Color-producing Agents Research Articles

Validation of 2015 Lake Erie MODIS image spectral decomposition using visible derivative spectroscopy and field campaign data

Timely identification of color-producing agents (CPAs) in Lake Erie is a challenging, but vital aspect of monitoring harmful algal blooms (HABs). In particular, HABs that include large amounts of cyanobacteria (CyanoHABs) can be toxic to humans, posing a threat to drinking water, in addition to recreational and economic use of Lake Erie. The optical signal of Lake Erie is complex (Becker et al., 2009; Moore et al., 2017), typically comprised of phytoplankton, cyanobacteria, colored dissolved organic matter (CDOM), detritus, and terrigenous inorganic particles, varying in composition both spatially and temporally. The Kent State University (KSU) spectral decomposition method effectively partitions CPAs using a varimax-rotated, principal component analysis (VPCA) of visible reflectance spectra measured using lab, field or satellite instruments (Ali et al., 2013; Ortiz et al., 2017, 2013). We analyze 2015 imagery acquired by the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor and field samples collected during the early 2015 cyanoHAB season. We identified four primary CPA spectral signatures, and the spatial distribution of each identified CPA, in the reflectance spectra datasets of both the MODIS and lab-measured water samples. The KSU spectral decomposition method results in mixtures of specific pigments, pigment degradation products, and minerals that describe the optically complex water. We found very good agreement between the KSU VPCA spectral decomposition results and in situ measurements, indicating that this method may be a powerful tool for rapid CyanoHAB monitoring and assessment in large lakes using instruments that provide moderate resolution imagery (0.3 to 1 km2).

Field Spectroscopy as a Tool for Enhancing Water Quality Monitoring in the ACE Basin, SC

The Ashepoo, Combahee, Edisto (ACE) Basin in South Carolina is one of the largest undeveloped estuaries in the Southeastern United States. This system is monitored and protected by several government agencies to ensure its health and preservation. However, as populations in surrounding cities rapidly expand and land is urbanized, the surrounding water systems may decline from an influx of contaminants, leading to hypoxia, fish kills, and eutrophication. Conventional in situ water quality monitoring methods are timely and costly. Satellite remote sensing methods are used globally to monitor water systems and can produce an instantaneous synopsis of color-producing agents (CPAs), including chlorophyll-a, suspended matter (TSM), and colored-dissolved organic matter by applying bio-optical models. In this study, field, laboratory, and historical land use land cover (LULC) data were collected during the summers of 2002, 2011, 2015, and 2016. The results indicated higher levels of chlorophyll, ranging from 2.94 to 12.19 μg/L, and TSM values were from 60.4 to 155.2 mg/L between field seasons, with values increasing with time. A model was developed using multivariate, partial least squares regression (PLSR) to identify wavelengths that are more sensitive to chlorophyll-a (R2 = 0.49; RMSE = 1.8 μg/L) and TSM (R2 = 0.40; RMSE = 12.9 mg/L). The imbrication of absorption and reflectance features characterizing sediments and algal species in ACE Basin waters make it difficult for remote sensors to distinguish variations among in situ concentrations. The results from this study provide a strong foundation for the future of water quality monitoring and for the protection of biodiversity in the ACE basin.

Application of empirical and semi-analytical algorithms to MERIS data for estimating chlorophyll a in Case 2 waters of Lake Erie

Lake Erie is biologically the most active lake among the Great Lakes of North America, experiencing seasonal harmful algal blooms (HABs). The early detection of HABs in the Western Basin of Lake Erie (WBLE) requires a more efficient and accurate monitoring tool. Remote sensing is an efficient tool with high spatial and temporal coverage that can allow accurate and timely detection of the HABs. The WBLE is heavily influenced by the surrounding terrestrial ecosystem via rivers such as the Sandusky River and the Maumee River. As a result, the optical properties of the WBLE are influenced by multiple color producing agents (CPAs) such as phytoplankton, colored dissolved organic matter (CDOM), organic detri- tus, and terrigenous inorganic particles. The diversity of the CPAs and their non-linear interactions makes these waters optically complex, and the task of optical remote sensing for retrieving estimates of CPAs more challenging. Chlo- rophyll a, which is the primary light harvesting pigment in all phytoplankton, is used as a proxy for algal biomass. In this study, several published remote sensing algorithms and band ratio models were applied to the reflectance data from the full resolution MERIS sensor to remotely estimate chlorophyll a concentrations in the WBLE. Efficiency of the sensor and the algorithms performance were tested through a least squares regression and residual analysis. The results indicate that, among the suite of existing bio- optical models, the Simis semi-analytical algorithm provided the best model results for measures of algal bio- mass in the optically complex WBLE with R 2 of 0.65, RMSE 0.85 lg/l, (n = 71, P \ 0.05). The superior results of this model in detecting chlorophyll a are attributed to several factors including optimizing spectral regions that are less sensitive to CDOM and the incorporation of cor- rection factors such as absorption effects due to pure water (aw), backscatter (bb) from suspended matter and interfer- ence due to phycocyanin (d), a major accessory pigment in the WBLE.

Spatial and temporal evolution of the St. Lawrence River spectral profile: A 25-year case study using Landsat 5 and 7 imagery

Underwater light characteristics (quality and quantity) are important drivers of many ecological processes in aquatic systems. However, the mechanisms driving the underwater light climate in large rivers are poorly understood due to the complex interactions among color-producing agents (CPAs), which can vary in space and time. A compelling example of such a complex interaction among CPAs concerns the St. Lawrence River, where chromophoric dissolved organic matter and suspended inorganic particulate matter interact to structure the underwater spectral environment. Because such interactions can be complex, the combined net effect of CPAs on water color is not intuitive and remains to be evaluated. To resolve the spatial and temporal dynamics of the spectral profile in the St. Lawrence River, we analyzed a series of 44 Landsat 5 and 7 images, distributed between 1984 and 2009, which cover the main freshwater section of the river. Rather than following linear trends along the longitudinal axis, the spectral profiles on spectral bands (blue, green and red) presented three distinctive spatial patterns. We argue that matter injected by tributaries and discontinuous zones of the river strongly impacts water color by modulating the balance among CPAs. We also clearly demonstrate a marked trend between 1984 and 2009 toward decreasing reflectance values on bands 1, 2 and 3, presumably in response to changes in land use in the surrounding watershed.

An algorithm to retrieve chlorophyll, dissolved organic carbon, and suspended minerals from Great Lakes satellite data

An algorithm that utilizes individual lake hydro-optical (HO) models has been developed for the Great Lakes that uses SeaWiFS, MODIS, or MERIS satellite data to estimate concentrations of chlorophyll, dissolved organic carbon, and suspended minerals. The Color Producing Agent Algorithm (CPA-A) uses a specific HO model for each lake. The HO models provide absorption functions for the Color Producing Agents (CPAs) (chlorophyll (chl), colored dissolved organic matter (as dissolved organic carbon, doc), and suspended minerals (sm)) as well as backscatter for the chlorophyll, and suspended mineral parameters. These models were generated using simultaneous optical data collected with in situ measurements of CPAs collected during research cruises in the Great Lakes using regression analysis as well as using specific absorption and backscatter coefficients at specific chl, doc, and sm concentrations. A single average HO model for the Great Lakes was found to generate insufficiently accurate concentrations for Lakes Michigan, Erie, Superior and Huron. These new individual lake retrievals were evaluated with respect to EPA in situ field observations, as well as compared to the widely used OC3 MODIS retrieval. The new algorithm retrievals provided slightly more accurate chl values for Lakes Michigan, Superior, Huron, and Ontario than those obtained using the OC3 approach as well as providing additional concentration information on doc and sm. The CPA-A chl retrieval for Lake Erie is quite robust, producing reliable chl values in the reported EPA concentration ranges. Atmospheric correction approaches were also evaluated in this study.

Simulation of water colors in a shallow acidified lake, Lake Onneto, Japan, using colorimetric analysis and bio-optical modeling

Lake Onneto has unusually brilliant water colors, such as blue-green, green, and greenish yellow. We investigated the coloration mechanism in this lake by measuring the radiometric water color and inherent optical properties (IOPs), such as attenuation, absorption, and scattering coefficients of color-producing agents (CPAs) in the lake. The hue of the water body was determined using colorimetric analysis. In addition, the radiometric water color of the lake was simulated using bio-optical modeling. Results showed that the hues of the water body were different between two sampling stations (blue-green and green at 3.1 and 5.5 m of bottom depth, respectively). However, the total suspended solids (TSS) concentrations were almost identical between the two stations (0.47 and 0.42 mg l−1, respectively). In addition, the attenuation and absorption coefficients showed that the CPAs in the water body were only inorganic suspended solids (ISS) more than 0.7 μm in diameter, and specific CPAs, such as dissolved mineral ions and aqueous colloids, were not observed in the lake. Simulation of the radiometric water color showed that the radiation around 600 nm of wavelength (yellow region) steeply increased with decreasing bottom depth, indicating that the water colors of Lake Onneto are strongly governed by the light reflection from yellowish bottom sediments.

Verification and Application of a Bio-optical Algorithm for Lake Michigan Using SeaWiFS: a 7-year Inter-annual Analysis

In this paper we utilize 7 years of SeaWiFS satellite data to obtain seasonal and interannual time histories of the major water color-producing agents (CPAs), phytoplankton chlorophyll (chl), dissolved organic carbon (doc), and suspended minerals (sm) for Lake Michigan. We first present validation of the Great Lakes specific algorithm followed by correlations of the CPAs with coincident environmental observations. Special attention is paid to the satellite observations of the extensive episodic event of sediment resuspension and calcium carbonate precipitation out of the water. We then compare the obtained time history of the CPA's spatial and temporal distributions throughout the lake to environmental observations such as air and water temperature, wind speed and direction, significant wave height, atmospheric precipitation, river runoff, and cloud and lake ice cover. Variability of the onset, duration, and spatial extent of both episodic events and seasonal phenomena are documented from the SeaWiFS time series data, and high correlations with relevant environmental driving factors are established. The relationships between the CPAs retrieved from satellite data and environmental observations are then used to speculate on the future of Lake Michigan under a set of climate change scenarios.

Color of Inland and Coastal Waters

Analysis of the color of water bodies is concerned with understanding what gives water its color (the “forward” problem), and the use of the remotely observed color to learn about “color producing agents” within the water (the “inverse” problem). Recently the study of water color has been going through a rapid expansion in sensors, coverage area, and algorithms. Similarly, the interest in the derived color‐producing agents (CPA) is growing from both scientific interests and agencies in charge of monitoring water bodies; for example, for harmful algal blooms. Monitoring the color of water bodies remotely has the advantage of unparalleled coverage in space and time. However, this approach is limited by both the inability to operate when conditions are cloudy and the fact that the signal is emanating from only the surface layer of the water and from a different depth at each wavelength.

Validation of a radiometric color model applicable to optically complex water bodies

A radiometric color model relating the color of optically complex (non-Case I) waters to the organic and inorganic color-producing agents (CPA) responsible for that color has been previously applied to the waters of Lake Ontario, Canada and Lake Ladoga, Russia. Additional underwater optical measurements and water quality data from Lakes Erie and Michigan and several boreal lakes in northern Ontario, as well as comparable data from the European lakes Krasnoye, Zug, and Lucerne are utilized to illustrate both the validity of the model, as well as the universality of its application. It was found that, due to the limited availability of specific spectral scattering and absorption properties (optical cross-section spectra) for CPA indigenous to natural water bodies on a global scale, the use of optical cross-section spectra appropriate to the CPA indigenous to Lake Ontario and those appropriate to the CPA indigenous to Lake Ladoga provided a more-than-adequate surrogate for the water bodies considered herein (inclusive of river systems in the British Columbia Canadian Cordillera for which results of a study relating river water color to hydrographic basin features are also revisited). Similarities among the optical cross-section spectra pertinent to freshwater biota and similarities among the optical cross-section spectra pertinent to freshwater dissolved organic matter would appear to allow such a liberal use of site-specific aquatic optical properties. However, greatest discrepancies in indigenous inland water CPA optical cross-section spectra are consequences of global geologic diversities. A compensation for geologic diversities is illustrated by the use of Lake Ladoga cross-section spectra with the European lakes and the Lake Ontario cross-section spectra with the North American lakes, also providing validation of both the radiometric model and its universality. Such geologic similarities could alleviate labor-, time-, and cost-intensive determinations of optical cross-section spectra for many inland water bodies.

Modelling the Radiometric Color of Inland Waters: Implications to a) Remote Sensing and b) Limnological Color Scales

Utilizing a bio-optical model previously developed for Lake Ontario, the responsiveness of chromaticity coordinates (X, Y, Z), dominant wavelength (λ dom), and associated spectral purity (p) to the abundance of color-producing agents (CPA) residing within the Lake Ladoga water column was determined. CPA considered were phytoplankton (chl), suspended minerals (sm), and dissolved organic carbon (doc). Waters that contain simultaneously low concentrations of chl, sm, and doc are shown to appear blue to turquoise in color (472–500 nm). Highly turbid waters (i.e., waters containing high concentrations of chl and/or sm) with low concentrations of doc are shown to display colors ranging from green to brown (> 500 nm). Waters with large concentrations of doc, irrespective of turbidity, are shown to be invariably brownish (560–570 nm). With increasing CPA content, X, Y, and Z (and, consequently, λ dom) asymptotically approach constant limit values. An “end-point” dominant wavelength at about 572 nm appears to be intrinsically characteristic of all natural waters. It is shown that when one or more CPA exceeds a critical concentration, the spectral purity p asymptotically approaches values in the range 0.35 to 0.45 for all waters (exceptive of those containing solely chl in the restricted concentration range < 0.5 μg/L). Optical distinctiveness, particularly with respect to indigenous doc, of natural waters, while impacting the spectral purity of the “end-point” radiometric color, does not produce a comparable impact on the “end-point” color itself. This work reinforces the restrictive application of chromaticity analyses to the remote sensing of binary aquatic systems comprised of water plus one CPA. It also illustrates that neither panchromatic nor two-channel ratio images can provide unambiguous inference of water quality parameters. Correspondence between radiometric water color descriptors (X, Y, Z, λ dom, and p) and water color scales traditionally used in limnology is established, illustrating that the platinum-cobalt scale would be most appropriate for assessing waters that were radiometrically yellow, provided that the yellow hue were not invariably attributed to doc.

COLOR FORMATION IN RAW LEAD GLAZES1

Various amounts of practically all of the chromatic elements in the periodic system were introduced individually into a simple lead silicate glaze, and the resulting glazes were analyzed for color with an automatically recording photoelectric color analyzer. The spectrophotometric data were then converted into trichromatic and monochromatic specifications of color. An empirical relationship was then determined between the concentration of the chromatic element and the three attributes of color: hue, saturation, and brilliance. The effect of glaze thickness on color was also studied. It was found that the same laws connecting concentration of the coloring element with the three attributes of color held for all the elements used. A number of elements not previously mentioned in the literature as color-producing agents were found to give interesting colored glazes. In several cases an element was added to the base glaze in various forms, such as nitrates, chlorides, and oxides. If equivalent amounts were added, the color in all cases was the same regardless of the colorless radical.