Collin S. Roesler

In situ phytoplankton absorption, fluorescence emission, and particulate backscattering spectra determined from reflectance

An inverse model was developed to extract the absorption and scattering (elastic and inelastic) properties of oceanic constituents from surface spectral reflectance measurements. In particular, phytoplankton spectral absorption coefficients, solar-stimulated chlorophyll a fluorescence spectra, and particle backscattering spectra were modeled. The model was tested on 35 reflectance spectra obtained from irradiance measurements in optically diverse ocean waters (0.07 to 25.35 mg m−3 range in surface chlorophyll a concentrations). The universality of the model was demonstrated by the accurate estimation of the spectral phytoplankton absorption coefficients over a range of 3 orders of magnitude (ρ = 0.94 at 500 nm). Under most oceanic conditions (chlorophyll a<3 mg m−3) the percent difference between measured and modeled phytoplankton absorption coefficients was <35%. Spectral variations in measured phytoplankton absorption spectra were well predicted by the inverse model. Modeled volume fluorescence was weakly correlated with measured chl a; fluorescence quantum yield varied from 0.008 to 0.09 as a function of environment and incident irradiance. Modeled particle backscattering coefficients were linearly related to total particle cross section over a twentyfold range in backscattering coefficients (ρ = 0.996, n = 12).

Variability in Arctic sea ice optical properties

The optical properties of sea ice exhibit considerable spatial, temporal, and spectral variability. During a field experiment at Barrow, Alaska, we examined the horizontal variability of spectral albedo and transmittance as well as the vertical variability of in-ice radiance. Temporal changes were monitored under cold conditions in April and during the onset of melt in June. Physical properties, including ice structure and concentrations of particulate and dissolved material, were measured to provide a context for understanding the observed temporal, horizontal, vertical, and spectral variability in optical properties. For snow-covered first-year ice in April, wavelength-integrated (300–3000 nm) albedos were high (0.8) and spatially uniform, but there was considerable variability in transmittance. Transmittance at 440 nm ranged by more than a factor of 2 over horizontal distances of only 25 m, owing primarily to differences in snow depth, although spectral variations in transmittance indicate that absorbing organic materials in the ice column contribute significantly to the horizontal variability. Peak values of transmittance in April were 1% near 500 nm, decreasing at both longer and shorter wavelengths. At the onset of melt in June, the ice surface rapidly evolved into a variegated mixture of melting snow, bare ice, and melt ponds. Albedos were much lower and exhibited considerable spatial variability, ranging from 0.2 to 0.5 over distances of a few meters concomitant with the variation in surface characteristics. Transmission increased over the spring transition as surface characteristics evolved to decrease albedo and as inice structure was altered by heating to reduce attenuation within the ice. The exception to this trend occurred over a period of a few days when an algal bloom developed on the underside of the ice and transmission was significantly reduced. Variability in the in-ice spectral radiance values was observed between nearby sites in both first-year and multiyear ice. While the radiance measurements are strongly dependent on the incident solar radiance, under similar solar conditions there was an observed shift in the peak of the maximum in the spectral radiance from 460 nm in clean ice to between 500 and 550 nm in ice that contained particulates in the surface layer. More impressive spectral shifts were found in an old melt pond that had accumulated particles at its base. Not only was there a strong shift in the spectral nature of the radiance as a function of horizontal distance, but there also existed large changes vertically within the ice. The vertical variability in the radiance attenuation coefficient was spatially coherent with variations in both the physical structure of the ice, especially grain size, and the concentrations of particulate and dissolved materials entrapped in the ice. Not surprisingly, the short-lived algal layer on the underside of the ice resulted in changes in the radiance attenuation coefficient from approximately 1 m−1 in the interior ice to approximately 40 m−1 within that layer.

Optical Monitoring and Forecasting Systems for Harmful Algal Blooms: Possibility or Pipe Dream?

Monitoring programs for harmful algal blooms (HABs) are currently reactive and provide little or no means for advance warning. Given this, the development of algal forecasting systems would be of great use because they could guide traditional monitoring programs and provide a proactive means for responding to HABs. Forecasting systems will require near real‐time observational capabilities and hydrodynamic/biological models designed to run in the forecast mode. These observational networks must detect and forecast over ecologically relevant spatial/ temporal scales. One solution is to incorporate a multiplatform optical approach utilizing remote sensing and in situ moored technologies. Recent advances in instrumentation and data‐assimilative modeling may provide the components necessary for building an algal forecasting system. This review will outline the utility and hurdles of optical approaches in HAB detection and monitoring. In all the approaches, the desired HAB information must be isolated and extracted from the measured bulk optical signals. Examples of strengths and weaknesses of the current approaches to deconvolve the bulk optical properties are illustrated. After the phytoplankton signal has been isolated, species‐recognition algorithms will be required, and we demonstrate one approach developed for Gymnodinium breve Davis. Pattern‐recognition algorithms will be species‐specific, reflecting the acclimation state of the HAB species of interest.Field data will provide inputs to optically based ecosystem models, which are fused to the observational networks through data‐assimilation methods. Potential model structure and data‐assimilation methods are reviewed.

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.

Optics, particles, stratification, and storms on the New England continental shelf

In situ beam attenuation and chlorophyll fluorescence were correlated with concentration and bulk composition of particles in shelf waters during summer and spring under different physical forcing conditions to determine if optical parameters could be used as an additional tracer in examining the process of mixing in shelf waters. Time series measurements were made for two 18 day periods during high stratification (late summer 1996, Δσt = ∼3.0 kg m−3 surface to bottom) and low but rapidly developing stratification (spring 1997, Δσt = 0.05 to 0.5 kg m−3) in 70 m of water in a midshelf environment south of Marthas Vineyard, Massachusetts. When defined by hydrography and optical profiles, four layers were identified during the summer: the surface mixed layer, the particle/chlorophyll maximum, the midwater particle minimum, and the bottom nepheloid layer. Fast moving solitons perturbed the water column briefly, but no storms perturbed the system until large surface swells from Hurricane Edouard intensified and thickened the nepheloid layer. Bulk composition and optics of particles in and above the nepheloid layer were distinctly different after the passage of Hurricane Edouard. The hurricane passage demonstrated that intense atmospheric forcing greatly influences both hydrographic and optical properties in the entire water column, even when highly stratified (Δσt = ∼3.0 kg m−3, decreasing to 0.8 kg m−3 post hurricane), and causes massive resuspension, due initially to wave shear stress that was later dominated by current shear. Restratification progressed rapidly after the hurricane passed. During spring the water column started as a weakly stratified two-layer system hydrographically and optically but evolved into three layers as stratification developed. Strong spring storms affected both surface and bottom layers but with decreasing impact as the water column stratified.

Temporal and vertical variability in optical properties of New England shelf waters during late summer and spring

Relationships between optical and physical properties were examined on the basis of intensive sampling at a site on the New England continental shelf during late summer 1996 and spring 1997. During both seasons, particles were found to be the primary source of temporal and vertical variability in optical properties since light absorption by dissolved material, though significant in magnitude, was relatively constant. Within the particle pool, changes in phytoplankton were responsible for much of the observed optical variability. Physical processes associated with characteristic seasonal patterns in stratification and mixing contributed to optical variability mostly through effects on phytoplankton. An exception to this generalization occurred during summer as the passage of a hurricane led to a breakdown in stratification and substantial resuspension of nonphytoplankton particulate material. Prior to the hurricane, conditions in summer were highly stratified with subsurface maxima in absorption and scattering coefficients. In spring, stratification was much weaker but increased over the sampling period, and a modest phytoplankton bloom caused surface layer maxima in absorption and scattering coefficients. These seasonal differences in the vertical distribution of inherent optical properties were evident in surface reflectance spectra, which were elevated and shifted toward blue wavelengths in the summer. Some seasonal differences in optical properties, including reflectance spectra, suggest that a significant shift toward a smaller particle size distribution occurred in summer. Shorter timescale optical variability was consistent with a variety of influences including episodic events such as the hurricane, physical processes associated with shelfbreak frontal dynamics, biological processes such as phytoplankton growth, and horizontal patchiness combined with water mass advection.

Evolution of electromagnetic signatures of sea ice from initial formation to the establishment of thick first-year ice

The spatial and temporal distribution of new and young sea ice types are of particular interest because of the influence this can exert on the heat and mass balance of the polar sea ice. The objective of the present work is to characterize the temporal evolution of the electromagnetic (EM) signatures of sea ice from initial formation through the development of first-year (FY) ice on the basis of the temporal variations in the physical properties of the ice. The time series of young sea ice signatures, including microwave emissivity, radar backscatter, and visible and infrared spectral albedo, has been measured at successive stages in the growth and development of sea ice, both under laboratory and field conditions. These observations have been accompanied by studies of the physical properties that influence the interaction between radiation and the ice. This has resulted in a consistent data set of concurrent multispectral observations that covers essentially all phases of the development of the different types of sea ice from initial formation to thick FY ice. Mutually consistent theoretical models covering the entire wavelength range of the observations are applied to selected cases and successfully match the observations. Principal component analysis (PCA) of the data set suggests combinations of the set of frequencies to effectively distinguish among different stages in the temporal evolution of the sea ice.

Underway and Moored Methods for Improving Accuracy in Measurement of Spectral Particulate Absorption and Attenuation

Optical sensors have distinct advantages when used in ocean observatories, autonomous platforms, and on vessels of opportunity, because of their high-frequency measurements, low power consumption, and the numerous established relationships between optical measurements and biogeochemical variables. However, the issues of biofouling and instrument stability over time remain complicating factors when optical instruments are used over periods longer than several days. Here, a method for obtaining calibration-independent measurements of spectral particle absorption and attenuation is presented. Flow-through optical instrumentation is routinely diverted through a large–surface area 0.2-mm cartridge filter, allowing for the calculation of particle optical properties by differencing temporally adjacent filtered and whole water samples. This approach yields measurements that are independent of drift in instrument calibration. The method has advantages not only for coastally moored deployments, but also for applications in optically clear waters where uncertainties in instrument calibration can be a significant part of the signal measured. The differencing technique is demonstrated using WET Labs (Philomath, Oregon) ac-9 and ac-s multi- and hyperspectral absorption and attenuation meters. For the ac-s sensor, a correction scheme is discussed that utilizes the spectral shape of water absorption in the near-infrared to improve the accuracy of temperature and scattering-corrected spectra. Flow-through particulate absorption measurements are compared with discrete filter-pad measurements and are found to agree well (R 2 5 0.77; rmse 5 0.0174 m 21 ).

Field observations of the electromagnetic properties of first-year sea ice

An interdisciplinary field experiment was conducted during April and May of 1994 at Point Barrow, AK, to investigate the relationship between the electromagnetic and physical-biological properties of first-year sea ice. Electromagnetic signatures of bare and snow-covered first-year ice were measured over a broad spectral range, including ultraviolet through near-infrared albedo, microwave emissivity, and radar backscatter. Observations indicated that the scattering of visible light varied significantly with depth in response to changes in the size and orientation of the ice crystals and in the number of brine and air inclusions. The scattering of visible light was greatest in the surface layer where there were numerous inclusions, and crystals tended to be small and randomly oriented. Changes in albedo over small horizontal distances were found to be related to surface layer conditions, including the number of air bubbles and particulate levels. Even for bare ice, transmittances were small with peaks in the blue-green. Scattering exceeds absorption throughout the snow and ice except in the skeletal layer colonized by bottom ice algae. Passive microwave emissivities showed a substantial difference between snow-covered and snow-free sites due to the effects of impedance matching at longer frequencies and volume scattering at higher frequencies produced by the snow, Spatial variability in the emissivity was quite small except at 90 GHz, where the emissivity was most sensitive to the amount of depth hoar. Radar backscatter coefficients were 5-6 dB larger for oblique viewing angles over snow-covered ice.

Ocean Inherent Optical Property Determination from In-Water Light Field Measurements

An algorithm is described and evaluated for determining the absorption and backscattering coefficients a(z) and bb(z) from measurements of the nadir-viewing radiance Lu(z) and downward irradiance Ed(z). The method, derived from radiative transfer theory, is similar to a previously proposed one for Eu(z) and Ed(z)and both methods are demonstrated with numerical simulations and field data. Numerical simulations and a sensitivity analysis show that good estimates of a(z) and bb(z) can be obtained if the assumed scattering phase function is approximately correct. In an experiment in Long Island Sound, estimates of a(z) derived with these methods agreed well with those obtained from an in situ reflecting tube instrument.