EFFECTS OF INCREASED SOLAR ULTRAVIOLET
RADIATION ON TERRESTRIAL PLANTS
M.M. Caldwell (USA), A.H. Teramura (USA), M. Tevini (FRG ), J.F. Bornman
L.O. Björn (Sweden), and G. Kulandaivelu (India)
Physiological and developmental processes of plants are affected by UV-B radiation, even by the amount of UV-B in present-day sunlight. Plants also have several mechanisms to ameliorate or repair these effects and may acclimate to a certain extent to increased levels of UV-B. Nevertheless, plant growth can be directly affected by UV-B radiation. Response to UV-B also varies considerably among species and also cultivars of the same species. In agriculture, this will necessitate using more UV-B-tolerant cultivars and breeding new ones. In forests and grasslands, this will likely result in changes in species composition; therefore there are implications for the biodiversity in different ecosystems. Indirect changes caused by UV-B (such as changes in plant form, biomass allocation to parts of the plant, timing of developmental phases and secondary metabolism) may be equally, or sometimes more, important than damaging effects of UV-B. These changes can have important implications for plant competitive balance, herbivory, plant pathogens, and biogeochemical cycles. These ecosystem-level effects can be anticipated, but not easily predicted or evaluated. Research at the ecosystem level for solar UV-B is barely beginning. Other factors, including those involved in climate change such as increasing CO2 also interact with UV-B. Such reactions are not easily predicted, but are of obvious importance in both agriculture and in nonagricultual ecosystems.
Since the first reports of potential stratospheric ozone reduction over 20 years ago (e.g., [Johnston, 1971; Crutzen, 1972]), UV-B (280-315 nm) effects on higher plants have been the subject of considerable research. Approximately 350 papers have appeared, but the majority of these deal with herbaceous, agricultural plants under laboratory or glasshouse conditions. Fewer than 5% of the studies have been conducted under field conditions, and fewer still with plants from forests and other nonagricultural systems. While the laboratory and glasshouse studies provide information on mechanisms and processes of UV-B action, only the field studies can provide realistic assessments of what will happen as the stratospheric ozone layer thins.
Several reviews of this literature have appeared in the last five years [Bornman, 1989; Caldwell et al., 1989; Tevini and Teramura, 1989; Krupa and Kickert, 1989; Tevini, 1993; Bornman and Teramura, 1993; Caldwell and Flint, 1993, 1994a, 1994b; Tevini, 1994; Teramura and Ziska, 1994; Teramura and Sullivan, 1994]. The present chapter provides an overview with interpretation of the results for both agriculture and other ecosystems such as forests, grasslands, etc. It also includes a brief consideration of the potential effects of the breakdown products of new man-made chemicals being brought into use which are less offending to the ozone layer.
Figure 3.1 shows some of the effects of UV-B radiation on plant processes.
Fig. 3.1. The influence of the UV-B radiation on plant processs.
As explained in Chapter 1, the biological effectiveness of UV-B solar radiation needs to be considered in assessing what ozone reduction, and the resulting changes in solar radiation, potentially mean for plants and other organisms. The biological weighting functions used for this purpose often come from action spectra. Action spectra assumed to be relevant for plants (Figure 3.2) all indicate that the shorter UV-B wavelengths are the most important. However, the relative importance of shorter vs longer UV-B wavelengths (the slopes in Figure 3.2) vary considerably. Depending on these slopes, the Radiation Amplifications Factors (discussed in Chapter 1) vary enormously. Action spectra that do not decrease sharply with increasing wavelength result in small RAF values. Thus, the evaluation of weighting functions (and therefore action spectra) is critical. There is evidence that action spectra for many plant functions are steep indicating that ozone reduction translates into large increases in effective solar UV-B [Caldwell, 1971; Setlow 1974]. Some more recent spectra developed specifically for evaluating the ozone reduction problem show flatter slopes (and therefore lower RAF values) than the earlier work [Caldwell et al., 1986; Steinmueller, 1986; Quaite et al., 1992]. Still, these spectra are sufficiently steep so that ozone reduction must be taken seriously. Biological weighting functions also are needed to relate solar UV to UV from artificial sources used in experiments.
Fig. 3.2. Action spectra for DNA damage [Setlow, 1974], DNA dimer formation (a type of DNA damage) in alfalfa seedlings [Quaite et al., 1992a], growth inhibition in seedlings [Steinmüller, 1986], and generalized plant responses [Caldwell, 1971]. The generalized plant action spectrum was developed from action spectra available in 1971 for several processes of higher and lower plants. It has been widely used for calculating UV irradiance in experiments with higher plants. Solar spectral irradiance at 360 and 180 Dobson Units (DU) of atmospheric ozone is also shown. (A Dobson Unit is an expression used for describing thickness of the ozone layer at standard temperature and pressure; 1 mm ozone layer thickness is equivalent to 100 DU.) The solar irradiance is calculated for latitude 49° at solar noon on Julian date 173 using the model of Green et al., .Apart from action spectrum considerations, it is important in experiments to maintain a realistic balance between different spectral regions since both UV-A (315-400 nm) and visible (400-700 nm) radiation can have strong ameliorating effects on responses of plants to UV-B [Caldwell et al., 1994]. In growth chamber and greenhouse experiments, the visible and UV-A radiation is usually much less than in sunlight. Thus, even if realistic levels of UV-B are used in simulating ozone reduction, the plant response may be exaggerated relative to field conditions. Even under field conditions, if applied UV-B is not adjusted downward during cloudy periods, the UV-B sensitivity may be unduly pronounced. Unfortunately, the most expensive and difficult experiments, i.e., those conducted in the field with UV-B supplements adjusted for cloudiness and other atmospheric conditions, are seldom undertaken.
As indicated in subsequent sections, another approach which achieves appropriate spectral balance is to filter existing solar radiation to modify the UV-B component. This has been done either with materials such as plastic filters or using ozone gas as a filter [Tevini et al., 1990]. For the latter system, in identical growth chambers, different UV-B levels can be achieved by filtering sunlight using ozone (the ozone gas is passed through an envelope of UV-transparent Plexiglas). In these growth chambers other factors such as temperature and CO2 can be controlled. This is a considerable technical improvement on other UV-B modification experiments using filtering materials such as glass, plastic films, etc. Growth and other responses of intact seedlings can be evaluated in these chambers, but they are too small for larger plants. Of course, all these UV-B filtering approaches can only result in lowering, but not increasing, solar UV-B under the particular conditions experienced.
Enhanced UV-B radiation can have many direct and indirect effects on plants including inhibition of photosynthesis, DNA damage, changes in morphology, phenology, and biomass accumulation. Most of the work to date has concentrated on crop plants from temperate regions, whereas little has been done on tropical and nonagricultural plants. Sensitivity to UV-B radiation, defined as the relative change induced by UV-B on plant growth, morphology, or yield, depends on plant species, cultivar, developmental stage and experimental conditions.
In many plant species reduced leaf area and/or stem growth have been found in studies carried out in growth chambers, greenhouses and in the field [Tevini and Teramura, 1989; Johanson et al., 1994]. Studies where the ozone filter technique with special growth chambers under solar radiation (in Portugal at 39 ° latitude) showed that higher solar UV-B (close to ambient levels) can result in smaller plants with reduced leaf area compared with plants under reduced UV-B levels [Tevini et al., 1990, 1991a; Mark, 1992]. These observations correspond with results obtained in other studies with artificial UV-B in greenhouses, growth chambers [Tevini and Iwanzik, 1986] and in the field [Teramura and Murali, 1986]. To provide differently filtered solar UV-B on a relatively larger scale, different thicknesses of UV-transparent Plexiglas have been used in greenhouses. The greenhouses (also located in Portugal) were covered with either 3 or 5 mm Plexiglas, providing a 10% difference in weighted solar UV-B. Even with this small difference in solar UV-B attenuation, reductions in growth of different cultivars of bean were observed under the higher level of solar UV-B (with the 3-mm Plexiglas).
Two studies on rice cultivars from different geographical regions were carried out in greenhouses with an enhanced daily dose of UV-B radiation using lamps [Teramura et al., 1991]. Of 16 rice cultivars native to the Philippines, India, Thailand, China, Vietnam, Nepal and Sri Lanka, about one third showed statistically significant decreases in total biomass and leaf area. Tiller number, correlated with yield, was reduced in 6 of the cultivars (Figure 3.3). The Sri Lanka cultivar, Kurkaruppan, however, showed increases in both total biomass and tiller number, indicating that selective breeding might be a successful tool for obtaining UV-B tolerant cultivars [Teramura et al., 1991]. In another study with rice cultivars from the Philippines, total biomass changes were different among cultivars, with IR74 being the most sensitive and IR64 the least sensitive [Figure 3.3; Barnes et al., 1993]. Field experiments are currently underway in the Philippines using modulated UV-B lamp systems which should provide some realistic estimates of rice cultivar response to UV-B radiation.
Fig. 3.3.A. Summary of a greenhouse study examining 16 rice cultivars grown with and without UV-B radiation simulating a 20% ozone depletion over the equator. The proportion of cultivars which has significantly reduced biomass, leaf area, and tiller number are shown in percents [Teramura et al., 1991].
Fig. 3.3 B. Selected data from a study examining 22 rice cultivars grown with and without UV-B radiation simulating a 5% reduction in stratospheric ozone in Spring for the Philippines. The proportion of cultivars with significantly reduced total weight, leaf area, and tiller number are shown in percents [Barnes et al., 1993].
The molecular basis for many of the changes observed following UV-B exposure is not yet well defined. Responses may result from direct damage to essential cell components and by UV-B absorbed by specific photoreceptors or growth regulators [Ensminger and Schäfer, 1992; Ballaré et al., In press]. Preliminary experiments suggest that flavins may function as UV-B photoreceptors for the induction of pigment synthesis and inhibition of elongation [Ensminger and Schäfer, 1992; Ballaré et al., In press]. Elongation growth is influenced by the auxin, indole acetic acid, which absorbs in the UV-B range and could be photodegraded by high levels of UV-B radiation. Oxidative enzymes, such as the peroxidases, are increased by enhanced UV-B radiation, and may be involved in plant hormone regulated growth responses, as shown in sunflower [Ros 1990]. The levels of another plant hormone, ethylene, which causes greater radial growth and less elongation, are increased after irradiation with UV-B in, e.g., sunflower seedlings [Ros 1990] and cultured shoots of pear plants [Predieri et al. 1993].
Ultraviolet-B radiation can alter both the time of flowering [Caldwell, 1968; Ziska et al., 1992; Saile-Mark, 1993; Staxén and Bornman, 1994] as well as the number of flowers in certain species. For example, Rau et al.,  found substantial decreases in flowering from UV-B irradiation. Differences in timing of flowering may have important consequences for the availability of pollinators. The reproductive parts of plants, such as pollen and ovules, are rather well shielded from solar UV-B radiation. For example, anther walls can absorb more than 98% of incident UV-B [Flint and Caldwell, 1983]. In addition, the pollen wall contains UV-B absorbing compounds affording protection during pollination. Only after transfer to the stigma might pollen be susceptible to solar UV-B radiation. Germinating pollen can be sensitive at this time to UV-B [Flint and Caldwell, 1984]. The overall significance of this in the context of the ozone reduction problem is unclear and needs to be assessed.
Limitation of Growth Chamber Studies
In growth chambers and greenhouses the radiation conditions are usually quite different from those in the field. For example, the visible radiation which is used in photosynthesis (400 to 700 nm, photosynthetically active radiation, PAR) and the UV-B/UV-A/PAR ratios are different from those in the field. As mentioned earlier, if UV-A and PAR are low, the effects of UV-B may be much more severe (see above). In addition, other factors, such as temperature, water and nutrients differ from conditions in the field and this can alter response to UV-B radiation. However, experiments in controlled conditions are usually necessary as a first step in defining plant response to specific combinations of UV-B and other environmental factors. It is, however, important that these studies conducted under controlled conditions be verified as much as possible under field conditions.
Plants compose most of the living mass in terrestrial ecosystems. Although there can be effects of UV-B directly on microbes and animal life (e.g., [Blaustein et al., 1994; Gehrke et al., In press], see also Chapters 2 and 4), most of the ecosystem-level responses of solar UV-B are anticipated to be mediated through the effects on plants. As shown in Figure 3.4, the major anticipated effects of increased solar UV-B on agricultural and nonagricultural ecosystems (such as forests, grasslands, savannahs, deserts, tundra, etc.) may result from changes in plant growth and form and secondary chemical composition. Although the principal processes may be the same in highly managed agroecosystems and in nonagricultural ecosystems, their importance is thought to be different. Therefore, different schemes are presented for each in Figure 3.4. Some forest systems, such as plantations, can be considered as agricultural systems for these purposes.
Fig. 3.4. Possible important consequences of increased solar UV-B in highly managed systems such as agricultural and forest plantation systems and in nonagricultural, less intensively managed ecosystems.
In forests, grasslands, etc., overall primary plant productivity may not be greatly affected by ozone reduction even if the growth of some plants is diminished. However, since plant species differ greatly in growth responsivity to UV-B, it is anticipated that a productivity reduction of one species will probably lead to increased productivity of another, more UV-tolerant species. This is conceivable because more resources (e.g., light, moisture and nutrients) will be available to the tolerant species. Thus, the overall productivity of the system may well remain about the same while species composition may change. However, a change in the balance of species could have far-reaching consequences for many ecosystems.
Another mechanism whereby the competitive balance of plant species can be changed by increased UV-B is through changes in plant form. Even if plant production per se is not affected by increased UV-B, changes in plant form can result in changes in which species can more effectively compete for sunlight. This phenomenon has been demonstrated in several experiments. For example, in a five-year field study using modulated UV-B lamp systems, the competitive balance of two species (wheat and a common weed, wild oat) could be changed even though the increased UV-B had no effect on production and growth of these species if grown by themselves [Barnes et al., 1988]. A quantitative analysis of competition for sunlight in the mixed stands with and without supplemental UV-B showed that subtle changes in plant form of the two species were sufficient to change the balance of competition for sunlight which is necessary for photosynthesis [Ryel et al., 1990]. Therefore, one species can achieve some advantage over the other because one captures more sunlight for photosynthesis. In these experiments, the wheat benefited from increased UV-B and the weed suffered. However, in other mixtures of crop and weeds, the situation could well be reversed. Also, other changes in plant form, such as greater allocation of biomass to roots, might change competitive effectiveness of individual species for soil moisture and nutrients.
In grasslands and forests that are not managed intensively, similar changes in species composition may be experienced. Of course, in forests this would take a long time to be realized. Also, if there are only a few tree species present and they all are sensitive to solar UV-B and experience growth reduction, overall forest productivity could decrease. Ecosystem-level experiments with nonagricultural systems are only beginning [Johanson et al., In press].
Timing of Life Phases
The timing of life phases of plants is a combination of response to environmental factors and the genetic constitution of the plant. This timing of events such as flowering, entering and breaking of dormancy, and even senescence is important not only to the individual plant, but also in how plants interact with other plants and animals. For example, a shift in the timing of flowering can mean that a plant species might not have sufficient insect pollinators available at the new time of flowering either because the insects are not present or because other plant species are attracting these pollinators. Such changes could also conceivably be important in agricultural systems, but intervention with management options may make these changes less important. As indicated earlier in this chapter, increased UV-B has been shown to advance or delay (depending on species) the time of flowering in plants. There is little work at present on flowering responses and virtually nothing on other potential effects of UV-B on life phase timing.
Plant Secondary Metabolism
Another pathway by which increased solar UV-B can have an influence at the ecosystem level is through changes in secondary metabolism of plant tissues. Increased UV-B can alter secondary chemical composition. It has been shown repeatedly that flavonoids and related phenolic compounds increase when plants are exposed to increased UV-B. Apart from the UV-B protection afforded by increases of these compounds, there are many other ecological implications of changes in these and related compounds. These compounds are important for plants in deterring insects and other herbivores from consuming plant tissues and they play a role in resistance to pathogens. For example, McCloud and Berenbaum  have shown that UV-B can increase furanocoumarin content of plant tissue which, in turn, results in slower development of certain insect larvae during early life stages of the larvae. In some legume, conifer and in one woody dicotyledon (Vitis vinifera) UV-B has been shown to induce phytoalexin synthesis [Beggs and Wellmann 1994]; some phytoalexins are considered to be toxic to humans and many animal species, e.g., coumestol a compound with estrogenic properties which is induced in bean plants exposed to UV-B [Beggs et al. 1985].
Secondary compounds that are important as structural materials in plants, such as lignin, are also related to flavonoids and phenolic compounds. If the ratio of lignin to cellulose in plant tissues changes, it can alter the rate of decomposition. This has very important implications for biogeochemical cycles as discussed in Chapter 5.
Overall, the consequences of increased solar UV-B in forests, grasslands and other nonagricultural ecosystems may involve several complex pathways (Figure 3.4) rather than simply a reduction in overall ecosystem primary productivity. However, the effects of these more involved pathways are difficult to predict without conducting experiments with assemblages of plant species and long-term study of ecosystem responses. This has, thus far, received very little attention in actual research. Further discussion of implications for specific types of ecosystems follows later in this chapter.
Structural and biochemical changes induced by enhanced levels of UV-B radiation ultimately modify the penetration of UV radiation into the plant. For example, the induction of UV-screening pigments, typically flavonoids and certain other phenolic compounds, will reduce the penetration of UV-B radiation to underlying tissues. Increased wax on leaf surfaces also can contribute to reduced penetration of UV due to increased reflection from the leaf surface, although reflection for most leaves is usually not more than 10% [Robberecht et al., 1980]. At the structural level, increased length of inner leaf cells or increases in cell number, both palisade and spongy mesophyll, influence the penetration and spectral distribution of UV radiation across a leaf. Direct measurements of UV penetration have been done using a fibre optic microprobe [Bornman and Vogelmann, 1988; Day et al., 1992, 1993; see also Bornman and Teramura 1993; DeLucia et al., 1992; Cen and Bornman, 1993]. Ultraviolet radiation penetration varies among different plant species and this should be reflected in the sensitivity of these species. Penetration of UV-B was found to be greatest in herbaceous dicotyledons (broad-leaved plants) and was progressively less in woody dicotyledons, grasses and conifers [Day et al. 1992]. The UV penetration also changes with leaf age; younger leaves attenuate UV-B radiation less than do the more mature leaves, as was shown for some conifers [DeLucia et al. 1991, DeLucia et al. 1992].
Although different species exhibit different UV-B attenuation depending on pigments (such as flavonoids) and leaf structure, the level of attenuation can also change as more UV-B-absorbing pigments are synthesized in response to UV-B exposure. Species also differ in their ability to increase UV-B-absorbing pigment levels. Increased levels of flavonoids have been shown to directly reduce the levels of damage by UV-B [Tevini et al., 1991b]. The link between flavonoid levels and UV-B sensitivity is most vivid in extreme cases. For example, Li et al.  showed that mutants of a mustard species, Arabidopsis thaliana, that lacked flavonoids were extremely sensitive to the UV radiation. An anthocynanin-deficient mutant of maize was found to be more sensitive to DNA damage by UV-B than the normal plants [Stapleton and Walbot 1994]. In cabbage leaves, flavonoids were shown to accumulate in the epidermal layers in response to mild UV-B exposure. The flavonoids protected the underlying tissues from DNA damage in the form of thymine dimer formation [Beggs and Wellmann 1994]. In addition to adaptive responses involving pigment induction, changes in surface waxes and certain leaf structural characteristics may also contribute to reducing penetration of UV into the plant tissues. Increases in scavengers of free radicals and active oxygen species may also mitigate the negative effects of UV radiation [see Bornman and Teramura, 1993].
Another characteristic of plants is the ability to repair damage. This may be exemplified by the replacement of damaged components, e.g., proteins, or by the light and dark processes involved in repair of DNA damage. Under illumination, the enzyme DNA photolyase repairs the UV-induced production of pyrimidine dimers. Both visible and UV-A radiation drive this repair, underlining again the importance of a balanced spectral regime for experiments not conducted in the field. A very effective repair capacity by DNA-photolyase in several plant systems (cell cultures, isolated leaves, seedlings and mature plants of several species) has been demonstrated [Buchholz et al., In press; McLennan 1987]. To date, little research has been done on induction of pyrimidine dimers in plants; a recent exception is the work of Quaite et al. [1992 a,b] on alfalfa seedlings.
Plants in nature seldom are affected by only a single stress factor, such as UV-B radiation. Instead, plants typically respond to several factors acting in concert. Therefore, it is important to keep in mind that the effectiveness of UV-B radiation can be greatly modified by some of these other factors, in some cases aggravating, and in some cases ameliorating the overall UV effect.
Water stress commonly occurs in nature. In a field study specifically designed to test the interaction between UV-B radiation and water stress, Sullivan and Teramura  demonstrated that UV-B mediated reductions in photosynthesis and growth were observed only in well-watered soybeans. When soybeans were water stressed, the same UV-B dose produced no significant effect on either photosynthesis or growth. The interpretation of these observations was that water stress produced a large reduction in photosynthesis and growth that thereby masked the UV-B effect. Furthermore, water stressed plants produced a higher concentration of leaf flavonoids, which in turn, provided greater UV-B protection.
Increases of atmospheric CO2 and global warming are anticipated in scenarios of future climate change. Model calculations predict that the average global temperature will rise by 1.5 - 4.5° and that atmospheric CO2 concentration will double by the latter part of the next century [Intergovernmental Panel on Climate Change 1992]. When studied independently, plant growth responses to changes in UV-B radiation and atmospheric CO2 concentration generally are thought to be in opposite directions, thereby leading some to the hypothesis of a canceling of effects. To date, only a few experiments have been specifically designed to examine this important interaction. The first such study [Teramura et al., 1990b] included two cereal crops (rice and wheat) and one legume (soybean). In that study, the increase in growth and seed yield resulting from a CO2 enrichment was eliminated by UV-B radiation in rice, reduced in wheat and unaffected in soybean. This suggests that overestimates in production may be made for cereal crops if CO2 enrichment is considered without UV-B radiation.
In loblolly pine, Sullivan and Teramura  reported that the combined effects of UV-B and CO2 produce changes in the proportion of dry matter in roots compared with aboveground shoots. At present-day levels of CO2, UV-B caused more shoot production than roots, while the same UV-B dose resulted in more roots at elevated levels of CO2. The implications of these changes is that plant competition and, therefore, ultimately community composition might be altered by these changes in allocation patterns, as described earlier.
As mentioned above, global climate change will likely include increased global mean temperatures in addition to increased UV-B radiation. Unfortunately, only very limited information is available on the consequences of these combined effects. In a study with sunflower and maize seedlings in the ozone-filter cuvette system mentioned earlier, Tevini et al. [1991b] found that photosynthesis was unaffected (maize) or declined (sunflower) with higher temperature when the UV-B in sunlight was attenuated with the ozone filter. In contrast, overall seedling production was greater at higher temperatures in both species under conditions approaching ambient solar UV-B. This observation might be attributed to accelerated plant development at the higher temperature.
Many metals such as cadmium, nickel, copper and lead, which can accumulate to high concentrations from human activities, are toxic to plants. The stress imposed by high levels of metals may be further compounded with increased UV-B radiation. Dubé and Bornman  showed that in spruce (Picea abies) with low levels of supplemental UV-B, the effect of the addition of cadmium was greater than for either stress alone for some of the parameters measured. Mobilization of the essential trace element zinc can be reduced by UV-B radiation [Ambler et al., 1975].
The degree of susceptibility of plants to disease and insect attack may change under elevated levels of UV-B radiation. These effects will vary among species, cultivar and plant age. For example, certain diseases may be less damaging to the plant under conditions of high UV-B, while the severity of others may be increased. The latter was shown in a study where sugar beet grown under elevated levels of UV-B radiation was infected with a fungus Cercospora beticola leading to a deleterious additive effect from the two stress factors [Panagopoulos et al., 1992]. The timing of infection is also of importance. Plants first exposed to UV-B radiation may be more susceptible to subsequent infection as shown in a study of cucumber infected with Colleotrichum lagenarium and Cladosporium cucumerinum prior to UV-B exposure [Orth et al., 1990]. Infection after UV exposure had no effect on the severity of the disease.
Interaction of UV-B with tropospheric air pollutants is also of concern although little work has been thus far conducted in this area. One field study of soybean plants showed them to be sensitive to ozone in the air. However, they were not sensitive to UV-B supplements from lamps under the particular test conditions and there were no significant interactions of supplemental UV-B and ozone [Miller et al., 1994].
One of the primary concerns of future increases in solar UV-B radiation is its potential effect on global agriculture. Despite the enormous potential consequences, we cannot yet make a quantitative prediction of anticipated effects resulting from stratospheric ozone depletion. This results from the limitation in the controlled-environment studies as discussed earlier and the overall paucity of experiments performed in field trials. Even in comparisons of field studies, there are large differences in temperature, precipitation, soil types, etc. from year to year and in different locations. This adds to the difficulty in making generalizations about the effects.
A six-year field experiment was conducted to evaluate the effects of UV-B supplementation in two commercially grown soybean cultivars [Teramura et al., 1990a]. These cultivars were specifically selected for their contrasting UV-B sensitivity previously determined by screening over 50 soybean cultivars under greenhouse conditions. (For perspective, nearly 2 out of 3 of the cultivars screened for UV-B sensitivity in the greenhouse exhibited sensitivity to UV-B.) In the field experiments, artificial lamps with selected filters were used in addition to the normal solar radiation. Plants were exposed to either ambient levels of solar UV-B or ambient radiation supplemented with UV-B emitted by the lamps. When evaluated over the entire six-year period, yield in the sensitive cultivar was reduced by 19 to 25%, in four of the six years (Figure 3.5). The other two years were characterized as hot and dry and all plants in the field experienced considerable water stress. As shown in subsequent field and greenhouse studies, the effectiveness of UV-B radiation is masked when the plants were subjected to other stresses such as drought (see section above).
In the same study, while the sensitive soybean cultivar exhibited decreased yield, production increased by 4 to 22% in the tolerant soybean cultivar. Such cultivar differences in the response to UV-B radiation may be important in future plant breeding considerations, since it suggests that UV-B tolerance already naturally exists in the modern soybean germplasm. A number of studies have shown that in addition to the wide range of sensitivity found among species, an impressive array of cultivar responses also have been observed. Figure 3.3 shows the proportion of sensitive rice cultivars screened under greenhouse conditions. Note that in this example, UV-B radiation elicits both positive as well as negative responses in rice. Similarly, wide-ranging cultivar differences have been reported in a number of plant ranging from crops such as soybean to forest tree species such as loblolly pine.
Fig. 3.5. Summary of yields in a 6-year field study using two soybean cultivars: Essex previously identified as UV-B sensitive and Williams, UV-B tolerant. The studies utilized filtered fluorescent sun lamps to simulate a 25% ozone depletion over College Park, Maryland (39° N latitude). Values presented above represent percent changes from control plants receiving only ambient levels of UV-B radiation [Teramura et al., 1990a]. Asterisks represent years with drought.
In addition to quantitative changes in crop yield, evidence exists for qualitative changes as well. For instance, in the study mentioned above, UV-B radiation also resulted in small changes on the order of 1 to 5% in the protein and oil content of the soybean seed.
A wide range of experimental protocols and methodologies have been used by different investigators which complicates the assessment of overall effects of elevated UV-B on crops. Several experiments reveal effects of UV-B and several do not. Because of the relatively few field studies that have been conducted, a quantitative prediction of the potential consequences for global food production resulting from increased solar UV-B is not now possible.
Despite the fact that over two thirds of global terrestrial productivity occurs in forest ecosystems, little information exists on the effects of UV-B radiation on forest tree species. Tropical forests, though representing nearly one half of global productivity and much of the total tree species diversity, have received very little attention thus far in respect to the ozone reduction problem. Although little ozone reduction has thus far occurred in the tropics, only a small decrease of ozone at these latitudes results in an absolute increase of UV-B since solar UV-B is already very intense (see Chapter 1). One recent study showed that excluding existing solar UV-B with filters can result in increased growth of some tropical tree species [Searles and Caldwell, in press]. However, for the most part, the effects of UV-B radiation on tropical tree species have been largely ignored.
Fortunately, there is some information for midtemperate latitude tree species. Because they are long lived, trees present a unique opportunity to observe the longer-term accumulative aspects of UV-B exposure. In one field study using loblolly pine [Sullivan and Teramura, 1992], seedlings from several different geographic regions were grown for three consecutive years under UV-B lamps in a field experiment. Seedlings were exposed to either ambient solar UV-B or ambient levels supplemented with the UV-B from lamps, similar to the soybean study above. After only the first year of UV-B exposure, reductions were observed in biomass of seedlings derived from several geographic areas. By the end of the third year, these biomass reductions were several-fold larger. These overall growth reductions were generally associated with small decreases in both roots and shoots, but not necessarily accompanied by reductions in photosynthesis. This may be due to changes in needle growth or shifts of allocation as has been found for some crop species. Therefore, these results suggest that the UV-B effects may be accumulative in long-lived plants such as trees, and that even small changes in UV-B radiation might have significant effects over the life time of the trees. Even in the absence of direct UV-B effects on tree biomass, UV-B radiation may still have ecological implications. Changes in plant architecture or biomass allocation could result in alterations in tree seedling competition, ultimately affecting patterns of forest succession.
Although absolute UV-B irradiance is naturally very low in subarctic and Arctic ecosystems such as tundra, there is experimental evidence that the plants in such a system react to increases in UV-B associated with realistic levels of ozone depletion. In a recent field study in northern Sweden, natural dwarf shrub vegetation containing two evergreen species (Empetrum hermaphroditum and Vaccinium vitis idæa) and two deciduous species (V. myrtillus and V. uliginosum) was exposed to artificially enhanced UV-B radiation. Leaf thickness was increased (V. vitis idæa) or decreased (the deciduous species) while stem growth over a two-year period was retarded more in evergreen species than in deciduous species. This suggests that a UV-B increase over an extended time could result in species composition changes [Johanson et al. In press]. Not only growth inhibitions, but also species-specific morphological changes have been observed, which, with time, may result in altered community composition.
Trifluoroacetic acid (TFA) is a breakdown product of HFC134a, HCFC123 and 124 [WMO/UNEP 1994] that is anticipated to achieve a final average concentration less than one thousandth of that required for toxicity of plants, making general impacts on the earth's vegetation by TFA unlikely. Substantially higher concentrations of TFA may occur in areas with low precipitation since the same quantity of TFA would be concentrated in less rainfall. Toxic concentrations in soils could conceivably build up in areas with high evaporation rate or lacking runoff, where salt stress excludes vegetation except for very specialized species. Although plants in such areas are of minor importance both economically and for global primary production, some are essential for migratory birds and other wildlife. Further investigations should therefore be focused on such areas and species.
Mechanisms of UV-B action on plant systems are reasonably well understood when compared with our ability to assess potential consequences of enhanced UV-B at the level of ecosystems. As global change involves not only increased solar UV-B, but also increased atmospheric CO2 concentrations and temperature changes, realistic assessments of the effect of stratospheric ozone reduction need to consider interacting factors. Effects of enhanced UV-B on terrestrial ecosystems are anticipated, both in agriculture, and in nonagricultural areas such as forests, tundra, etc., but prediction of exact consequences, and sometimes even the direction of these changes, is not currently possible.