Robert G. Steward

Hyperspectral remote sensing for shallow waters: 2. Deriving bottom depths and water properties by optimization

In earlier studies of passive remote sensing of shallow-water bathymetry, bottom depths were usually derived by empirical regression. This approach provides rapid data processing, but it requires knowledge of a few true depths for the regression parameters to be determined, and it cannot reveal in-water constituents. In this study a newly developed hyperspectral, remote-sensing reflectance model for shallow water is applied to data from computer simulations and field measurements. In the process, a remote-sensing reflectance spectrum is modeled by a set of values of absorption, backscattering, bottom albedo, and bottom depth; then it is compared with the spectrum from measurements. The difference between the two spectral curves is minimized by adjusting the model values in a predictor-corrector scheme. No information in addition to the measured reflectance is required. When the difference reaches a minimum, or the set of variables is optimized, absorption coefficients and bottom depths along with other properties are derived simultaneously. For computer-simulated data at a wind speed of 5 m/s the retrieval error was 5.3% for depths ranging from 2.0 to 20.0 m and 7.0% for total absorption coefficients at 440 nm ranging from 0.04 to 0.24 m(-1). At a wind speed of 10 m/s the errors were 5.1% for depth and 6.3% for total absorption at 440 nm. For field data with depths ranging from 0.8 to 25.0 m the difference was 10.9% (R2 = 0.96, N = 37) between inversion-derived and field-measured depth values and just 8.1% (N = 33) for depths greater than 2.0 m. These results suggest that the model and the method used in this study, which do not require in situ calibration measurements, perform very well in retrieving in-water optical properties and bottom depths from above-surface hyperspectral measurements.

Hyperspectral remote sensing for shallow waters. I. A semianalytical model.

For analytical or semianalytical retrieval of shallow-water bathymetry and/or optical properties of the water column from remote sensing, the contribution to the remotely sensed signal from the water column has to be separated from that of the bottom. The mathematical separation involves three diffuse attenuation coefficients: one for the downwelling irradiance (K(d)), one for the upwelling radiance of the water column (K(u)(C)), and one for the upwelling radiance from bottom reflection (K(u)(B)). Because of the differences in photon origination and path lengths, these three coefficients in general are not equal, although their equality has been assumed in many previous studies. By use of the Hydrolight radiative-transfer numerical model with a particle phase function typical of coastal waters, the remote-sensing reflectance above (R(rs)) and below (r(rs)) the surface is calculated for various combinations of optical properties, bottom albedos, bottom depths, and solar zenith angles. A semianalytical (SA) model for r(rs) of shallow waters is then developed, in which the diffuse attenuation coefficients are explicitly expressed as functions of in-water absorption (a) and backscattering (b(b)). For remote-sensing inversion, parameters connecting R(rs) and r(rs) are also derived. It is found that r(rs) values determined by the SA model agree well with the exact values computed by Hydrolight (~3% error), even for Hydrolight r(rs) values calculated with different particle phase functions. The Hydrolight calculations included b(b)/a values as high as 1.5 to simulate high-turbidity situations that are occasionally found in coastal regions.

Reflectance model for quantifying chlorophyll a in the presence of productivity degradation products

Marine colored dissolved organic matter (CDOM) (also gilvin or yellow substance) absorbs light at an exponentially decreasing rate as a function of wavelength. From 410 nm to about 640 nm, particulate phytoplankton degradation products including pheopigments, detritus, and bacteria have absorption curves that are similar in shape to that of CDOM. In coastal areas and areas downstream from upwelling regions, these constituents of seawater often absorb much more light than do living phytoplankton, leading to errors in satellite-derived chlorophyll estimates as high as 133%. Proposed NASA sensors for the 1990s will have spectral channels as low as 412 nm, permitting the development of algorithms that can separate the absorption effects of CDOM and other phytoplankton degradation products from those due to biologically viable pigments. A reflectance model has been developed to estimate chlorophyll a concentrations in the presence of CDOM, pheopigments, detritus, and bacteria. Nomograms and lookup tables have been generated to describe the effects of different mixtures of chlorophyll a and these degradation products on the R(412) : R(443) and R(443) : R(565) remote-sensing reflectance or irradiance reflectance ratios. These are used to simulate the accuracy of potential ocean color satellite algorithms, assuming that atmospheric effects have been removed. For the California Current upwelling and offshore regions, with chlorophyll a ≤ 1.3 mg m−3, the average error for chlorophyll a retrievals derived from irradiance reflectance data for degradation product-rich areas was reduced from ±61% to ±23% by application of an algorithm using two reflectance ratios (R(412) : R(443) and R(443) : R(565)) rather than the commonly used algorithm applying a single reflectance ratio (R(443) : R(565)).

Model for the interpretation of hyperspectral remote-sensing reflectance

Remote-sensing reflectance is easier to interpret for the open ocean than for coastal regions because the optical signals are highly coupled to the phytoplankton (e.g., chlorophyll) concentrations. For estuarine or coastal waters, variable terrigenous colored dissolved organic matter (CDOM), suspended sediments, and bottom reflectance, all factors that do not covary with the pigment concentration, confound data interpretation. In this research, remote-sensing reflectance models are suggested for coastal waters, to which contributions that are due to bottom reflectance, CDOM fluorescence, and water Raman scattering are included. Through the use of two parameters to model the combination of the backscattering coefficient and the Q factor, excellent agreement was achieved between the measured and modeled remote-sensing reflectance for waters from the West Florida Shelf to the Mississippi River plume. These waters cover a range of chlorophyll of 0.2-40 mg/m(3) and gelbstoff absorption at 440 nm from 0.02-0.4 m(-1). Data with a spectral resolution of 10 nm or better, which is consistent with that provided by the airborne visible and infrared imaging spectrometer (AVIRIS) and spacecraft spectrometers, were used in the model evaluation.

Estimating primary production at depth from remote sensing

By use of a common primary-production model and identical photosynthetic parameters, four different methods were used to calculate quanta (Q) and primary production (P) at depth for a study of high-latitude North Atlantic waters. The differences among the four methods relate to the use of pigment information in the upper water column. Methods 1 and 2 use pigment biomass (B) as an input and a subtropical, empirical relation between K(d) (diffuse attenuation coefficient) and B to estimate Q at depth. Method 1 uses measured B, but Method 2 uses B derived from the Coastal Zone Color Scanner (subtropical algorithm) as inputs. Methods 3 and 4 use the phytoplankton absorption coefficient (a(ph)) instead of B as input, and Method B uses empirically derived a(ph)(440) and K(d) values, and Method 4 uses analytically derived a(ph)(440) and a (total absorption coefficient) values based on the same remote measurements as Method 2. When the calculated and the measured values of Q(z) and P(z) were compared, Method 4 provided the closest results [for P(z), r(2) = 0.95 (n = 24), and for Q(z), r(2) = 0.92 (n = 11)]. Method 1 yielded the worst results [for P(z), r(2) = 0.56 and for Q(z), r(2) = 0.81]. These results indicate that one of the greatest uncertainties in the remote estimation of P can come from a potential mismatch of the pigment-specific absorption coefficient (a(ph)*), which is needed implicitly in current models or algorithms based on B. We point out that this potential mismatch can be avoided if we arrange the models or algorithms so that they are based on the pigment absorption coefficient (a(ph)). Thus, except for the accuracy of the photosynthetic parameters and the above-surface light intensity, the accuracy of the remote estimation of P depends on how accurately a(ph) can be estimated, but not how accurately B can be estimated. Also, methods to derive a(ph) empirically and analytically from remotely sensed data are introduced. Curiously, combined application of subtropical algorithms for both B and K(d) to subarctic waters apparently compensates to some extent for effects that are due to their similar and implicit pigment-specific absorption coefficients for the calculation of Q(z).

Remote sensing reflectance and inherent optical properties of oceanic waters derived from above-water measurements

Remote-sensing reflectance and inherent optical properties of oceanic properties of oceanic waters are important parameters for ocean optics. Due to surface reflectance, Rrs or water-leaving radiance is difficult to measure from above the surface. It usually is derived by correcting for the reflected skylight in the measured above-water upwelling radiance using a theoretical Fresnel reflectance value. As it is difficult to determine the reflected skylight, there are errors in the Q and E derived Rrs, and the errors may get bigger for high chl_a coastal waters. For better correction of the reflected skylight,w e propose the following derivation procedure: partition the skylight into Rayleigh and aerosol contributions, remove the Rayleigh contribution using the Fresnel reflectance, and correct the aerosol contribution using an optimization algorithm. During the process, Rrs and in-water inherent optical properties are derived at the same time. For measurements of 45 sites made in the Gulf of Mexico and Arabian Sea with chl_a concentrations ranging from 0.07 to 49 mg/m3, the derived Rrs and inherent optical property values were compared with those from in-water measurements. These results indicate that for the waters studied, the proposed algorithm performs quite well in deriving Rrs and in- water inherent optical properties from above-surface measurements for clear and turbid waters.

The Hydraulic Equivalence of Mica

ABSTRACT Settling experiments performed on silt- to fine-sand-sized mica flakes with a holographic micro-velocimeter revealed that mica is the hydraulic equivalent of quartz spheres having diameters a factor of 4 to 12 times smaller. Mica in the very fine to fine-sand sizes has been traditionally used by sedimentologists to delineate areas of deposition or nondeposition and potential winnowing of fines, and is here found to be the hydraulic equivalent of silt-sized particles but not of clay. Experiments also showed that mica flakes tend to settle at orientations which are neither perpendicular nor parallel to the gravitational vector and tend to generally maintain their orientation throughout. Equations for the settling of a disc in Lerman and others (1974) and that developed by Komar and Reimers (1978) are shown to be mathematically similar for the coarse-silt to fine-sand ranges of discs and are adequate predictors of settling rates of mica flakes. A comparison of the hydraulic equivalency of quartz spheres to coarse-silt- through fine-sand-sized mica flakes is presented.

Calculated quantum yield of photosynthesis of phytoplankton in the Marine Light‐Mixed Layers (59°N, 21°W)

The quantum yield o of photosynthesis (mol C (mol photons)−1) was calculated at six depths for the waters of the Marine Light-Mixed Layer (MLML) cruise of May 1991. As there were photosynthetically available radiation (PAR) but no spectral irradiance measurements for the primary production incubations, three ways are presented here for the calculation of the absorbed photons (AP) by phytoplankton for the purpose of calculating o. The first is based on a simple, nonspectral model; the second is based on a nonlinear regression using measured PAR values with depth; and the third is derived through remote sensing measurements. We show that the results of o calculated using the nonlinear regression method and those using remote sensing are in good agreement with each other, and are consistent with the reported values of other studies. In deep waters, however, the simple nonspectral model may cause quantum yield values much higher than theoretically possible.

Effects of fluorescence and water Raman scattering on models of remote-sensing reflectance

The modeling of oceanic remote sensing reflectance typically employs absorption and scattering parameters for the various constituents present in marine waters. Trans-spectral light sources such as fluorescence and Raman scattering are not generally parameterized in these models. Bioluminescence is not considered to be a significant contributor to water-leaving radiance measurements obtained mid-day, and has not been included in the models either. In this paper we present evidence of effects due to these three phenomena by comparing model results to remote sensing reflectances measured at several stations during the 1988 California Coastal Transition Zone (CTZ) Experiment. Differences between modeled and measured Rrs(A) values are discussed from the perspective of in-situ light source contributions.

Image analysis techniques for holograms of dynamic oceanic particles

A holographic image analysis system has been developed to measure position, velocity, size, and shape of microscopic particles settling in 3-D space. Images of particles recorded sequentially on individual holographic frames are reconstructed using an in-line far-field configuration, registered in 3-D space, and particle displacements (velocities) between sequential frames are determined. Particle settling velocity is calculated using elapsed time between frames. Digitized video signals of the reconstructed holographic images are processed to determine particle size, shape, and area to facilitate identification by shape from frame to frame and to calculate particle specific gravities. A cataloging system was developed to provide efficient data management.