Toward closure of upwelling radiance in coastal waters

Example of Monte Carlo Method Uncertainty Evaluation for Above-Water Ocean Colour Radiometry

We describe a method to evaluate an uncertainly budget for the in situ Ocean Colour Radiometric measurements. A Monte Carlo approach is chosen to propagate the measurement uncertainty inputs through the measurements model. The measurement model is designed to address instrument characteristics and uncertainty associated with them. We present the results for a particular example when the radiometers were fully characterised and then use the same data to show a case when such characterisation is missing. This, depending on the measurement and the wavelength, can increase the uncertainty value significantly; for example, the downwelling irradiance at 442.5 nm with fully characterised instruments can reach uncertainty values of 1%, but for the instruments without such characterisation, that value could increase to almost 7%. The uncertainty values presented in this paper are not final, as some of the environmental contributors were not fully evaluated. The main conclusion of this work are the significance of thoughtful instrument characterisation and correction for the most significant uncertainty contributions in order to achieve a lower measurements uncertainty value.

Re-examining the effect of particle phase functions on the remote-sensing reflectance

Even though it is well known that both the magnitude and detailed angular shape of scattering (phase function, PF), particularly in the backward angles, affect the color of the ocean, the current remote-sensing reflectance (R_rs) models typically account for the effect of its magnitude only through the backscattering coefficient (b_b). Using 116 volume scattering function (VSF) measurements previously collected in three coastal waters around the U.S. and in the water of the North Atlantic Ocean, we re-examined the effect of particle PF on R_rs in four scenarios. In each scenario, the magnitude of particle backscattering (i.e., b_bp) is known, but the knowledge on the angular shape of particle backscattering is assumed to increase from knowing nothing about the shape of particle PFs to partially knowing the particle backscattering ratio (B_p), the exact backscattering shape as defined by β˜_p(γ≥90°) (particle VSF normalized by the particle total scattering coefficient), and the exact backscattering shape as defined by the χ_p factor (particle VSF normalized by the particle backscattering coefficient). At sun zenith angle=30°, the nadir-viewed R_rs would vary up to 65%, 35%, 20%, and 10%, respectively, as the constraints on the shape of particle backscattering become increasingly stringent from scenarios 1 to 4. In all four scenarios, the R_rs variations increase with both viewing and sun angles and are most prominent in the direction opposite the sun. Our results show a greater impact of the measured particle PFs on R_rs than previously found, mainly because our VSF data show a much greater variability in B_p, β˜_p(γ≥90°), and χ_p than previously known. Among the uncertainties in R_rs due to the particle PFs, about 97% can be explained by χ_p, 90% by β˜_p(γ≥90°), and 27% by B_p. The results indicate that the uncertainty in ocean color remote sensing can be significantly constrained by accounting for χ_p of the VSFs.

Absorption correction and phase function shape effects on the closure of apparent optical properties

We present a closure experiment between new inherent optical properties (IOPs: absorption a, scattering b, backscattering b_b) and apparent optical properties (AOPs: remote-sensing reflectance R_rs, irradiance reflectance R, and anisotropic factor at nadir Q_n) data of Ionian and Adriatic seawaters, from very clear to turbid waters, ranging across one order of magnitude in R_rs. The internal consistency of the IOP-AOP matchups was investigated though radiative transfer closure. Using the in situ IOPs, we predicted the AOPs with the commercial radiative transfer solver Hydrolight. Closure was limited by two unresolved issues, one regarding processing of in situ data and the other related to radiative transfer modeling. First, different correction methods of the absorption data measured by the Wetlabs ac-s produced high variations in simulated reflectances, reaching 40% for the highest reflectances in our dataset. Second, the lack of detailed volume scattering function measurements forces us to adopt analytical functions that are consistent with a given particle backscattering ratio. The analytical phase functions named Fournier-Forand and two-term Kopelevich presented reasonable angular shapes with respect to measurements at a few backward angles. Between these phase functions, induced changes were within 4% for R_rs, within 11% for R, and within 10% for Q_n. Additionally, closure of Q_n was generally not successful considering radiometric uncertainties. Simulated Q_n overestimated low values and underestimated high values, especially at 665 nm, where Hydrolight was unable to predict measured Q_n values greater than 6 sr. The physical nature of Q_n makes this mismatch almost independent of the measured IOPs, thus precluding Q_n tuning by varying the former. The non-closure of Q_n might be caused by an inaccurate phase function and, to a lesser extent, by the modeling of the incoming radiance. For the future, this remains the task of accurate absorption and phase function measurements, especially at red wavelengths.

Optical closure in marine waters from in situ inherent optical property measurements

Optical closure using radiative transfer simulations can be used to determine the consistency of in situ measurements of inherent optical properties (IOPs) and radiometry. Three scattering corrections are applied to in situ absorption and attenuation profile data for a range of coastal and oceanic waters, but are found to have only very limited impact on subsequent closure attempts for these stations. Best-fit regressions on log-transformed measured and modelled downwards irradiance, Ed, and upwards radiance, Lu, profiles have median slopes between 0.92 - 1.24, revealing a tendency to underestimate Ed and Lu with depth. This is only partly explained by non-inclusion of fluorescence emission from CDOM and chlorophyll in the simulations. There are several stations where multiple volume scattering function related data processing steps perform poorly which suggests the potential existence of unresolved features in the modelling of the angular distribution of scattered photons. General optical closure therefore remains problematic, even though there are many cases in the data set where the match between measured and modelled radiometric data is within 25% RMS%E. These results are significant for applications that rely on optical closure e.g. assimilating ocean colour data into coupled physical-ecosystem models.

On the non-closure of particle backscattering coefficient in oligotrophic oceans

Many studies have consistently found that the particle backscattering coefficient (bbp) in oligotrophic oceans estimated from remote-sensing reflectance (Rrs) using semi-analytical algorithms is higher than that from in situ measurements. This overestimation can be as high as ~300% for some oligotrophic ocean regions. Various sources potentially responsible for this discrepancy are examined. Further, after applying an empirical algorithm to correct the impact from Raman scattering, it is found that bbp from analytical inversion of Rrs is in good agreement with that from in situ measurements, and that a closure is achieved.

Retrieving quantum yield of sun-induced chlorophyll fluorescence near surface from hyperspectral in-situ measurement in productive water

Magnitude and quantum yield (eta) of sun induced chlorophyll fluorescence are determined in widely varying productive waters with chlorophyll concentrations from 2- 200 mg/m(3). Fluorescence was estimated using linear fitting of in-situ measured surface reflectance with elastic and inelastic reflectance spectra. Elastic reflectance spectra were obtained from Hydrolight simulations with measured absorption and attenuation spectra as inputs. Eta is then computed based on a depth integrated fluorescence model and compared with Hydrolight calculation results. Despite the large variability of coastal environments examined the ? values are found to vary over a relatively narrow range 0.1%-1% with mean values of 0.33%+/-0.17%.

Bio-optical properties and ocean color algorithms for coastal waters influenced by the Mississippi River during a cold front

During the passage of a cold front in March 2002, bio-optical properties examined in coastal waters impacted by the Mississippi River indicated that westward advective flows and increasing river discharge containing high concentrations of nonalgal particles contributed significantly to surface optical variability. A comparison of seasonal data from three cruises indicated spectral models of absorption and scattering to be generally consistent with other coastal environments, while their parameterization in terms of chlorophyll (Chl) alpha concentration showed seasonal variability. The exponential slope of the colored dissolved organic matter (CDOM) averaged 0.0161+/-0.00054 nm(-1) and nonalgal absorption averaged 0.011 nm(-1) with deviations from general trends observed due to anomalous water properties. Although the phytoplankton specific absorption coefficients varied over a wide range [0.02 to 0.1 m2 (mg Chl)(-1) at 443 nm] being higher in offshore surface waters, values of phytoplankton absorption spectra at the Sea-Viewing Wide Field-of-View Sensor (SeaWiFS) wave bands were highly correlated to modeled values. Particulate scattering characteristics were similar to observations for other coastal waters, while backscattering ratios were on average lower in phytoplankton-dominated surface waters (0.011+/-0.003) and higher in low Chl near-bottom waters (0.0191+/-0.0045). Average percent differences in remote sensing reflectance Rrs derived from modeled and in-water radiometric measurements were highest in the blue wave bands (52%) and at locations with more stratified water columns. SeaWiFS estimates of Chl and CDOM absorption derived using regional empirical algorithms were highly correlated to in situ data.

In situ optical variability and relationships in the Santa Barbara Channel: implications for remote sensing

Relationships and variability of bio-optical properties in coastal waters are investigated. Optical proxies indicate that these coastal waters are optically complex and highly variable and are categorized as follows: (1) relatively clear and dominated by high index of refraction, biogenic particles, (2) more turbid, consisting of mostly inorganic particles and little phytoplankton, (3) extremely turbid with high concentrations of inorganic particles, and (4) more turbid and dominated by biogenic particles. We present a method, alternative to traditional remote-sensing algorithms, of classifying coastal waters [the Spectral Angle Mapper (SAM)] and utilize the SAM to successfully isolate plume conditions in time series of downwelling irradiance and total absorption coefficient. We conclude with a discussion of the use of the SAM for coastal management operations.

Uncertainties of inherent optical properties obtained from semianalytical inversions of ocean color

We present a method to quantify the uncertainties in the in-water constituent absorption and backscattering coefficients obtained from an inversion of remotely sensed reflectance (rrs). We first find a set of positive inversion solutions within a given uncertainty range around the values of the inverted rrs. The uncertainties of the solutions are then computed based on the statistics of these solutions. We demonstrate the uncertainty calculation algorithm using a specific semianalytic inversion model applied to both a field and a simulated data set. When the associated uncertainties are taken into account, the inverted parameters are generally within the uncertainties of the measured (or simulated) parameters, highlighting the success of the inversion and the method to obtain uncertainties. The specific inversion we use, however, fails to retrieve two spectral parameters within a usable range. The method presented is general and can be applied to all existing semianalytical inversion algorithms.

Evidence for wavelength dependence of the scattering phase function and its implication for modeling radiance transfer in shelf seas

More than 90% of stations from the Irish and Celtic Seas are found to have significantly higher back-scattering ratios in the blue (470 nm) than in the red (676 nm) wave band. Attempts to obtain optical closure by use of radiance transfer modeling were least successful for stations at which backscattering ratios are most strongly wavelength dependent. Significantly improved radiance transfer simulation results were obtained with a modified scattering correction algorithm for AC-9 absorption measurements that took wavelength dependency in the scattering phase function into account.

Measured and modeled radiometric quantities in coastal waters: toward a closure

Accurate radiative transfer modeling in the coupled atmosphere-sea system is increasing in importance for the development of advanced remote-sensing applications. Aiming to quantify the uncertainties in the modeling of coastal water radiometric quantities, we performed a closure experiment to intercompare theoretical and experimental data as a function of wavelength lambda and water depth z. Specifically, the study focused on above-water downward irradiance E(d)(lambda, 0+) and in-water spectral profiles of upward nadir radiance L(u)(lambda, z), upward irradiance E(u)(lambda, z), downward irradiance E(d)(lambda, z), the E(u)(lambda, z)/L(u)(lambda, z) ratio (the nadir Q factor), and the E(u)(lambda, z)/E(d)(lambda, z) ratio (the irradiance reflectance). The theoretical data were produced with the finite-element method radiative transfer code ingesting in situ atmospheric and marine inherent optical properties. The experimental data were taken from a comprehensive coastal shallow-water data set collected in the northern Adriatic Sea. Under various measurement conditions, differences between theoretical and experimental data for the above-water E(d)(lambda, 0+) and subsurface E(d)(lambda, 0-) as well as for the in-water profiles of the nadir Q factor were generally less than 15%. In contrast, the in-water profiles of L(u)(lambda, z), E(d)(lambda, z), E(u)(lambda, z) and of the irradiance reflectance exhibited larger differences [to approximately 60% for L(u)(lambda, z) and E(u)(lambda, z), 30% for E(d)(lambda, z), and 50% for the irradiance reflectance]. These differences showed a high sensitivity to experimental uncertainties in a few input quantities used for the simulations: the seawater absorption coefficient; the hydrosol phase function backscattering probability; and, mainly for clear water, the bottom reflectance.