International audienceOzone profile trends over the period 2000 to 2016 from several merged satellite ozone data sets and from ground-based data by four techniques at stations of the Network for the Detection of Atmospheric Composition Change indicate significant ozone increases in the upper stratosphere, between 35 and 48 km altitude (5 and 1 hPa). Near 2 hPa (42 km), ozone has been increasing by about 1.5 % per decade in the tropics (20° S to 20° N), and by 2 to 2.5 % per decade in the 35° to 60° latitude bands of both hemispheres. At levels below 35 km (5 hPa), 2000 to 2016 ozone trends are smaller and not statistically significant. The observed trend profiles are consistent with expectations from chemistry climate model simulations. Using three to four more years of observations and updated data sets, this study confirms positive trends of upper stratospheric ozone already reported, e.g., in the WMO/UNEP Ozone Assessment 2014, or by Harris et al. (2015). The additional years, and the fact that nearly all individual data sets indicate these increases, give enhanced confidence. Nevertheless, a thorough analysis of possible drifts and differences between various data sources is still required, as is a detailed attribution of the observed increases to declining ozone depleting substances and to stratospheric cooling. Ongoing quality observations from multiple independent platforms are key for verifying that recovery of the ozone layer continues as expected

Simultaneous measurements of upwelling and downwelling ultraviolet (UV) spectral irradiance were used to deduce the wavelength dependence and the solar zenith angle dependence of the albedo of the ground surface (long grass) at Lauder, New Zealand (45.05°S, 169.68°E). In the UVB region the deduced albedos are approximately 1%. At longer wavelengths the albedo increases to approximately 2% at 400 nm, and 4.5% at 450 nm. These albedos are significantly smaller than those generally quoted for the visible region. The deduced albedos tend to increase at larger solar zenith angles, demonstrating that the surface is not strictly Lambertian.

Five UV spectroradiometers representative of the types used in monitoring programs on several continents were intercompared at the Fraunhofer Institute for Atmospheric Environmental Research (IFU), Garmisch‐Partenkirchen, Southern Germany, during a campaign in August 1994. Global spectral irradiances between 290 and 410 nm were measured over a range of solar zenith angles from 30° to 80°. Scans were synchronized to enable useful comparisons to be made under changing weather conditions, which included overcast, partly cloudy, and clear skies. No exchange of data was allowed between participating groups until after the campaign. At wavelengths longer than 310 nm, the spectra generally agreed to within ±5%. At wavelengths shorter than 310 nm, differences between instruments were larger, especially at larger solar zenith angles. Causes of differences are discussed. For all instruments, deviations in erythemally weighted irradiances were always less than 7% from the mean. The agreement between measurement systems is sufficient to allow an investigation of geographical differences in UV, under all observing conditions. UV doses measured at sites in the southern hemisphere are systematically larger than those measured at the corresponding northern latitudes. During the summer months the daily doses at the South Pole exceed those at mid‐latitudes in the Northern Hemisphere. Further investigations must be performed to establish a global UV‐climatology.

Abstract. The purpose of this work is determine the extent of closure between measurements and models of UV irradiances at diverse sites using state of the art instruments, models, and the best available data as inputs to the models. These include information about aerosol optical depth (unfortunately not extending down as far into the UVB region as desirable because such information is not generally available), ozone column amounts, as well as vertical profiles of temperature. We concentrate on clear-sky irradiances, and report the results in terms of UV Index (UVI).Clear-sky data from one year of measurements at each of four diverse sites (Lauder -New Zealand, Mauna Loa Observatory -Hawaii, Boulder -Colorado, and Melbourne -Australia) have been analysed in detail, also taking account of different measurements of ozone, including satellite-derived values, as well as ground measured values, both from Dobson instruments and as retrieved from the UV spectra under study. Previous studies have generally focussed on data from a single site, and for shorter periods. As such, it is the most comprehensive study of its kind to date.At Lauder, which is the cleanest low altitude site, we obtained agreement between measurement and model at 5% level, which is consistent with the best agreement found previously. At Mauna Loa Observatory, similar agreement was achieved, but model calculations need to allow for reflections from cloud that are present below the observatory. At this site, there are occasional problems with using satellitederived ozone. At Boulder, mean agreements were similar but the dispersion around the mean was slightly larger, corresponding to larger uncertainties in the aerosol inputs to the model. However, at Melbourne, which is the only nonCorrespondence to: J. Badosa (jordi.badosa@gmail.com) NDACC (Network for the Detection of Atmospheric Composition Change) site, there remain unexplained discrepancies.The measured values are significantly lower than the calculated values. We investigate the extent to which this discrepancy can be explained by incomplete knowledge of aerosol extinctions in the UV at this site. We conclude that further information about aerosol optical depth and single scattering albedo in the UVB region is needed to resolve the issues. For more polluted sites (the four considered locations show in general small aerosol load), the uncertainties in the aerosol input parameters would lead to less confidence on the modelling approach. At the three NDACC sites, the closure provided by the study gives confidence in both the measurements and our ability to model them. This study revealed a limitation in the use of PTFE diffusers when temperatures are lower than approximately 20 • C. It also documents the range of clear sky UVI values expected at these diverse sites.

Three UV spectroradiometers from the National Institute of Water and Atmospheric Research (NIWA) New Zealand, the Fraunhofer Institute (IFU) Germany, and the Australian Radiation Laboratory (ARL) Australia were intercompared at Lauder NZ on 23 February 1993. Over the spectral range 290–400 nm, the agreement between the IFU and NIWA instruments was better than 5%. At noon on this day, the irradiances measured by all three instruments agreed within ±10%, except at wavelengths shorter than 300 nm, where the ARL instrument gave higher readings. At larger solar zenith angles (SZA) the differences at short wavelengths were more pronounced, and at wavelengths above 300 nm the ARL measurements were systematically lower. The reasons for these differences are discussed. Having established the differences between the sets of instrumentation, spectra of maximum clear sky UV irradiances observed by these groups in New Zealand, Australia, and Europe are compared. The erythemally weighted irradiance observed in Melbourne Australia was the highest (0.35 W m−2). Respective maxima for Lauder NZ and for Neuherberg Germany were 85% and 66% of that in Australia. Differences are larger for DNA‐weighted UV.