A satellite-based hybrid model for trophic state evaluation in inland waters across China

Eutrophication is one of the major threats to the inland water ecosystem. Satellite remote sensing provides a promising way to monitor trophic state at large spatial scale in an efficient manner. Currently, most satellite-based trophic state evaluation approaches have focused on water quality parameters retrieval (e.g., transparency, chlorophyll-a), based on which trophic state was evaluated. However, the retrieval accuracies of individual parameter do not meet the demand for accurate trophic state evaluation, especially for the turbid inland waters. In this study, we proposed a novel hybrid model to estimate trophic state index (TSI) by integrating multiple spectral indices associated with different eutrophication level based on Sentinel-2 imagery. The TSI estimated by the proposed method agreed well with the in-situ TSI observations, with root mean square error (RMSE) of 6.93 and mean absolute percentage error (MAPE) of 13.77%. Compared with the independent observations from Ministry of Ecology and Environment, the estimated monthly TSI also showed good consistency (RMSE=5.91,MAPE=10.66%). Furthermore, the congruent performance of the proposed method in the 11 sample lakes (RMSE=5.91,MAPE=10.66%) and the 51 ungauged lakes (RMSE=7.16,MAPE=11.56%) indicated the favorable model generalization. The proposed method was then applied to assess the trophic state of 352 permanent lakes and reservoirs across China during the summers of 2016-2021. It showed that 10%, 60%, 28%, and 2% of the lakes/reservoirs are in oligotrophic, mesotrophic, light eutrophic, and middle eutrophic states respectively. Eutrophic waters are concentrated in the Middle-and-Lower Yangtze Plain, the Northeast Plain, and the Yunnan-Guizhou Plateau. Overall, this study improved the trophic state representativeness and revealed trophic state spatial distribution of Chinese inland waters, which has the significant meanings for aquatic environment protection and water resource management.

Estimating pixel-level uncertainty in ocean color retrievals from MODIS

The spectral distribution of marine remote sensing reflectance, R_rs, is the fundamental measurement of ocean color science, from which a host of bio-optical and biogeochemical properties of the water column can be derived. Estimation of uncertainty in these derived properties is thus dependent on knowledge of the uncertainty in satellite-retrieved R_rs (uc(R_rs)) at each pixel. Uncertainty in R_rs, in turn, is dependent on the propagation of various uncertainty sources through the R_rs retrieval process, namely the atmospheric correction (AC). A derivative-based method for uncertainty propagation is established here to calculate the pixel-level uncertainty in R_rs, as retrieved using NASA's multiple-scattering epsilon (MSEPS) AC algorithm and verified using Monte Carlo (MC) analysis. The approach is then applied to measurements from the Moderate Resolution Imaging Spectroradiometer (MODIS) on the Aqua satellite, with uncertainty sources including instrument random noise, instrument systematic uncertainty, and forward model uncertainty. The uc(R_rs) is verified by comparison with statistical analysis of coincident retrievals from MODIS and in situ R_rs measurements, and our approach performs well in most cases. Based on analysis of an example 8-day global products, we also show that relative uncertainty in R_rs at blue bands has a similar spatial pattern to the derived concentration of the phytoplankton pigment chlorophyll-a (chl-a), and around 7.3%, 17.0%, and 35.2% of all clear water pixels (chl-a ≤ 0.1 mg/m³) with valid uc(R_rs) have a relative uncertainty ≤ 5% at bands 412 nm, 443 nm, and 488 nm respectively, which is a common goal of ocean color retrievals for clear waters. While the analysis shows that uc(R_rs) calculated from our derivative-based method is reasonable, some issues need further investigation, including improved knowledge of forward model uncertainty and systematic uncertainty in instrument calibration.

Fusion of Drone-Based RGB and Multi-Spectral Imagery for Shallow Water Bathymetry Inversion

Shallow bathymetry inversion algorithms have long been applied in various types of remote sensing imagery with relative success. However, this approach requires that imagery with increased radiometric resolution in the visible spectrum be available. The recent developments in drones and camera sensors allow for testing current inversion techniques on new types of datasets with centimeter resolution. This study explores the bathymetric mapping capabilities of fused RGB and multispectral imagery as an alternative to costly hyperspectral sensors for drones. Combining drone-based RGB and multispectral imagery into a single cube dataset provides the necessary radiometric detail for shallow bathymetry inversion applications. This technique is based on commercial and open-source software and does not require the input of reference depth measurements in contrast to other approaches. The robustness of this method was tested on three different coastal sites with contrasting seafloor types with a maximum depth of six meters. The use of suitable end-member spectra, which are representative of the seafloor types of the study area, are important parameters in model tuning. The results of this study are promising, showing good correlation (R2 > 0.75 and Lin’s coefficient > 0.80) and less than half a meter average error when they are compared with sonar depth measurements. Consequently, the integration of imagery from various drone-based sensors (visible range) assists in producing detailed bathymetry maps for small-scale shallow areas based on optical modelling.

Water Mixing Conditions Influence Sentinel-2 Monitoring of Chlorophyll Content in Monomictic Lakes

Prompt estimation of phytoplankton biomass is critical in determining the ecological quality of freshwaters. Remote Sensing (RS) may provide new opportunities to integrate with situ traditional monitoring techniques. Nonetheless, wide regional and temporal variability in freshwater optical constituents makes it difficult to design universally applicable RS protocols. Here, we assessed the potential of two neural networks-based models, namely the Case 2 Regional CoastColour (C2RCC) processor and the Mixture Density Network (MDN), applied to MSI Sentinel-2 data for monitoring Chlorophyll (Chl) content in three monomictic volcanic lakes while accounting for the effect of their specific water circulation pattern on the remotely-sensed and in situ data relation. Linear mixed models were used to test the relationship between the remote sensing indices calculated through C2RCC (INN) and MDN (IMDN), and in situ Chl concentration. Both indices proved to explain a large portion of the variability in the field data and exhibited a positive and significant relationship between Chl concentration and satellite data, but only during the mixing phase. The significant effect of the water circulation period can be explained by the low responsiveness of the RS approaches applied here to the low phytoplankton biomass, typical of the stratification phase. Sentinel-2 data proved their valuable potential for the remote sensing of phytoplankton in small inland water bodies, otherwise challenging with previous sensors. However, caution should be taken, since the applicability of such an approach on certain water bodies may depend on hydrological and ecological parameters (e.g., thermal stratification and seasonal nutrient availability) potentially altering RS chlorophyll detection by neural networks-based models, despite their alleged global validity.

Dynamic partitioning of tropical Indian Ocean surface waters using ocean colour data - management and modelling applications

Over the past few decades, partitioning of the surface ocean into ecologically-meaningful spatial domains has been approached using a range of data types, with the aim of improving our understanding of open ocean processes, supporting marine management decisions and constraining coupled ocean-biogeochemical models. The simplest partitioning method, which could provide low-latency information for managers at low cost, remains a purely optical classification based on ocean colour remote sensing. The question is whether such a simple approach has value. Here, the efficacy of optical classifications in constraining physical variables that modulate the epipelagic environment is tested for the tropical Indian Ocean, with a focus on the Chagos marine protected area (MPA). Using remote sensing data, it was found that optical classes corresponded to distinctive ranges of wind speed, wind stress curl, sea surface temperature, sea surface slope, sea surface height anomaly and geostrophic currents (Kruskal-Wallis and post-hoc Tukey honestly significantly different tests, α = 0.01). Between-class differences were significant for a set of sub-domains that resolved zonal and meridional gradients across the MPA and Seychelles-Chagos Thermocline Ridge, whereas between-domain differences were only significant for the north-south gradient (PERMANOVA, α = 0.01). A preliminary test of between-class differences in surface CO₂ concentrations from the Orbiting Carbon Observatory-2 demonstrated a small decrease in mean pCO₂ with increasing chlorophyll (chl), from 418 to 398 ppm. Simple optical class maps therefore provide an overview of growth conditions, the spatial distribution of resources - from which habitat fragmentation metrics can be calculated, and carbon sequestration potential. Within the 17 year study period, biotic variables were found to have decreased at up to 0.025%a^-1 for all optical classes, which is slower than reported elsewhere (Mann-Kendall-Sen regression, α = 0.01). Within the MPA, positive Indian Ocean Dipole conditions and negative Southern Oscillation Indices were weakly associated with decreasing chl, fluorescence line height (FLH), eddy kinetic energy, easterly wind stress and wind stress curl, and with increasing FLH/chl, sea surface temperature, SSH gradients and northerly wind stress, consistent with reduced surface mixing and increased stratification. The optical partitioning scheme described here can be applied in Google Earth Engine to support management decisions at daily or monthly scales, and potential applications are discussed.

Spectral and Radiometric Measurement Requirements for Inland, Coastal and Reef Waters

This paper studies the measurement requirements of spectral resolution and radiometric sensitivity to enable the quantitative determination of water constituents and benthic parameters for the majority of optically deep and optically shallow waters on Earth. The spectral and radiometric variability is investigated by simulating remote sensing reflectance (Rrs) spectra of optically deep water for twelve inland water scenarios representing typical and extreme concentration ranges of phytoplankton, colored dissolved organic matter and non-algal particles. For optically shallow waters, Rrs changes induced by variable water depth are simulated for fourteen bottom substrate types, from lakes to coastal waters and coral reefs. The required radiometric sensitivity is derived for the conditions that the spectral shape of Rrs should be resolvable with a quantization of 100 levels and that measurable reflection differences at at least one wavelength must occur at concentration changes in water constituents of 10% and depth differences of 20 cm. These simulations are also used to derive the optimal spectral resolution and the most sensitive wavelengths. Finally, the Rrs spectra and their changes are converted to radiances and radiance differences in order to derive sensor (noise-equivalent radiance) and measurement requirements (signal-to-noise ratio) at the water surface and at the top of the atmosphere for a range of solar zenith angles.

Evaluation of glint correction approaches for fine-scale ocean color measurements by lightweight hyperspectral imaging spectrometers

Low-power, lightweight, off-the-shelf imaging spectrometers, deployed on above-water fixed platforms or on low-altitude aerial drones, have significant potential for enabling fine-scale assessment of radiometrically derived water quality properties (WQPs) in oceans, lakes, and reservoirs. In such applications, it is essential that the measured water-leaving spectral radiances be corrected for surface-reflected light, i.e., glint. However, noise and spectral characteristics of these imagers, and environmental sources of fine-scale radiometric variability such as capillary waves, complicate the glint correction problem. Despite having a low signal-to-noise ratio, a representative lightweight imaging spectrometer provided accurate radiometric estimates of chlorophyll concentration-an informative WQP-from glint-corrected hyperspectral radiances in a fixed-platform application in a coastal ocean region. Optimal glint correction was provided by a spectral optimization algorithm, which outperformed both a hardware solution utilizing a polarizer and a subtractive algorithm incorporating the reflectance measured in the near infrared. In the same coastal region, this spectral optimization approach also provided the best glint correction for radiometric estimates of backscatter at 650 nm, a WQP indicative of suspended particle load.

The Fundamental Contribution of Phytoplankton Spectral Scattering to Ocean Colour: Implications for Satellite Detection of Phytoplankton Community Structure

There is increasing interdisciplinary interest in phytoplankton community dynamics as the growing environmental problems of water quality (particularly eutrophication) and climate change demand attention. This has led to a pressing need for improved biophysical and causal understanding of Phytoplankton Functional Type (PFT) optical signals, in order for satellite radiometry to be used to detect ecologically relevant phytoplankton assemblage changes. Biophysically and biogeochemically consistent phytoplankton Inherent Optical Property (IOP) models play an important role in achieving this understanding, as the optical effects of phytoplankton assemblage changes can be examined systematically in relation to the bulk optical water-leaving signal. The Equivalent Algal Populations (EAP) model is used here to investigate the source and magnitude of size- and pigment- driven PFT signals in the water-leaving reflectance, as well as the potential to detect these using satellite radiometry. This model places emphasis on the determination of biophysically consistent phytoplankton IOPs, with both absorption and scattering determined by mathematically cogent relationships to the particle complex refractive indices. All IOPs are integrated over an entire size distribution. A distinctive attribute is the model’s comprehensive handling of the spectral and angular character of phytoplankton scattering. Selected case studies and sensitivity analyses reveal that phytoplankton spectral scattering is most useful and the least ambiguous driver of the PFT signal. Key findings are that there is the most sensitivity in phytoplankton backscatter ( b b ϕ ) in the 1–6 μ m size range; the backscattering-driven signal in the 520 to 570 nm region is the critical PFT identifier at marginal biomass, and that, while PFT information does appear at blue wavelengths, absorption-driven signals are compromised by ambiguity due to biomass and non-algal absorption. Low signal in the red, due primarily to absorption by water, inhibits PFT detection here. The study highlights the need to quantitatively understand the constraints imposed by phytoplankton biomass and the IOP budget on the assemblage-related signal. A proportional phytoplankton contribution of approximately 40% to the total b b appears to a reasonable minimum threshold in terms of yielding a detectable optical change in R r s . We hope these findings will provide considerable insight into the next generation of PFT algorithms.

Optical inversions of the water column based on glider measurements

We demonstrate a method for estimating absorption and backscattering coefficients by inverting glider-measured profiles of the downwelling irradiance and upwelling radiance. The inversion method was validated against approximately 1,300 profiles of data from 22 glider missions within the Gulf of Maine over a 10 year period. The backscattering coefficient at 532 nm was estimated with a mean absolute error of 21% and bias of 0.01% compared to measured values. We could only quantitatively evaluate the absorption coefficient against the fluorometry data, but found that profiles of fluorescence and absorption were in quantitative agreement. With absorption and backscattering coefficients acting as a basis for studying the biogeochemical parameters of the constituents in the water column, these results show the potential of bio-optical gliders for studying marine ecosystems under varying sky conditions.

An overview of approaches and challenges for retrieving marine inherent optical properties from ocean color remote sensing

Ocean color measured from satellites provides daily global, synoptic views of spectral waterleaving reflectances that can be used to generate estimates of marine inherent optical properties (IOPs). These reflectances, namely the ratio of spectral upwelled radiances to spectral downwelled irradiances, describe the light exiting a water mass that defines its color. IOPs are the spectral absorption and scattering characteristics of ocean water and its dissolved and particulate constituents. Because of their dependence on the concentration and composition of marine constituents, IOPs can be used to describe the contents of the upper ocean mixed layer. This information is critical to further our scientific understanding of biogeochemical oceanic processes, such as organic carbon production and export, phytoplankton dynamics, and responses to climatic disturbances. Given their importance, the international ocean color community has invested significant effort in improving the quality of satellite-derived IOP products, both regionally and globally. Recognizing the current influx of data products into the community and the need to improve current algorithms in anticipation of new satellite instruments (e.g., the global, hyperspectral spectroradiometer of the NASA Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission), we present a synopsis of the current state of the art in the retrieval of these core optical properties. Contemporary approaches for obtaining IOPs from satellite ocean color are reviewed and, for clarity, separated based their inversion methodology or the type of IOPs sought. Summaries of known uncertainties associated with each approach are provided, as well as common performance metrics used to evaluate them. We discuss current knowledge gaps and make recommendations for future investment for upcoming missions whose instrument characteristics diverge sufficiently from heritage and existing sensors to warrant reassessing current approaches.

Discrimination of phytoplankton functional groups using an ocean reflectance inversion model

Ocean reflectance inversion models (ORMs) provide a mechanism for inverting the color of the water observed by a satellite into marine inherent optical properties (IOPs), which can then be used to study phytoplankton community structure. Most ORMs effectively separate the total signal of the collective phytoplankton community from other water column constituents; however, few have been shown to effectively identify individual contributions by multiple phytoplankton groups over a large range of environmental conditions. We evaluated the ability of an ORM to discriminate between Noctiluca miliaris and diatoms under conditions typical of the northern Arabian Sea. We: (1) synthesized profiles of IOPs that represent bio-optical conditions for the Arabian Sea; (2) generated remote-sensing reflectances from these profiles using Hydrolight; and (3) applied the ORM to the synthesized reflectances to estimate the relative concentrations of diatoms and N. miliaris. By comparing the estimates from the inversion model with those from synthesized vertical profiles, we identified those conditions under which the ORM performs both well and poorly. Even under perfectly controlled conditions, the absolute accuracy of ORM retrievals degraded when further deconstructing the derived total phytoplankton signal into subcomponents. Although the absolute magnitudes maintained biases, the ORM successfully detected whether or not Noctiluca miliaris appeared in the simulated water column. This quantitatively calls for caution when interpreting the absolute magnitudes of the retrievals, but qualitatively suggests that the ORM provides a robust mechanism for identifying the presence or absence of species.

Sensitivity in reflectance attributed to phytoplankton cell size: forward and inverse modelling approaches

Synoptic scale knowledge of the size structure of phytoplankton communities can offer insight in to primary ecosystem diversity and biogeochemical variability from operational to the decadal scales. Accordingly, obtaining estimates of size and other phytoplankton functional type descriptors within known confidence limits from remotely sensed data has become a major objective to extend the use of ocean colour data beyond chlorophyll a retrievals. Here, a new forward and inverse modelling structure is proposed to determine information about the cell size of phytoplankton communities using Standard size distributions of two layered spheres to derive a full suite of algal inherent optical properties for a coupled radiative transfer model. This new capability allows explicit quantification of the remote sensing reflectance signal attributable to changes in phytoplankton cell size. Inversion of this model reveals regions within the parameter space where ambiguity may limit potential of inversion algorithms. Validation of the algorithm within the Benguela upwelling system using independent data shows promise for ecosystem applications and further investigation of the interaction between phytoplankton functional types and optical signals. The results here suggest that the utility of assemblage related signals in spectral reflectance is highly sensitive to algal biomass, the presence of other absorbing and scattering constituents and the resultant constituent-specific inherent optical property budget. As such, optimal methods for determining phytoplankton size from (in situ or satellite) ocean colour data will likely rely on appropriately spectrally dense and optimised sensors, well characterised measurement errors including those from atmospheric correction, and an ability to appropriately limit ambiguity within the context of regional inherent optical properties.

Generalized ocean color inversion model for retrieving marine inherent optical properties

Ocean color measured from satellites provides daily, global estimates of marine inherent optical properties (IOPs). Semi-analytical algorithms (SAAs) provide one mechanism for inverting the color of the water observed by the satellite into IOPs. While numerous SAAs exist, most are similarly constructed and few are appropriately parameterized for all water masses for all seasons. To initiate community-wide discussion of these limitations, NASA organized two workshops that deconstructed SAAs to identify similarities and uniqueness and to progress toward consensus on a unified SAA. This effort resulted in the development of the generalized IOP (GIOP) model software that allows for the construction of different SAAs at runtime by selection from an assortment of model parameterizations. As such, GIOP permits isolation and evaluation of specific modeling assumptions, construction of SAAs, development of regionally tuned SAAs, and execution of ensemble inversion modeling. Working groups associated with the workshops proposed a preliminary default configuration for GIOP (GIOP-DC), with alternative model parameterizations and features defined for subsequent evaluation. In this paper, we: (1) describe the theoretical basis of GIOP; (2) present GIOP-DC and verify its comparable performance to other popular SAAs using both in situ and synthetic data sets; and, (3) quantify the sensitivities of their output to their parameterization. We use the latter to develop a hierarchical sensitivity of SAAs to various model parameterizations, to identify components of SAAs that merit focus in future research, and to provide material for discussion on algorithm uncertainties and future emsemble applications.

Estimation of hyperspectral inherent optical properties from in-water radiometry: error analysis and application to in situ data

An inverse algorithm is developed to retrieve hyperspectral absorption and backscattering coefficients from measurements of hyperspectral upwelling radiance and downwelling irradiance in vertically homogeneous waters. The forward model is the azimuthally averaged radiative transfer equation, efficiently solved by the EcoLight radiative transfer model, which includes the effects of inelastic scattering. Although this inversion problem is ill posed (the solution is ambiguous for retrieval of total scattering coefficients), unique and stable solutions can be found for absorption and backscattering coefficients. The inversion uses the attenuation coefficient at one wavelength to constrain the inversion, increasing the algorithm's stability and accuracy. Two complementary methods, Monte Carlo simulation and first-order error propagation, are used to develop uncertainty estimates for the retrieved absorption and backscattering coefficients. The algorithm is tested using both simulated light fields from a chlorophyll-based case I bio-optical model and radiometric field data from the 2008 North Atlantic Bloom Experiment. The influence of uncertainty in the radiometric quantities and additional model parameters on the inverse solution for absorption and backscattering is studied using a Monte Carlo approach, and an uncertainty budget is developed for retrievals. All of the required radiometric and inherent optical property measurements can be made from power-limited autonomous platforms. We conclude that hyperspectral measurements of downwelling irradiance and upwelling radiance, with a single-wavelength measurement of attenuation, can be used to estimate hyperspectral absorption to an accuracy of ±0.01 m(-1) and hyperspectral backscattering to an accuracy of ±0.0005 m(-1) from 350 to 575 nm.

Impact of signal-to-noise ratio in a hyperspectral sensor on the accuracy of biophysical parameter estimation in case II waters

Errors in the estimated constituent concentrations in optically complex waters due solely to sensor noise in a spaceborne hyperspectral sensor can be as high as 80%. The goal of this work is to elucidate the effect of signal-to-noise ratio (SNR) on the accuracy of retrieved constituent concentrations. Large variations in the magnitude and spectral shape of the reflectances from coastal waters complicate the impact of SNR on the accuracy of estimation. Due to the low reflectance of water, the actual SNR encountered for a water target is usually quite lower than the prescribed SNR. The low SNR can be a significant source of error in the estimated constituent concentrations. Simulated and measured at-surface reflectances were used in this study. A radiative transfer code, Tafkaa, was used to propagate the at-surface reflectances up and down through the atmosphere. A sensor noise model based on that of the spaceborne hyperspectral sensor HICO was applied to the at-sensor radiances. Concentrations of chlorophyll-a, colored dissolved organic matter, and total suspended solids were estimated using an optimized error minimization approach and a few semi-analytical algorithms. Improving the SNR by reasonably modifying the sensor design can reduce estimation uncertainties by 10% or more.

Inherent optical property estimation in deep waters

We develop two algorithms for determining two inherent optical properties (IOPs) from radiometric measurements in vertically homogeneous waters. The first algorithm is for estimation of the ratio of the backscattering to absorption coefficients from measurements of only the vertically upward radiance and the downward planar irradiance at depths where the light field is in the asymptotic regime. The second algorithm enables estimation of the absorption coefficient from measurement of the diffuse attenuation coefficient in the asymptotic regime after use of the first algorithm. Multiplication of the two estimates leads to an estimate for the backscattering coefficient. The algorithms, based upon the use of a simplified phase function and the asymptotic eigenmode, are shown to potentially provide good starting conditions for iteratively determining the absorption and backscattering coefficients of a wide variety of waters. The uncertainty in the estimates defines a subspace for IOPs that may reduce ambiguity in such iterative solutions. Because of the ease of estimating the backscattering to absorption ratio from in-water measurements, this IOP deserves further investigation as a proxy for biogeochemical quantities in the open ocean.

Biogeo-optics: particle optical properties and the partitioning of the spectral scattering coefficient of ocean waters

We propose a direct method of partitioning the particulate spectral scattering coefficient of the marine hydrosol based on the concurrent determination of the concentrations of particulate mineral and organic matter (the total mass of optically active scattering material exclusive of water) with the particulate spectral scattering coefficient. For this we derive a Model II multiple linear regression model. The multiple linear regression of the particulate spectral scattering coefficient against the independent variables, the concentrations of particulate inorganic matter and particulate organic matter, yields their mass-specific spectral scattering cross sections. The mass-specific spectral scattering cross section is simply the particle scattering cross section normalized to the particle mass, a fundamental optical efficiency parameter for the attenuation of electromagnetic radiation [Absorption and Scattering of Light by Small Particles, (Wiley-Interscience, 1983), pp. 80-81, 289]. It is possible to infer the optical properties of the suspended matter from the mass-specific spectral scattering cross sections. From these cross sections we partition the particulate spectral scattering coefficient into its major components.