Introduction and background

The penetration of UVR into natural waters leads to exposure of organisms and nonliving matter to energetic photons. The response of organisms and nonliving matter to UVR exposure may change the UV transparency of the aquatic environment. When the intensity (irradiance) of UVR underwater is sufficient to cause biotic damage then a stressful range of depths exists in the water column. Ecological consequences of this stress will depend on the depth of penetration and spectral shifts in the underwater irradiance relative to the depth of the mixed layer ([1], Chapter 4). Spectral shifts may influence vision and behavior of aquatic organisms, especially those with UVR receptors (Chapter 14). They may also influence survival and productivity of aquatic organisms because of the interplay between damaging UV-B wavelengths and the sometimes-beneficial UV-A wavelengths ([2], Chapters 11-13). Ecologists can benefit from improved understanding of patterns and determinants of the intensity and spectral quality of underwater UVR, especially in view of future changes in global climate and in stratospheric ozone. While column ozone and clouds have a strong influence on damaging UVR that reaches the Earth's surface, other factors, especially the concentration and optical qualities of DOM, are more important to the penetration of UVR in the water column.

The topic of underwater optics has been extensively treated with a marine emphasis by Preisendorfer [3], Jerlov [4], and Mobley [5], and with added coverage of freshwater systems and photosynthesis by Kirk [6]. Reviews of solar radiation penetration into natural waters began with Smith and Tyler [7], and with emphasis on UVR and varying emphasis on marine and freshwater environments include Smith and Baker [8], Baker and Smith [9], Kirk [10], Booth and Morrow [11], and Whitehead et al. [12]. Xenopoulos and Schindler [13] recently reviewed underwater UVR in the context of terrestrial ecosystems and climate change. While marine and freshwater systems are often treated separately in the optical literature, here their optical properties will be compared to reveal common features as well as information gaps and different approaches used by oceanographers and limnologists.

Solar radiation is typically measured underwater as irradiance, the energy striking a unit of surface area (e.g., W m~2), and is further characterized by its wavelength (units, nanometers, nm). Spectral irradiance is reported as energy integrated over a waveband, which may be narrow (e.g., 1 nm) or broad (e.g., UV-A, 320-400 nm; UV-B, 280-320 nm). The solar UVR spectrum, about 10% of the incoming energy reaching the Earth's surface, includes UV-A and UV-B wavebands. Photosynthetic organisms use wavelengths starting at about 400 nm and extending to 700 nm (Photosynthetically Active Radiation or PAR) in the process of photosynthesis. Roughly half of the incoming solar energy is represented by infrared wavelengths from 700 to 2000 nm. Ozone in the atmosphere and DOM in natural waters strongly absorb UV-B wavelengths. Water molecules in the atmosphere and in aquatic systems strongly absorb far-red and infrared wavelengths.

UVR will pass through the air-water interface if it is not reflected. Reflection depends on the angle of incidence and follows Fresnel's Law [6]. Penetration of irradiance coming directly from the sun depends on the solar zenith angle (SZA) for a flat (calm) surface and decreases as SZA rises (i.e. for decreasing sun elevation). Windy conditions can increase the penetration of direct solar irradiance for high SZA but may have the opposite effect at low SZA. Penetration of the diffuse component of sunlight has been modeled [14,15]. The proportions of direct and diffuse solar radiation vary with wavelength, atmospheric conditions, and sun angle (some examples are shown in Figure 4A below). For UV-B wavelengths the distribution is often similar to that for all wavelengths under an overcast sky or when the sun is near the horizon, when models predict 96% transmission, [14]. The effects of surface waves and variations in direct versus diffuse radiation combine to complicate the interpretation of optical field measurements at shallow depths unless temporal and spatial averaging are adopted at appropriate scales [16].

Underwater light is similar in many coastal and freshwater environments. Figure 1 shows an example of underwater UVR and PAR spectra at several depths in a typical coastal ocean site (San Diego, California; unpublished data provided by J.H. Morrow). Water absorbs strongly in the red and longer wavelengths (> 600 nm) with the result that the right side of the curves shows rapid attenuation with depth. Photosynthetic pigments in phytoplankton (such as chlorophyll a) absorb blue (450 nm) wavelengths most strongly but phyto-

400 500 600

Wavelength (nm)

Figure 1. Typical coastal ocean underwater spectra of downwelling cosine irradiance with moderate levels of CDOM and algae. San Diego coastal waters, 5 miles offshore (5 January 2000), using Biospherical Instruments PRR-800 multichannel reflectance profiler (J.H. Morrow, unpublished data). This spectrum is plotted on a log scale to show similar percentage changes over the range of intensity that spans seven decades of magnitude. Except for the peak around 685 nm (caused by algal fluorescence), deep irradiance beyond 520 nm is likely to be caused by Raman scattering from shorter wavelengths.

0.00001

400 500 600

Wavelength (nm)

Figure 1. Typical coastal ocean underwater spectra of downwelling cosine irradiance with moderate levels of CDOM and algae. San Diego coastal waters, 5 miles offshore (5 January 2000), using Biospherical Instruments PRR-800 multichannel reflectance profiler (J.H. Morrow, unpublished data). This spectrum is plotted on a log scale to show similar percentage changes over the range of intensity that spans seven decades of magnitude. Except for the peak around 685 nm (caused by algal fluorescence), deep irradiance beyond 520 nm is likely to be caused by Raman scattering from shorter wavelengths.

plankton also absorb at shorter wavelengths. For the component of DOM called CDOM, absorption increases exponentially from the mid-visible wavelengths into the UVR range. The red peak (centered at 685 nm) that is evident at greater depths (most noticeable in the 20 m curve of Figure 1) is fluorescence emitted by photosynthetic cells - a small fraction of the visible light they absorb. The characteristic spectrum of underwater light (note the blue-green peak at 500 nm in the 160 m curve of Figure 1) is caused by combining the strong red absorption of water, the blue absorption by photosynthetic cells, and the violet and UVR absorption by DOM. The color of light emerging from deep natural waters is a product of this selective absorption of the medium and backscattering in the upward direction that increases at shorter wavelengths.

UV transparency of natural waters can be described empirically by two measures that are wavelength-specific and inter-related: the downwelling diffuse attenuation coefficient, Kd, and the percent attenuation depth, Zn%. A downwelling diffuse attenuation coefficient is nominally proportional to the concentration of substances in the water that absorb or scatter UVR [17,42]. It is typically calculated for specific wavelengths (A) from measurements of downwelling irradiance (Edtx) by fitting the following equation (in units of m_1) [8] to irradiance versus depth data:

where Z is geometric depth measured in vertical metres from the mean surface, £d 0- represents downwelling irradiance just below the water surface, and Ed{Zj.) is the downwelling irradiance at depth Z (m) and wavelength X (nm). Figure 2

Wavelength (nm)

Figure 2. Spectral diffuse attenuation of downwelling irradiance from Figure 1 compared with Kw for pure seawater estimated by Smith and Baker [18]. The phytoplankton concentration (based on chlorophyll a fluorescence) was highest in the upper 30 m. The curve labeled "Kd 1-30 m minus phytoplankton" was calculated by regression of spectral Kd against chlorophyll fluorescence for a range of depths, a method that also removes effects of scattering and absorption (including that of CDOM) that covary with phytoplankton fluorescence.

Wavelength (nm)

Figure 2. Spectral diffuse attenuation of downwelling irradiance from Figure 1 compared with Kw for pure seawater estimated by Smith and Baker [18]. The phytoplankton concentration (based on chlorophyll a fluorescence) was highest in the upper 30 m. The curve labeled "Kd 1-30 m minus phytoplankton" was calculated by regression of spectral Kd against chlorophyll fluorescence for a range of depths, a method that also removes effects of scattering and absorption (including that of CDOM) that covary with phytoplankton fluorescence.

shows diffuse attenuation spectra computed from irradiance data shown in the previous figure. In this example the upper mixed layer of water shows the highest attenuation because of the higher concentration of phytoplankton there. Compared to surface waters, the spectral attenuation at 35-40 m depth is reduced in the blue and UV wavelengths because the chlorophyll concentration is 42% lower than at shallower depths. To show the residual effects on spectral attenuation caused by other substances (DOM and suspended non-algal particles), adjusted Kd values were calculated (by regression of Kd versus algal biomass as estimated from chlorophyll fluorescence) and are plotted as "1-30 m minus phytoplankton" in Figure 2. An estimate of attenuation by pure seawater, Kw, from Smith and Baker [18] is presented as a contrast to "Kd minus phytoplankton" and suggests the magnitude of attenuation by these other substances. Subtracting Kw from "1-30 m minus phytoplankton" yields an exponential curve (exponent = —0.015) typical of CDOM (Figure 6 below).

Rearranging equation (1) and substituting the symbol ( (Greek "z") for Ln(£d,o-/£d,z), yields Kirk's [6] general equation for the depth (in meters) at which irradiance for a specific wavelength is reduced from 100% just below the surface to n% in a uniformly mixed water column:

Kirk [6] calls C the "optical depth" while Mobley [5] uses "optical depth" differently. Using Kirk's definition for and using/for the fraction of surface irradiance reaching Z„»/o, the general solution for optical depth is ( = Ln{f~with

C= 1, 2.3, and 4.6 corresponding to (fx 100) = 37%, 10%, and 1% respectively. Calculations of Z10«/o and Zi»/o have been used increasingly in the UY literature [e.g., 10,12,19], but infer significance primarily from the PAR waveband where Z\o/o is considered the bottom and Z10% the midpoint of the euphotic zone for photosynthesis [6]. No general name for these measures is generally accepted, although Williamson [20,21] used Za (which he referred to as "the attenuation depth") for the specific case of the 1 % attenuation depth. This author suggests that Zno/o x be referred to as the "% attenuation depth" to serve as a general term for the depth at which irradiance is reduced to n% of the value just below the surface. Equation (2) can be reduced to:

Lacking biological or physical significance for a particular percentage within UV wavelengths, the 37% attenuation depth has the advantage that it is less likely to extend below the mixed layer in the water column. Because Kd is typically determined from mixed layer measurements, the 1% and 10% attenuation depths are more likely to misrepresent the penetration of UVR (specifically, whenever the computed value of Zn% extends below the mixed layer). This is because Kd often changes below the mixed layer, either increasing (e.g., Figures 5A and 12, also [22-24]) or decreasing (e.g Figure 2, also [19]) compared to surface Kd values. Figure 3 (adapted from Whitehead et al. [12] using data from Table 1) shows Z37o/o attenuation depths for 320 nm UVR for the mixed layer at a number of freshwater and marine sites. Also included in Figure 3 are estimates for pure water calculated by Smith and Baker [18] and a new estimate for pure water (described in Section 3.3.4).

The diffuse attenuation coefficient (Kd) is one of several "apparent optical properties" (AOPs) of natural waters described by Preisendorfer [25]. Unlike inherent optical properties (IOPs) described below, AOP's depend on the quality of incident light as well as the optical qualities of the water. In spite of this apparent limitation (and in part because the differences between AOP's and IOP's were said to be small in many instances [26]), a case was argued for the standard use of Kd to characterize natural waters for purposes of optical comparisons and bio-optical models [27,28]. Gordon [17,29] provided a practical means to adjust Kd measurements to remove much of its dependence on the ambient light field. In particular, Gordon [17] established that, after adjustment (described below), Kd$ averaged from surface to Z10o/o is proportional to the summed concentrations of constituent optical compounds.

In contrast to AOP's, "inherent optical properties" (IOP's) depend solely on the water and its optically active constituents. The IOP's include the beam absorption coefficient "a", beam scattering coefficient "b", and beam attenuation coefficient "c", which are related as follows:

The absorption coefficient "a" is the sum of absorption by constituent components (including the solvent, water) and is proportional to the concentration of absorbing substances. Similarly, the scattering coefficient is the sum of constitu-

Was this article helpful?

0 0
Solar Power Sensation V2

Solar Power Sensation V2

This is a product all about solar power. Within this product you will get 24 videos, 5 guides, reviews and much more. This product is great for affiliate marketers who is trying to market products all about alternative energy.

Get My Free Ebook


Post a comment