Modeling climateUV interactions in aquatic ecosystems

The complex spatial and temporal interactions between climate change and UVR make it difficult to model and predict effects of these processes on aquatic ecosystems. Ozone depletion, as modified by greenhouse gases, will alter the levels of UV incident at the water's surface. In contrast, climate change is more likely to alter the underwater UV environment by influencing watershed and in-lake processes (Figure 6). Removing riparian or littoral vegetation, clear-cutting forests, increasing water usage, and other anthropogenic or climate-induced changes all may alter water transparency, surface water temperatures, and mixing depths. The exposure of aquatic ecosystems to underwater UV involves a complex set of feedback loops between CDOM, UV, PAR, IR, and temperature (Figures 1 and 6). These diagrams (Figures 1 and 6) do not even consider the complex interactions between UV damage and variability in the

Interaction Among Aquatic Organisms

Figure 6. Conceptual model of how the key ecosystem level processes involving CDOM (Figure 1) and the molecular, organism, population, and community level effects of temperature and UV (Figure 5) can be integrated within the context of climate change and ozone depletion. This makes lakes particularly good models for examining the interactive effects of temperature and UVR on natural ecosystems.

Figure 6. Conceptual model of how the key ecosystem level processes involving CDOM (Figure 1) and the molecular, organism, population, and community level effects of temperature and UV (Figure 5) can be integrated within the context of climate change and ozone depletion. This makes lakes particularly good models for examining the interactive effects of temperature and UVR on natural ecosystems.

responses of organisms from the cellular and subcellular to the population and community levels (Figure 5).

Some progress has been made in recent years toward quantitative modeling of the diverse processes that influence the effects of UV on aquatic ecosystems. At the heart of these models is the biological weighting function (BWF) concept. Biological weighting functions quantify the wavelength-dependence of photon-specific effects of UV, where shorter wavelengths of UV are generally much more damaging than longer wavelengths [89,90]. In order to estimate the biological effectiveness of a given set of UV conditions, the incident UV spectrum must be multiplied by the BWF for the organism of interest (Figure 7). While this quantitative approach has its limitations, it is essential to understanding how climate change-induced variations in the quality and quantity of CDOM alter the incident UV spectrum, and how aquatic organisms and processes will respond. For example, even within a single lake there are strong variations in the wavelength-specific absorptivity of CDOM that alter the spectral slope and thus the selective attenuation of different wavelengths of UV within lakes [16]. Another example is found in endorheic systems where CDOM is exposed to sunlight over a period of many years and consequently this CDOM absorbs very little UV compared to younger CDOM. Climate change is likely to reduce stream flows, create more endorheic systems, and reduce CDOM inputs from

Figure 7. Biologically effective exposure (BEE) for mortality in the cladoceran Daphnia pulicaria as estimated from a 7 h solar phototron exposure experiment. BEE is estimated by multiplying the biological weighting function (BWF, an estimate of the wavelength-specific effects of UV), times the cumulative solar energy spectrum (here a 7 h exposure period during midday). [Modified from [59], with permission.]

280 300 320 340 360 380 400

Wavelength (nm)

Figure 7. Biologically effective exposure (BEE) for mortality in the cladoceran Daphnia pulicaria as estimated from a 7 h solar phototron exposure experiment. BEE is estimated by multiplying the biological weighting function (BWF, an estimate of the wavelength-specific effects of UV), times the cumulative solar energy spectrum (here a 7 h exposure period during midday). [Modified from [59], with permission.]

wetlands. All of these factors will potentially increase the UV exposure of ecosystems. The strong changes in the spectral composition of UV related to changes in quality and quantity of CDOM as well as to changes in water depth must be accounted for when estimating the impact of these increased UV exposures. The responses of both living organisms and chemical processes are also strongly wavelength dependent, and must be accounted for with weighting functions.

To date, BWFs have been developed for a variety of aquatic organisms ranging from marine phytoplankton [89-91], to marine copepods and fish [92,93], freshwater copepods [94], and freshwater cladocerans [95]. The application of BWFs is dependent upon the reciprocity principle, which states that the effect of a given dose is independent of the rate at which the dose is administered [89,96]. In essence, the reciprocity principle is not valid if a high dose rate administered over a short period of time gives a different response than a low dose rate of the same total dose administered over a longer time period.

Some of the first modeling efforts to quantify the effects of ozone depletion on aquatic ecosystems were with phytoplankton photosynthesis in Antarctica where the close proximity of regions of high and low ozone facilitated intercom-parisons [97,98]. These studies indicated that inhibition of phytoplankton photosynthesis related to ozone depletion was <5%. In more recent years enough information has accumulated on BWFs for photosynthesis in phytoplankton that comparative summaries are possible. Work on the Rhode River, a sub-estuary of the Chesapeake Bay, has demonstrated short-term variation in the BWFs for phytoplankton, but no significant patterns of seasonal variation [99]. Comparisons among systems in this same study demonstrated striking similarities for the BWFs from the Rhode River and Antarctic phytoplankton.

Three recent modeling efforts have made particularly important progress towards quantifying and integrating the various factors that influence the effects of UV on aquatic organisms. Neale and others [55] developed a model for estimating the effects of ozone depletion and vertical mixing on photosynthesis in Antarctic phytoplankton. This modeling effort included sensitivity analyses of the effects of variations in ozone depletion, phytoplankton sensitivity to UV, vertical mixing, and cloud cover. The results indicated that a 50% reduction in stratospheric ozone could reduce water column integrated photosynthesis by up to 8.5%. The other factors tested were predicted to have a stronger influence than ozone depletion. This included effects of variations in phytoplankton sensitivity to UV (±46%), vertical mixing (±37%), and cloud cover (± 15%). Subsequent modeling efforts by Neale include separate terms for repair as well as damage by UV [100],

Huot et al. [40] developed a numerical simulation model for estimating DNA damage in marine bacterioplankton based on spectral scalar irradiance, biological weighting functions for DNA damage and photoenzymatic repair, and vertical mixing. The effects of chlorophyll and CDOM on spectral composition were incorporated into the irradiance portion of the model, and excision repair was included. One of the strengths of this model is the explicit inclusion of photoenzymatic repair (PER) of UV-damaged DNA, based here on the action spectrum for PER in Escherichia coli. This is an important step forward because of the dependence of the application of B WFs on the validity of the reciprocity principle [89,96]. In the presence of PER, the reciprocity principle may be invalid [101], and the ability to extrapolate BWFs to different exposure regimes is severely compromised [64]. The results of the Huot et al. model agreed reasonably well with estimates of net DNA damage (cyclobutane pyrimidine dimers) from field measurements and experimental incubations of dosimeters and natural bacterioplankton.

Kuhn et al. [41] have developed a numerical simulation model for looking at UV effects on the embryos of marine copepods and fish that also includes spectral irradiance, BWFs for embryo mortality, and vertical mixing processes. Their model differs from that of Huot's in that it uses downwelling rather than scalar irradiance, and it models mortality (daily survival) rather than net DNA damage. The results of the simulations suggest that cod embryos (Gadus morhua) are insensitive to environmental UVR in the St. Lawrence estuary, with the average daily survival rates on the order of 99%. The calanoid copepod Calanus finmarchicus was found to be more sensitive to environmental UV with daily survival rates reduced to about 90%. These modeling efforts do not include the potential for more chronic sublethal effects on these organisms and in this sense they may underestimate UV impacts.

Sensitivity analyses were carried out with both of these simulation models to examine which factors were most important in regulating UV damage in aquatic ecosystems. Both models found that changes in UV attenuation by the water column due primarily to mixing processes are of primary importance in determining UV damage [40,41]. Over the range of DOM concentrations for the marine systems tested by Huot et al. (0-0.5 g m-3, much lower than most freshwater systems), variations in DOM were found to be of little importance.

Extrapolating these results to freshwater systems where CDOM concentrations are generally much higher and more variable and vertical gradients in UV much steeper would suggest that variations in the quantity or quality of CDOM is critical to the extent of UV damage. The sensitivity analyses gave contrasting results for the importance of ozone depletion. Net DNA damage was found to be quite sensitive to changes in atmospheric ozone [40], while ozone depletion had only a minimal influence on mortality in C.finmarchicus [41].

These kinds of quantitative models are essential if we are to integrate the complex processes involving the effects of climate change on UV effects in aquatic ecosystems. The strong wavelength-selective absorption of UV by CDOM, and the variation in the optical characteristics of this CDOM, make such a quantitative approach essential. For example, CDOM with a higher aromaticity will tend to absorb more short wavelength UV than CDOM with lower aromaticity. If one simply estimates UV exposure from changes in CDOM without taking into account the spectral absorption of the CDOM as well as the spectral sensitivity of the organisms or processes of interest, this could result in misleading interpretations or predictions. On the other hand, collecting quantitative data to validate these models will be a great challenge, particularly for higher trophic levels where reciprocity [101], temperature dependent photoen-zymatic repair [68] and active behavioral avoidance [102] may vary greatly among taxa and even within taxa for different life history stages. Other environmental factors such as pH, temperature tolerance, photosensitizers, and toxicants will also have to be taken into account.

0 0

Post a comment