Mapping UV radiation in the East Antarctic sea ice zone

Ozone depletion has been a well-researched phenomenon since its discovery in the early 1980s. It is now scientifically accepted that if all other factors (cloudiness, turbidity, surface albedo, etc.) remain the same, decreases in stratospheric ozone concentrations provoke an increase in ultraviolet B (UVB) radiation (in the wavelength range 280-325 nm) at the Earth’s surface. Experimental evidence also shows that exposure to UVB radiation can decrease algal productivity, and cause damage to various forms of aquatic larvae and other organisms. Biologically-effective levels of solar UV radiation penetrate water columns well, especially in the clear waters found in the Antarctic pack ice zone. To determine the relationship (if any) between increased biologically-weighted UV doses and damage to various Antarctic organisms, scientists need to know the magnitude of these increases, not only at the surface but throughout the ice and the water column. This research initially produced surface maps of levels of cloudy-sky erythemal UV, derived from satellite data, for the East Antarctic marginal ice zone (split into four regions for ease of presentation) for the years 1996-2005. Case studies of the relationships between levels of cloudy-sky erythemal UV (at the surface of the sea ice and ocean) and ozone, sea ice concentration and the cloud modification factor are presented by way of various analyses for a selection of sub-areas within the four regions. These analyses show that the levels of erythemal UV in the sub-areas are strongly affected by levels of ozone and by sea ice concentration. The results also indicate that the Shackleton and Amery regions are more affected by lower levels of ozone associated with the Antarctic ozone hole than the Mertz and Dalton regions. Trend analyses for the nine-year study period indicated significant trends for erythemal UV, sea ice concentration and ozone for some of the selected sub-areas. The austral spring of 2001 stood out as a period of unusually high erythemal UV values. To create maps of cloudy-sky erythemal UV radiation, the radiative transfer model UVSPEC was employed to develop a look-up table (LUT) of levels of erythemal UV radiation for various combinations of ozone, solar zenith angle and surface UV albedo. IDL software was then written to estimate a value of clear-sky erythemal UV at the Earth’s surface, using the erythemal UV LUT, for each 5 x 5 km pixel of an Advanced Very High Resolution Radiometer (AVHRR) Polar Pathfinder image for each day that AVHRR irradiance and TOMS ozone data were available for the temporal coverage of the project. The main confounding issue in the determination of erythemal UV radiation at the Earth’s surface is the cloud cover, as this can vary not just on a daily basis but also throughout a day. The amount of cloud in a pixel was taken into account by multiplying clear-sky erythemal UV by a cloud modification factor based on the classification of AVHRR images using an expert system. A subsequent undertaking used the ASPeCt sea ice and snow thickness dataset to produce maps of erythemal UV values at predetermined depths within the sea ice and the water column.Field work completed during the SIPEX voyage to the East Antarctic sea ice zone in 2007 allowed the estimation of UVB attenuation coefficients for ice and snow – these values were used, along with a known UVB attenuation coefficient for water, when determining sub-surface erythemal UV values. Initially, the erythemal UV values were calculated at 5, 10, 15 and 20 m in the water column. To take into consideration the fact that phytoplankton do not necessarily remain at a specific depth, average erythemal UV values over the upper 5, 10, 15 and 20 m of the water column were also calculated. Erythemal UV values at the top of the sea ice (beneath the snow cover), mid-way through the ice, at the bottom of the ice and at a fixed depth of 20 cm within the ice were calculated specifically for sea ice algae research. The erythemal UV values at the bottom of the ice were used to further calculate erythemal UV in the water column at the re-established depths beneath the ice (taking into account the draft of the sea ice). The sea ice concentration was incorporated into the calculations to determine pixel-averaged values of erythemal UV at 5, 10, 15 and 20 m depth. A final step was to use spectral biological weighting functions to develop a set of non-linear conversions between erythemal UV dose rate and more biologically-relevant dose rates. Including these aspects in the model allowed the erythemal UV values in the water column to be converted to a phytoplankton-specific dose rate, and the estimated erythemal UV values within a slab of sea ice to be converted to a sea-ice-algae-specific dose rate. The logistical problems of undertaking in situ research in Antarctica require alternative methods such as satellite remote sensing and radiative transfer modelling. This project has used satellite data and a radiative transfer model (RTM) to create maps of erythemal UV values for the study area. Further research will be required to assess (either from in situ measurements of primary production or with the use of remote sensing of ocean colour) how significant the effects of increased levels of erythemal UV are on the primary production of the Southern Ocean