 100–113

[57] Smith B, Pahlevan N, Schalles J, Ruberg S, Errera R, Ma R, Giardino C, Bresciani M, Barbosa C, Moore T. . (2021). A chlorophyll-a algorithm for Landsat-8 based on mixture density networks. Frontiers in Remote Sensing, 1: 623678

[58] Smith M E, Robertson Lain L, Bernard S. (2018). An optimized chlorophyll a switching algorithm for MERIS and OLCI in phytoplankton-dominated waters. Remote Sensing of Environment, 215: 217–227

[59] Spyrakos E, O’Donnell R, Hunter P D, Miller C, Scott M, Simis S G, Neil C, Barbosa C C, Binding C E, Bradt S. . (2018). Optical types of inland and coastal waters. Limnology and Oceanography, 63(2): 846–870

[60] Tan H, Oishi T, Tanaka A, Doerffer R, Tan Y. (2017). Chlorophyll-a specific volume scattering function of phytoplankton. Optics Express, 25(12): A564–A573

[61] Tiyasha T M, Tung C M. (2020). A survey on river water quality modelling using artificial intelligence models: 2000–2020. Journal of Hydrology, 585: 124670

[62] Vangi E, D’Amico G, Francini S, Giannetti F, Lasserre B, Marchetti M, Chirici G. (2021). The new hyperspectral satellite PRISMA: imagery for forest types discrimination. Sensors, 21(4): 1182

[63] Xu Y, Feng L, Zhao D, Lu J. (2020). Assessment of Landsat atmospheric correction methods for water color applications using global AERONET-OC data. International Journal of Applied Earth Observation and Geoinformation, 93: 102192

[64] Zhang S. (2008). Parimputation: From imputation and null-imputation to partially imputation. IEEE Intelligent Informatics Bulletin, 9(1): 32–38

[65] Zhang S. (2012). Nearest neighbor selection for iteratively kNN imputation. Journal of Systems and Software, 85(11): 2541–2552

[66] Zhao M, Bai Y, Li H, He X, Gong F, Li T. (2022). Fluorescence line height extraction algorithm for the geostationary ocean color imager. Remote Sensing, 14(11): 2511

[67] Zhou D, Wang D. (2015). Quantitative estimation of chlorophyll-a and suspended solids in Taihu based on Landsat TM. Environmental Science & Technology, 38(6P): 362–367

[68] Zhu W, Yu Q. (2012). Inversion of chromophoric dissolved organic matter from EO-1 Hyperion imagery for turbid estuarine and coastal waters. IEEE Transactions on Geoscience and Remote Sensing, 51(6): 3286–3298

RIGHTS & PERMISSIONS

Higher Education Press 2025