Increased exposure of Southern Ocean ... - Semantic Scholar

GEOPHYSICAL RESEARCH LETTERS, VOL. 31, L09304, doi:10.1029/2004GL019633, 2004

Increased exposure of Southern Ocean phytoplankton to ultraviolet radiation Dan Lubin Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California, USA

Kevin R. Arrigo and Gert L. van Dijken Department of Geophysics, Stanford University, Stanford, California, USA Received 4 March 2004; accepted 9 April 2004; published 4 May 2004.

[ 1 ] Satellite remote sensing of both surface solar ultraviolet radiation (UVR) and chlorophyll over two decades shows that biologically significant ultraviolet radiation increases began to occur over the Southern Ocean three years before the ozone ‘‘hole’’ was discovered. Beginning in October 1983, the most frequent occurrences of enhanced UVR over phytoplankton-rich waters occurred in the Weddell Sea and Indian Ocean sectors of the Southern Ocean, impacting 60% of the surface biomass by the late 1990s. These results suggest two reasons why more serious impacts to the base of the marine food web may not have been detected by field experiments: (1) the onset of UVR increases several years before dedicated field work began may have impacted the most sensitive organisms long before such damage could be detected, and (2) most biological field work has so far not taken place in Antarctic waters most extensively subjected INDEX TERMS: 9310 Information Related to enhanced UVR. to Geographic Region: Antarctica; 0360 Atmospheric Composition and Structure: Transmission and scattering of radiation; 3349 Meteorology and Atmospheric Dynamics: Polar meteorology; 3360 Meteorology and Atmospheric Dynamics: Remote sensing; 4853 Oceanography: Biological and Chemical: Photosynthesis. Citation: Lubin, D., K. R. Arrigo, and G. L. van Dijken (2004), Increased exposure of Southern Ocean phytoplankton to ultraviolet radiation, Geophys. Res. Lett., 31, L09304, doi:10.1029/2004GL019633.

1. Introduction [2] The discovery of the springtime Antarctic ozone decrease [Farman et al., 1985], and the related measurement of substantially enhanced UVR irradiances at the Antarctic earth surface [Lubin et al., 1989], has motivated field research on UVR-induced damage in a variety of Antarctic marine organisms [Karentz, 1991]. Such experiments have quantified the inhibition of photosynthesis in many species of Southern Ocean phytoplankton, and have suggested that an adverse impact to the base of the marine food web is detectable [Smith et al., 1992]. Because of Antarctica’s extreme remoteness, remote sensing must be used to extrapolate these physiological results to the entire Southern Ocean. In the first use of satellite data for this purpose [Arrigo et al., 2003], it was seen that an extensive adverse response of Southern Ocean phytoplankton under a deep ozone ‘‘hole’’ is not automatic, because (1) UVR-induced photoinhibition Copyright 2004 by the American Geophysical Union. 0094-8276/04/2004GL019633$05.00

is confined largely to the upper few meters of the water column, and (2) the largest UVR enhancements are often over the continent. In this study, the 20 year time series of both NASA Total Ozone Mapping Spectrometer (TOMS) and passive microwave (MW) sea ice data [Cavalieri et al., 1999] are used to demonstrate that enhanced UVR irradiances have increased in frequency and geographic extent over the Southern Ocean, and that these enhanced UVR irradiances intersect with an increasingly large fraction of the phytoplankton population.

2. Satellite Data and Analysis [3] Spectral UVR at the Southern Ocean (below 60
S) surface is retrieved by a radiative transfer method described by Arrigo et al. [2003], but which includes a more robust parameterization for sea ice albedo as a function of sea ice concentration [Lubin and Morrow, 2001], such that surface UVR (280– 400 nm) and photosynthetically active radiation (400– 800 nm) can be retrieved over any ice concentration. Examples of wavelength-integrated UV-B (280 – 315 nm) are shown in Figure 1, for 05 October of 1979 and 1992. The austral spring of 1979 can be considered a control season, in that the anthropogenic ozone depletion had only recently commenced and was largely confined over the Antarctic continent. Compared to 1979, we see that the same date in 1992 exhibited UV-B enhancements of a factor of two or more over the Bellingshausen Sea, Antarctic Peninsula, and Indian Ocean sector of the Southern Ocean. Since 1997, NASA’s Sea-viewing Wide Field-of-view Sensor (SeaWiFS) [McClain et al., 1998] has collected enough Southern Ocean data that a complete climatology of surface chlorophyll a distribution has been derived for both spring (SON) and summer (DJF). This study evaluates the daily intersection of enhanced UVR with these climatological surface phytoplankton distributions. [4] Figure 2a compares satellite-based UVR retrievals over Palmer Station with surface measurements [Weiler and Penhale, 1994]; the response of surface UVR to daily variations in both ozone and cloud opacity is captured by the satellite method. We need to choose a meaningful definition of enhanced UVR in the context of anthropogenic ozone depletion. First, we must choose a biological weighting function (BWF) that relates the spectral UVR irradiance to the relevant physiological process (i.e., photoinhibition in phytoplankton); here we use the BWF reported by Neale et al. [1998a] to calculate the photoinhibition dose rate F from the satellite retrievals of spectral downwelling surface irradiance [Madronich, 1993]. For Antarctic solar eleva-

L09304

1 of 4

L09304

LUBIN ET AL.: ANTARCTIC UV RADIATION AND PHYTOPLANKTON

L09304

For Palmer Station, these two climatological dose rates are shown in Figure 2b, along with the actual satellite-derived dose rate. We see that F > Fclr on 54 days and F > Fc on 75 days. Some of these instances occur after a normal ozone column has returned above the station (day 298, Figure 2c).

Figure 1. Examples of satellite-retrieved UVR and chlorophyll a over the Southern Ocean; top: instantaneous local noon UV-B irradiance reaching the ocean surface on 05 October of 1979 (top) and 1992 (middle); bottom: seasonally averaged SeaWiFS surface chlorophyll a abundance for spring. tions, this BWF yields a radiation amplification factor in the range 0.26 –0.54. Second, we must consider the large daily and interannual variability in both cloud opacity and sea ice concentration. For a given location, one can define a ‘‘climatological clear sky’’ dose rate, Fclr, computed theoretically from the 1979 ozone abundance and sea ice concentration averaged over the 20 year data set. One can also define a ‘‘climatological’’ dose rate under typical cloud cover, Fc, computed from the 1979 ozone and both the average sea ice concentration and average cloud opacity.

Figure 2. Satellite retrievals from Palmer Station, Antarctica, during spring 1992, illustrating the steps involved with determining if the dose rate should be considered enhanced due to ozone depletion. (a) comparison of the satellitederived spectral irradiance at 305 nm with NSF UV Monitor measurements. (b) Time series of the satellite-derived UVR dose rate for photoinhibition in Antarctic phytoplankton (magenta), the dose rate under climatological conditions Fc (1979 ozone and interannually averaged cloud effective optical depth and sea ice cover, dashed curve), the threshold for exceeding the climatological dose rate with a 95% confidence level, Fc + 1.96s (solid black curve), and the dose rate Fclr that would prevail under 1979 ozone, clear skies, and interannually averaged sea ice cover (blue). (c) TOMS total column ozone for 1979 and 1992. (d) MW total sea ice concentration for 1992, and averaged over all available years.

2 of 4

L09304

LUBIN ET AL.: ANTARCTIC UV RADIATION AND PHYTOPLANKTON

L09304

sistently exceeded 10% during September. Beginning in 1982, this biomass fraction under enhanced UVR consistently exceeded 10% during October. Between 1982 and 1994, this biomass fraction for November varied between 3– 17%, depending on the duration and depth of the ozone hole. By the late 1990s, the ozone hole lasted long enough that impacted monthly average biomass fractions greater than 10% were observed during December. Furthermore, although the areal extent of the ozone hole has increased by only 10% since 1994, the fraction of Southern Ocean

Figure 3. Monthly averaged fraction of the Southern Ocean surface biomass lying under enhanced UVR dose rates for photoinhibition in phytoplankton. [5] For a more conservative definition of enhanced UVR, we can label a given UVR dose rate enhanced if it exceeds the climatological dose rate by 1.96 sF (for a 95% confidence level), where sF is the standard deviation in the dose rate resulting from interannual variability in cloud cover and sea ice, with ozone held constant [Lubin and Jensen, 1995]. In Figure 2b, we see that F > Fc + 1.96sF on only 17 days and never after the ozone column has returned to normal. With this definition, enhanced UVR dose rates result primarily from the ozone depletion (Figure 2c). Examining Figure 2d, we see that F < Fc on most days in November and December, due to the 1992 sea ice concentration being lower than the climatological value. The presence of sea ice brings about multiple reflection effects between the high albedo surface and the atmosphere/cloud base that enhance UVR both over the ice and over the adjacent open water [Gardiner, 1987]. After day 300, the occasional instances of F > Fc result from cloud free or optically thinner cloud cover relative to climatological values (Figure 2a). Based on these radiative transfer considerations, we adopt this more conservative criterion for labeling a UVR dose rate as enhanced. [6] For each grid cell, the standard deviation of the wavelength-integrated UV-A irradiances (315 – 400 nm) was computed from all the years’ retrievals, to estimate the standard deviation sF that results from interannual variability in cloud cover and sea ice (mostly independent of ozone variability). To evaluate the biomass fraction under enhanced UVR, if F > Fc + 1.96sF in a grid cell, the average surface-layer chlorophyll a abundance (mg m 3) in the grid cell is computed from seasonally averaged SeaWiFS retrievals, weighted by the grid cell area (km2), and added to the sum either over the entire Southern Ocean (Figure 3) or over a given longitude bin (Figure 4).

3. Results [7] When we examine the time evolution of the total Southern Ocean surface biomass under enhanced UVR (Figure 3), we notice that a considerable biomass fraction was impacted early in the history of the ozone hole (>5% of biomass during spring by 1980), even with our conservative definition of enhanced UVR. By 1987, the monthly-averaged surface biomass fraction under enhanced UVR con-

Figure 4. Meridional breakdown of the October monthly average surface biomass fraction lying under enhanced UVR dose rates for photoinhibition in phytoplankton. Bins are of width 20
in longitude (10
for two bins near the Antarctic Peninsula), bounded in the north by 60
S and in the south by the Antarctic coastline or permanent ice shelf edge.

3 of 4

L09304

LUBIN ET AL.: ANTARCTIC UV RADIATION AND PHYTOPLANKTON

phytoplankton exposed to enhanced UVR has more than doubled over that same time period. This is ultimately due to polar vortex dynamics, which have consistently placed increasingly severe ozone depletions over the productive Weddell Sea region and over the Indian Ocean sector that frequently extend as far north as 60oS. During recent ozone hole events, more than 30% of the average monthly phytoplankton biomass fraction was exposed to increased UVR. [8] There is a strong meridional variability in these UVR enhancements (Figure 4). In the years prior to the discovery of the Antarctic ozone hole, enhanced UVR during October began to intersect more than 10% of the surface biomass in the Weddell Sea, Queen Maud Land, and Indian Ocean sectors of the Southern Ocean. In the years immediately following discovery, during which intensive biological field work began, the bias in enhanced UVR toward these regions continued but the Ross Sea sector also showed substantially impacted biomass fractions (>20%) during 1987 and 1990. During the early and mid-1990s, the dynamics of the polar vortex consistently resulted in the largest impacted biomass fractions occurring throughout the Indian Ocean sector as far east as the Amery Ice Shelf. By the late 1990s, the ozone hole had increased in size and depletion such that monthly-averaged surface biomass fractions under enhanced UVR consistently exceeded 10% in most sectors, with the largest increases (40 – 60%) occurring in the Weddell Sea and Indian Ocean sectors due to the consistent geographic bias in polar vortex dynamics noted above. A secondary maximum in impacted biomass fractions (reaching 30%) occurred in the Ross Sea.

4. Discussion [9] These results have implications for understanding scientific progress to date and for planning future field work. To date there has been no direct detection of an overall decrease in primary production throughout a given Antarctic region that can be directly attributable to the ozone ‘‘hole.’’ The closest result to such a detection was obtained by the ICECOLORS experiment [Smith et al., 1992], which was fortunate to carry out physiological water column experiments as the stratospheric ozone above the research stations alternated between depleted and climatologically normal abundances in the Western Antarctic Peninsula (WAP) region. However, Figure 4 shows that the WAP region (65
W) is not a sector most strongly impacted by extensive enhanced UVR. To date, most ecological research related to ozone depletion has taken place in the WAP region and Ross Sea, due to the proximity of major research bases and scheduling of research vessels. Our satellite retrievals suggest that field experiments carried out in the Weddell Sea [Neale et al., 1998b], Indian Ocean sector, and the vicinity of the Amery Ice Shelf, might have a better chance at directly detecting major ecological impacts. Future work should also consider the surface and in-water irradiances at finer spatial resolutions than our large-scale retrievals (i.e., 50 km herein), as these details affect the life-cycle of phytoplankton near the sea ice edge. [10] There is also the issue of a missing baseline in physiological understanding. Once the ozone hole was discovered, field programs of the late 1980s quantified biological factors such as UVR-induced inhibition of pho-

L09304

tosynthesis, cellular repair mechanisms, and protective pigmentation in marine organisms [Weiler and Penhale, 1994]. To fully understand the potential for ecological damage, one must have a priori understanding of these factors throughout a representative sample of species that comprise the base of the Antarctic marine food web. This understanding was not garnered until after the ozone hole was discovered, which, as the satellite data reveal, occurred several years after large fractions of the Southern Ocean biomass were subjected to enhanced UVR. Thus it is possible that some of the more highly UVR-sensitive species or components of the Antarctic marine ecosystem may have suffered damage that may never be known [Karentz, 1991]. Another possibility is that the Southern Ocean ecosystem is actually highly resilient to enhanced UVR, and no damage is detectable. The Antarctic ozone hole therefore provides not only a lesson in how human activity can have a major impact on the atmosphere, but also underscores the need to survey major marine ecosystems as completely as possible so that anthropogenic impacts can be properly assessed. [11] Acknowledgments. This research was supported by the NASA Atmospheric Chemistry Modeling and Analysis (ACMAP) and the Goddard Earth Science and Technology (GEST) programs.

References Arrigo, K. R., D. Lubin, G. L. van Dijken, O. Holm-Hansen, and E. Morrow (2003), Impact of a deep ozone hole on Southern Ocean primary production, J. Geophys. Res., 108(C5), 3154, doi:10.1029/2001JC001226. Cavalieri, D., et al. (1999), Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data [CD-ROM], Natl. Snow and Ice Data Cent., Boulder, Colo. Farman, J. C., B. G. Gardiner, and J. D. Shanklin (1985), Large losses of total ozone over Antarctica reveal seasonal ClO(x)/NO(x) interaction, Nature, 315, 207 – 210. Gardiner, B. G. (1987), Solar radiation transmitted to the ground through cloud in relation to surface albedo, J. Geophys. Res., 92, 4010 – 4018. Karentz, D. (1991), Ecological considerations of Antarctic ozone depletion, Antarct. Sci., 3, 3 – 11. Lubin, D., and E. H. Jensen (1995), Effects of clouds and stratospheric ozone depletion on ultraviolet radiation trends, Nature, 377, 710 – 713. Lubin, D., and E. Morrow (2001), Ultraviolet radiation environment of Antarctica 1: Effect of sea ice on top-of-atmosphere albedo and on satellite retrievals, J. Geophys. Res., 106, 33,453 – 33,461. Lubin, D., J. E. Frederick, C. R. Booth, T. Lucas, and D. Neuschuler (1989), Measurements of enhanced springtime ultraviolet radiation at Palmer Station, Antarctica, Geophys. Res. Lett., 16, 783 – 785. Madronich, S. (1993), UV radiation in the natural and perturbed atmosphere, in Environmental Effects of Ultraviolet Radiation, edited by M. Tevini, pp. 17 – 69, A. F. Lewis, New York. McClain, C. R., M. L. Cleave, G. C. Feldman, W. W. Gregg, S. B. Hooker, and N. Kuring (1998), Science quality SeaWiFS data for global biosphere research, Sea Technol., 39, 10 – 16. Neale, P. J., R. F. Davis, and J. J. Cullen (1998a), Interactive effects of ozone depletion and vertical mixing on photosynthesis of Antarctic phytoplankton, Nature, 392, 585 – 589. Neale, P. J., J. J. Cullen, and R. F. Davis (1998b), Inhibition of marine photosynthesis by ultraviolet radiation: Variable sensitivity of phytoplankton in the Weddell-Scotia Confluence during the austral spring, Limnol. Oceanogr., 43, 433 – 448. Smith, R. C., et al. (1992), Ozone depletion: Ultraviolet radiation and phytoplankton biology in Antarctic waters, Science, 255, 952 – 959. Weiler, C. S., and P. A. Penhale (Eds.) (1994), Ultraviolet Radiation in Antarctica: Measurement and Biological Effects, Antarct. Res. Ser., vol. 62, 257 pp., AGU, Washington, D. C.

K. R. Arrigo and G. L. van Dijken, Department of Geophysics, Stanford University, Stanford, CA 94305, USA. D. Lubin, Scripps Institution of Oceanography, UCSD, La Jolla, CA 92093, USA. ([email protected])

4 of 4