This research examines a 17-year database of UV-A (320–400 nm) and visible (400–600 nm) solar irradiance obtained by a scanning spectroradiometer located at the South Pole. The goal is to define the variability in solar irradiance reaching the polar surface, with emphasis on the influence of cloudiness and on identifying systematic trends and possible links to the solar cycle. To eliminate changes associated with the varying solar elevation, the analysis focuses on data averaged over 30–35 day periods centered on each year's austral summer solstice. The long-term average effect of South Polar clouds is a small attenuation, with the mean measured irradiances being about 5–6% less than the clear-sky values, although at any specific time clouds may reduce or enhance the signal that reaches the sensor. The instantaneous fractional attenuation or enhancement is wavelength dependent, where the percent deviation from the clear-sky irradiance at 400–600 nm is typically 2.5 times that at 320–340 nm. When averaged over the period near each year's summer solstice, significant correlations appear between irradiances at all wavelengths and the solar cycle as measured by the 10.7 cm solar radio flux. An approximate 1.8 ± 1.0% decrease in ground-level irradiance occurs from solar maximum to solar minimum for the wavelength band 320–400 nm. The corresponding decrease for 400–600 nm is 2.4 ± 1.9%. The best-estimate declines appear too large to originate in the sun. If the correlations have a geophysical origin, they suggest a small variation in atmospheric attenuation with the solar cycle over the period of observation, with the greatest attenuation occurring at solar minimum.

The ability to detect subtle trends in upper stratospheric ozone places strict requirements on satellite-based Solar Backscatter Ultraviolet (SBUV) sensors intended for this purpose. Simulation of the long-term change in backscattered radiance indicates trends in the range 1.0–1.6% per decade depending on wavelength for an ozone depletion scenario based on chlorofluorocarbon (CFC) chemistry. The maximum percentage change in the measured quantity, as would be sensed by the Nimbus-7 SBUV experiment and future similar instruments, is roughly a factor of two less than the maximum percentage change in ozone, the quantity of geophysical interest. Furthermore, interannual variations, whose magnitude can be estimated from the SBUV data set, could obscure a CFC-related trend in radiance when viewed over a period less than a decade. Even when such atmospheric noise and possible solar cycle variations are neglected, the detection of an ozone trend of the magnitude predicted for CFC-related chemistry requires a set of radiance measurements that are internally consistent to 1% or better over a time span of a decade.

Oscillator strengths and predissociation linewidths deduced in recent studies predict a dissociation rate for O2 in the Schumann-Runge bands which is significantly larger in the upper stratosphere and lower mesosphere than previously believed. Error bars on molecular parameters required in the cross-section calculation translate into uncertainties in the dissociation rate which are less than ±10% at all altitudes where the Schumann-Runge bands are agronomically significant.

A network of scanning spectroradiometers has acquired a multiyear database of visible solar irradiance, covering wavelengths from 400 to 600 nm, at four sites in the high-latitude Southern Hemisphere, from 55 degrees S to 90 degrees S. Monthly irradiations computed from the hourly measurements reveal the character of the seasonal cycle and illustrate the role of cloudiness as functions of latitude. Near summer solstice, the combined influences of solar elevation and the duration of daylight would produce a monthly irradiation with little latitude dependence under clear skies. However, the attenuation associated with local cloudiness varies geographically, with the greatest effect at the most northern locations, Ushuaia, Argentina and Palmer Station on the Antarctic Peninsula. Near summer solstice, the South Pole experiences the largest monthly irradiation of the sites studied, where relatively clear skies contribute to this result. Scaling factors derived from radiative-transfer calculations combined with the measured 400-600 nm irradiances allow estimating irradiances integrated over the wavelength band 400-700 nm. This produces a climatology of photosynthetically active radiation for each month of the year at each site.

Measurements of solar ultraviolet irradiance obtained by a Brewer spectrophotometer located in Chicago, Illinois, reveal a wavelength‐dependent excess attenuation associated with cloudy skies beyond that expected from the column ozone amount. When each observation is expressed as a fraction of the irradiance that would have existed under clear skies, a quantity called the transmission ratio, values for a wavelength band near 305 nm are generally smaller than those for a band near 345 nm under dense cloud cover. Transmission ratios computed for wavelength bands near 310 nm and 315 nm lie in between those for the above wavelengths. The observations are consistent with absorption of radiation in the interstitial air of a cloud, while the magnitude and wavelength dependence are like those expected from ozone. A radiative transfer calculation shows that ozone amounts typical of the troposphere are accompanied by enhanced absorption when placed in a medium that is optically thick in scattering. This mechanism constitutes a coupling between tropospheric air quality and the attenuation of sunlight provided by cloudy skies in the chemically active ultraviolet portion of the spectrum.

Continuous measurements of solar ultraviolet spectral irradiance exist for four sites in the high latitude Southern Hemisphere, from 55° to 90° South, and span a time period in excess of a decade. This database allows comprehensive analyses of variability in ground-level radiation over a wide range of time scales. The behavior of irradiances within a single month reflects the combined influences of changing solar elevations, cloudiness and ozone amounts. Histograms assembled for corresponding months over the decadal time scale reveal maximum instantaneous irradiances that can vary by a factor of two or more among different years. An important portion of each year encompasses the months from October through December. The annual ozone loss develops during this period, and in some years this is sufficiently dramatic to distort the seasonal cycle in erythemal irradiance expected solely on the basis of solar elevation. Under unperturbed conditions, the largest monthly-integrated irradiances would occur in December at all wavelengths from 290 to 400 nm. However, in years of unusually low springtime ozone amounts the maximum monthly-integrated erythemal irradiances can appear in November.

Surface measurements of total and diffuse UV irradiance at the seven narrowband wavelength channels of the ultraviolet multifilter rotating shadow‐band radiometer (UVMFR) were used to determine total column ozone and aerosol optical depth for two 6‐month periods in 1997 and 1999 at a site in the Blue Ridge Mountains of North Carolina. The retrieved column ozone displayed a seasonal dependence and consistent agreement with the Total Ozone Mapping Spectrometer (TOMS). The mean ratio of retrieved ozone to TOMS ozone was 0.98 with standard deviations of 0.02 and 0.01 for 1997 and 1999, respectively. Aerosol optical depth at 317, 325, 332, and 368 nm was derived for a 6‐month period of 1999. The seasonal trend exhibited is influenced by the persistent summertime haze that occurs in the region. The retrieved aerosol optical depths are used as input in a radiative transfer model to investigate the effect of their realistic values on the calculation of the UV index (UVI) forecasted by the National Weather Service. The percentage change in calculated surface erythemally weighted UV (versus calculations using the standard UVI aerosol inputs) ranges from a 4% increase to a nearly 50% decrease, dependent upon the aerosol optical depth and amount of absorption by aerosols. Based on our measurements, it was found that during the summertime the UV index can deviate by up to −5 index units from the forecast using the standard aerosol inputs.

Ground-based measurements of solar UV spectral irradiance made from Ushuaia, Argentina at latitude 55 degrees S reveal a large degree of variability among corresponding months of different years over the period from September 1990 through April 1998. The magnitude and wavelength dependence of year-to-year changes in monthly spectral UV-B irradiation are consistent with expectations based on the behavior of column ozone and cloudiness. When combined with satellite measurements of column ozone, a regression model fit to the ground-based data set allows estimates of monthly UV-B irradiation over a time frame of two decades, 1978-1998, during several months of the year. Results show a general increase in ground-level irradiation at 305.0 nm from the end of the 1970s to the early 1990s during calendar months from September through December. This is followed by generally smaller irradiances through the middle to late 1990s for all months except November, where the increase continues through the end of the data record. The long-term variability in monthly irradiation over the time period studied is more complicated than can be described by a simple linear trend.

International agreements for the limitation of ozone-depleting substances have already resulted in decreases in concentrations of some of these chemicals in the troposphere. Full compliance and understanding of all factors contributing to ozone depletion are still uncertain; however, reasonable expectations are for a gradual recovery of the ozone layer over the next 50 years. Because of the complexity of the processes involved in ozone depletion, it is crucial to detect not just a decrease in ozone-depleting substances but also a recovery in the ozone layer. The recovery is likely to be detected in some areas sooner than others because of natural variability in ozone concentrations. On the basis of both the magnitude and autocorrelation of the noise from Nimbus 7 Total Ozone Mapping Spectrometer ozone measurements, estimates of the time required to detect a fixed trend in ozone at various locations around the world are presented. Predictions from the Goddard Space Flight Center (GSFC) two-dimensional chemical model are used to estimate the time required to detect predicted trends in different areas of the world. The analysis is based on our current understanding of ozone chemistry, full compliance with the Montreal Protocol and its amendments, and no intervening factors, such as major volcanic eruptions or enhanced stratospheric cooling. The results indicate that recovery of total column ozone is likely to be detected earliest in the Southern Hemisphere near New Zealand, southern Africa, and southern South America and that the range of time expected to detect recovery for most regions of the world is between 15 and 45 years. Should the recovery be slower than predicted by the GSFC model, owing, for instance, to the effect of greenhouse gas emissions, or should measurement sites be perturbed, even longer times would be needed for detection.

This research examines the behavior of ground-level solar UV radiation as measured by eight broadband meters in the continental United States during the period from late 1994 to late 1998, The goal is to define the variability that occurs in UV irradiance over time scales ranging from one to several years. The monthly integrated irradiances, from latitude 32 degreesN to 47 degreesN, contain large annual cycles and latitudinal gradients which depend on season. Seven of the eight sites show a maximum in July, a behavior related to proximity to the summer solstice, with modifications associated with the annual cycle in column ozone, A large interannual variability in monthly integrated irradiance appears over the 4 year period studied. A comparison of corresponding months during different years shows differences in irradiance of 20% or more in one-third of the cases analyzed. When the solar zenith angle (SZA) is held fixed in the range 60-65 degrees, a substantial annual cycle in UV irradiance remains where the maximum monthly mean irradiance is 1.4-1.9 times the minimum, depending on location. Furthermore, the annual cycle at fixed SZA is not in phase with the normal seasonal cycle, Maximum irradiances at fixed SZA tend to occur in the October to December period, while minima cluster in April through July. The annual cycle in ozone, with maximum column values in spring and minima in autumn, explains the general character of the fixed-SZA data, although changes in cloudiness are significant contributors to interannual variability.

Detection of long‐term, linear trends is affected by a number of factors, including the size of trend to be detected, the time span of available data, and the magnitude of variability and autocorrelation of the noise in the data. The number of years of data necessary to detect a trend is strongly dependent on, and increases with, the magnitude of variance (σN2) and autocorrelation coefficient (ϕ) of the noise. For a typical range of values of σN2 and ϕ the number of years of data needed to detect a trend of 5%/decade can vary from ∼10 to >20 years, implying that in choosing sites to detect trends some locations are likely to be more efficient and cost‐effective than others. Additionally, some environmental variables allow for an earlier detection of trends than other variables because of their low variability and autocorrelation. The detection of trends can be confounded when sudden changes occur in the data, such as when an instrument is changed or a volcano erupts. Sudden level shifts in data sets, whether due to artificial sources, such as changes in instrumentation or site location, or natural sources, such as volcanic eruptions or local changes to the environment, can strongly impact the number of years necessary to detect a given trend, increasing the number of years by as much as 50% or more. This paper provides formulae for estimating the number of years necessary to detect trends, along with the estimates of the impact of interventions on trend detection. The uncertainty associated with these estimates is also explored. The results presented are relevant for a variety of practical decisions in managing a monitoring station, such as whether to move an instrument, change monitoring protocols in the middle of a long‐term monitoring program, or try to reduce uncertainty in the measurements by improved calibration techniques. The results are also useful for establishing reasonable expectations for trend detection and can be helpful in selecting sites and environmental variables for the detection of trends. An important implication of these results is that it will take several decades of high‐quality data to detect the trends likely to occur in nature.

In this study we model the effects of continental and urban aerosols and their variation with humidity on the transmission of ultraviolet and visible radiation to the Earth's surface. Normalizing the transmission to that of an aerosol‐free atmosphere, we examine the mechanisms behind two wavelength‐dependent aerosol effects. The first is a dip in the normalized transmission at wavelengths below around 320 nm, which is caused by a coupling between multiple scattering by the aerosol particles and absorption by ozone and by the rapidly increasing absorption coefficient of tropospheric water‐soluble aerosols below 340 nm, based on limited available refractive index data in the UV. The second effect is an increase in normalized transmission with wavelength from 320 nm through the visible, which is caused by the decrease with wavelength in the Mie scattering coefficients of tropospheric water‐soluble, soot, and stratospheric sulfate aerosols. Using our continental aerosol model, at 0% relative humidity we compute aerosol optical depths of 0.72 at 310 nm and 0.35 at 550 nm, which reduce atmospheric transmission by 12.8% at 310 nm and by 7.9% at 550 nm. With our urban aerosol model we compute aerosol optical depths of 1.82 at 310 nm and 0.87 at 550 nm, which reduce transmission by 34.5% at 310 nm and by 21.1% at 550 nm. Absorption by the aerosols is a significant contributor to this, reducing transmission by 20.9% at 310 nm and by 8.6% at 550 nm beyond nonabsorbing aerosols. For average summer humidity conditions our continental aerosol model predicts an increase in optical depth to 1.26 at 310 nm and to 0.65 at 550 nm, leading to a reduction in transmission of 15.2% at 310 nm and 9.7% at 550 nm, as compared with an aerosol‐free atmosphere. For average summer humidity conditions our urban aerosol model predicts an increase in optical depth to 3.22 at 310 nm and to 1.65 at 550 nm, leading to a reduction in transmission of 40.0% at 310 nm and 25.3% at 550 nm, as compared with an aerosol‐free atmosphere. Comparison with ground‐based data indicates that our estimated optical depths are toward the high end of values seen for continental aerosols, being more typical of industrial than rural areas.

A modification to the conventional delta‐Eddington radiative transfer scheme was implemented in order to create a simple and efficient model particularly suitable for handling the highly anisotropic scattering properties exhibited by cloud drops and aerosol particles in the atmosphere. The modification includes separating out the singly scattered radiation and applying the conventional Eddington approximation to the multiply scattered radiation alone. The singly scattered irradiance is computed analytically without approximation, preserving more of the angular dependence in the radiance than conventional two‐stream style solutions. The single‐scattering‐separate delta‐Eddington model was found to handle strong absorption in the stratosphere and the anisotropic scattering within optically thin cloud layers better than both the simple two‐stream and conventional Eddington models, and to produce results similar to a 22‐stream discrete ordinates model.

Ground-based measurements of solar UV irradiance combined with calculations using satellite-based ozone data are able to define the variability in UV sunlight at Palmer Station and McMurdo Station, Antarctica over time scales of years. Special attention focuses on the spring and summer seasons. Satellite data show that the annual ozone loss that occurs during October was greater in 1991-1992 than in 1979-1980, This led to average noontime UVB irradiances computed for clear skies in the latter period that exceeded those in the earlier time by 50-65%, However, a biologically weighted irradiance for suppression of photosynthesis in phytoplankton indigenous to the area near McMurdo Station increased by at most 5% over this period in response to the change in ozone owing to an important contribution from the UVA, At Palmer Station the behavior of ozone and cloudiness can mesh so as to produce the largest noontime UVB irradiances of the year in October as opposed to near summer solstice in December and January. Interannual variability in UVB irradiance during October, the month of the major ozone loss, is larger at Palmer than at McMurdo during the time spanned by ground-based irradiance measurements, being 1990-1994. However, interannual variations in cloudiness were more important than changes in ozone in causing the observed year-to-year variability at Palmer Station, The opposite situation prevailed at McMurdo during October, where interannual variations in ozone were responsible for most of the year-to-year differences in UVB received at the ground.

This study considers the effects of aerosols in cloudy atmospheres on the wavelength dependence of atmospheric transmission in the ultraviolet and visible parts of the solar spectrum. Normalizing the transmission to that of a cloud‐ and aerosol‐free atmosphere, we examine the competing influences of clouds and aerosols on the shape of the transmission function when clouds and aerosols are mixed; while pure water clouds cause the normalized transmission to decrease with wavelength from around 320 nm through the visible, aerosol particles cause the normalized transmission to increase with wavelength from 320 nm through the visible. The results show that when clouds are superimposed on an aerosol profile with the cloud drops and aerosol particles externally mixed, the shape of the normalized transmission spectrum is dominated by the effect of the cloud drops, unless the optical depth of the aerosols begins to approach the optical depth of the cloud. This is the case for an optically thin stratus cloud and an urban aerosol profile. When cloud drops and aerosol particles are internally mixed through coagulation, the shape of the normalized transmission spectrum is again dominated by the effect of the cloud drops, unless there is an unrealistically high volume fraction of strongly absorbing aerosols embedded in the droplets. While measurements of the mass fraction of absorbing aerosols such as soot in cloud and rainwater range from 3.0×10−9 to 6.9×10−6, a soot volume fraction of 1×10−4 is necessary to cause the normalized transmission to increase with wavelength from 320 nm through the visible. The model results are also shown to be consistent with Brewer spectrophotometer irradiance measurements under cloudy and hazy conditions.

Observations of stratospheric NO, NO2, and O3 from the Halogen Occultation Experiment (HALOE) are examined over the southern midlatitudes (35°–45°S) in the summers following the eruption of Mount Pinatubo, from early 1992 through mid‐1995. The tracers HF and CH4, also observed by HALOE, are used to distinguish between the effects of chemistry and those of transport processes on the distribution of total reactive nitrogen (NOy) and ozone. Results show that at 17 mbar (about 28 km), in parcels of similar photochemical age, the abundances of summertime NO + NO2 (NOx.) increased dramatically between 1992 and 1993 and then leveled off in subsequent years. The 1992–1993 increase is coincident in time with a sharp drop in aerosol surface area density as the Pinatubo aerosol cleared the atmosphere and gives evidence of the sensitivity of NOx to sulfate aerosol even at these altitudes, where the cycling of NOy through HNO3 is rapid. Results also show that summertime ozone abundances at 17 mbar declined by nearly 10% between 1992 and 1993 and then, like NOx, remained about constant for the next three summers. The trend in ozone is opposite in sign to that observed at lower altitudes in the aftermath of the Pinatubo eruption and demonstrates the importance of the NOx catalytic cycles in the ozone loss budget above 26 km.

Surface ultraviolet (UV) radiation measurements from the Robertson‐Berger (RB) meter network and existing documentation of these data were examined to determine long‐term variations of UV. RB meter data from 14 sites in the United States were analyzed for trends over the period 1974–1991. A more in‐depth analysis of the RB meter data, including the use of supporting geophysical data, was carried out for four of the locations. Results based on analysis of data from the 14 sites show a significant negative trend of the order of −6% per decade overall, reasonably consistent with annual trends obtained by Scotto et al. [1988] using similar data for the period 1974–1985. However, when allowance is made for mean level shifts in the data for several of the stations around 1979, which may be due to calibration and other instrument‐related problems, the resulting overall trend is found to be of the order of +2% per decade and not statistically significant. An additional trend analysis using only RB meter data since 1979 at the 14 sites is also performed and leads to overall trend results similar to those from the analysis which allows for mean level shifts in the data. The more detailed analysis of data from four of the stations for the period 1979–1991 is performed to investigate the extent to which the trend behavior in the RB meter measurements can be explained by the behavior of other geophysical quantities such as cloudiness and total ozone. In particular, radiative transfer model‐based calculations of ultraviolet irradiance based on satellite data from the total ozone mapping spectrometer are compared with the RB meter measurements to help explain their behavior. Generally, inconsistencies are found between the trend behavior in RB meter measurements and radiative transfer calculations, with the RB data showing substantial downward movement relative to the calculations for three of the four sites. Significant evidence exists to indicate that problems with the network render the existing RB meter measurements unreliable for long‐term trend detection. Different reasonable treatments of the data result in dramatically different trend results. Without further information, the data, by themselves, do not allow for definitive trend analysis results.

Abstract Measurements of the ground-level solar irradiance from Palmer Station, Antarctica, and Ushuaia, Argentina, reveal a systematic wavelength dependence in the attenuation provided by cloudy skies. As wavelength increases from 350 to 600 nm, the measured cloudy-sky irradiance, expressed as a fraction of the clear-sky value, decreases. Results from Ushuaia for a solar zenith angle of 45° show that a cloudy sky that reduces the spectral irradiance at 500 nm to 50% of that for clear skies is accompanied by irradiances at 350 and 600 nm, which are approximately 59% and 49%, respectively, of the clear sky value. A weaker wavelength dependence appears in the data for Palmer Station. The observed behavior can arise from Rayleigh backscattering of sunlight beneath the cloud, followed by reflection of this upwelling radiation from the cloud base back to the ground. This sequence of events is most effective at short wavelengths and leads to cloudy skies providing less overall attenuation as wavelength decreases.

Observations of O3, HCl, NO, and NO2 from the Halogen Occultation Experiment (HALOE) provide a means to investigate chemical change in the lower stratosphere over Antarctica during the first 23 days of October 1992. Two long‐lived species also observed by HALOE, HF and CH4, are used as tracers to identify a series of air parcels having similar Cly and NOy abundances. The set of parcels chosen using tracer analysis show uniformly low O3 mixing ratios, less than 1 ppm on the 480 K surface (about 18–20 km). HCl mixing ratios for those parcels with less than 1 ppm of ozone nearly tripled during the time period, and NO+NO2 abundances rose sharply, by a factor of 6 or 7. These trends in HCl, NO, and NO2 agree qualitatively with model calculations which show that the formation of HCl proceeds quickly when O3 levels fall so low that (1) the rate of the reaction Cl+O3→ClO+O2 slows and (2) the rate of the reaction NO+ClO→NO2+Cl becomes faster than the rate of the competing reaction NO+O3→NO2+O2. Under these conditions, Cl increases at the expense of ClO, and HCl is formed via the reaction Cl+CH4→HCl+CH3. Stratospheric chlorine is thus shifted from reactive species to the long‐lived, reservoir molecule HCl. The repartitioning of the active chlorine family in favor of HCl halts the processes that destroy ozone and makes available active nitrogen in the form of NO and NO2. The investigation confirms earlier results and validates tracer analysis as a reliable method to probe chemical change in the stratosphere.