Past UV-B flux from fossil pollen: prospects for climate, environment and evolution


A considerable amount of experimental work has been devoted to investigating the effects of UV-B radiation on organisms and ecosystems, but very little is known about how UV-B fluxes have changed through time, despite this being an issue of great interest to a wide range of scientists, including geophysicists, climatologists, ecologists and biologists, and their ‘palaeo’ colleagues. In this issue of New Phytologist, Willis et al. (pp. 553–560) present the first quantitative reconstruction of UV-B flux over thousands of years, opening up new avenues of research on climate, environment and evolution.

‘This research has the potential to reveal new aspects of the evolutionary history of organisms.’

In the last 10 yr, a number of pioneering papers have established the technical basis for tracking UV-B flux through time by determining the concentration of UV-B-absorbing compounds in the walls of spores and pollen grains (Rozema et al., 2001; Blokker et al., 2005; Lomax et al., 2008), as well as in cuticles, seed coats and wood (Rozema et al., 2009). These absorbing compounds (e.g. p-coumaric acid) are chemically relatively stable and may be preserved under favourable environmental conditions. Building on these foundations, the work by Willis et al. achieves two main goals: to confirm the positive relationship between UV-B-absorbing compounds in modern pollen and the geographical variation of UV-B radiation; and to infer variations of UV-B flux from measurements of p-coumaric concentrations in Pinus spp. pollen at a site in Norway over the past 9500 yr. This reconstruction is preliminary, as the dose–response relationship between solar UV-B and the content of UV-B-absorbing compounds in plants is still to be precisely determined (Rozema et al., 2009). However, the analyses carried out on modern pollen by Willis et al. support the expectation that future research on extant material will answer the questions of how much the chemical nature of sporopollenin differs among various plant groups, and which other factors may affect the content of UV-B-absorbing compounds in pollen. Calibration on modern conditions may not be fully valid for pre-instrumental times, as with any fossil proxy, because complex climatic and environmental parameters may have changed the response of UV-B-absorbing compounds and their conservation over long time-scales. However, the significant correspondence of the UV-B variations reconstructed by Willis et al. with modelled solar flux over the last 9500 yr demonstrates the considerable potential of this field of research, and suggests that the relationship between UV-B-absorbing compounds and surface UV-B radiation found in modern samples may be reasonably extended back in time, at least to the time-scale of the whole postglacial period.

At this stage of research, when the first step has been taken and matters of controversy have not yet come to light, it is the right time to speculate about future prospects. Three main directions of research can be envisaged, answering questions about climate, environment and evolution, respectively.


Reconstructions of past UV-B fluxes may be an important tool with which to infer variations of ozone in the stratosphere and their effects on climate change. UV-B radiation depends on changes in solar activity (in addition to changes in the Earth’s orbital configuration, cloud cover, volcanic activity and aerosols). Increased solar activity leads to increased UV-C radiation in the upper atmosphere, which in turn enhances the photochemical formation of ozone and hence the absorption of UV-B radiation. The relationships between ozone and climate change are very complex, as they can work in both directions: changes in ozone can induce changes in climate, and vice versa (McKenzie et al., 2011). Investigation of the nature of these relationships in the past may lead to a better understanding of the physical interactions among solar activity, ozone and climate. Past solar variability has been reconstructed on time-scales of centuries to millennia using cosmogenic radionuclides, such as 10Be, stored in terrestrial archives. Past climate changes have been reconstructed using various proxy data, for example, tree rings, fossil pollen, ocean sediments, coral and historical data. Using both solar and climate proxies, possible connections of solar activity with changes in climate variables have been found, including the location and intensity of the InterTropical Convergence Zone (ITCZ), periods of mid-continental droughts, ocean currents, cloud formation, and monsoon strength (Gray et al., 2010). The paper by Willis et al. opens up the intriguing possibility of comparing solar and climate proxies with biological proxy data for ozone-modulated UV-B. UV-B reconstructions at a time-scale of 10–103 yr seem adequate for this line of research, allowing analyses of the effects of ozone changes during time intervals for which the variations in solar irradiance may be reconstructed using independent methods. This may also improve our knowledge of the relative importance of human activity in ongoing climate change. In this respect, assessing modes and rates of UV-B flux variations through time, including their possible periodical oscillations, is of great interest to policy makers and may have important societal effects.


Changes in solar UV-B radiation have many different direct and indirect effects on terrestrial and aquatic ecosystems, and on global biogeochemistry, with implications for the cycling of carbon, nitrogen and other elements (Ballaréet al., 2011). Complex conceptual models have been developed to explain the potential effects of enhanced UV radiation on biogeochemical cycles, but, as the nature of many of these interactions is that they act over medium to long time-scales, understanding of the consequences remains limited (Zepp et al., 2007). Experimental data and meta-analyses of published information suggest that increases in UV-B radiation may have inhibitory effects on biomass accumulation. However, consequences for plant growth are generally modest, as plants acclimate to changes in UV-B radiation through several defence responses (Ballaréet al., 2011). In any case, the time needed to verify the possible effects of UV-B changes on natural ecosystems may be much longer than a human life-time and may require centennial/millennial records of past vegetation. Thus, comparisons of past UV-B flux variations with terrestrial ecosystem dynamics as reconstructed using pollen analysis may prove very useful. Long-term observations may also clarify whether variations in UV-B radiation resulting from changes in climate and land use may have more important consequences for terrestrial ecosystems than the increases in UV-B caused by ozone depletion (Ballaréet al., 2011). Considering that the effects of UV-B radiation on terrestrial and aquatic ecosystems are expected to vary significantly between regions, comparison of pollen records and reconstructed UV-B fluxes at different latitudes may help to determine the regional response of natural terrestrial ecosystems to variations in UV-B radiation through time. Time-scales of 102–104 yr and geographical scales including continental/regional distributions of ecosystems/communities/taxa, for which a wealth of palaeoecological data is already available, may be appropriate to study the effects of changes in UV-B radiation on terrestrial ecosystems.


Changes in UV-B may induce mutations in plants and may be expected to influence all aspects of plant genetics up to and including speciation rates, eventually affecting the mode and tempo of evolution (Willis et al., 2009). Hypotheses regarding the possible links between changing UV-B concentrations caused by catastrophic ozone depletions and mass extinctions have been advanced by a number of authors (Bjorn & Mckenzie, 2007), for example in relation to the world-wide ecological crisis at the end of the Permian (Visscher et al., 2004). Conversely, other authors have suggested increased speciation rates during the Eocene and early Miocene in environments with high UV-B values (Willis et al., 2009; Flenley, 2011). These hypotheses are very appealing, but there is not yet any possibility of testing them. If UV-B-absorbing compounds can survive unaltered for millions of years in the wall of pollen grains and other plant fragments (Rozema et al., 2009), we can expect that future research on UV-B-absorbing compounds in the fossil record will elucidate whether causal relationships exist between changes in UV-B flux and extinction, origination and turnover rates of plant and animal species. This research, dealing with time-scales of 104–107 yr, has the potential to reveal new aspects of the evolutionary history of organisms.

Conclusive remarks

In summary, future reconstructions of past variations of UV-B flux may be of great interest across disciplines. Development of this new field of research at different scales, from very detailed analyses of recent material to analyses of long-term variations over millions of years, and from local to continental geographical scales, is made possible by the versatility of palynological research, which takes advantage of considerable experience in palaeoenvironmental reconstructions at different spatial and temporal scales. At the same time, the study of UV-B-absorbing compounds from fossil pollen may open new frontiers in palynological research. Recent scientific developments show that traditional pollen analysis, based on pollen morphology, may be profitably supplemented with new genetic, isotopic and biochemical techniques, allowing determination of the ancient DNA of tree species (Magyari et al., 2011), reconstruction of the abundance of C4 grasses in past landscapes (Nelson et al., 2008), and assessment of past UV-B fluxes in the environment (Willis et al.). Some time is needed for these techniques to be properly calibrated and validated, but the seed has been planted and will hopefully bear fruit.