Click
Here
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 111, D17312, doi:10.1029/2005JD006580, 2006
for
Full
Article
Ultraviolet climatology over Argentina
Eduardo Luccini,1 Alexander Cede,2 Rubén Piacentini,1,3 Carlos Villanueva,4
and Pablo Canziani5
Received 10 August 2005; revised 17 March 2006; accepted 26 May 2006; published 14 September 2006.
[1] Satellite-derived climatologies of UV index (UVI) and erythemal daily dose were
determined for Argentina at a geographical resolution of 0.5 latitude by 0.5 longitude.
Total Ozone Mapping Spectrometer climatology of key input parameters was used to
calculate the clear-sky ground-level solar UV irradiance. NASA Surface meteorology and
Solar Energy cloud cover data were used to determine the UV attenuation by clouds.
Two cases were tested: (1) monthly averages of total cloud cover and (2) monthly averages
for three ranges of cloud cover percentage (0–10%, 10–70%, and 70–100%). Case 2,
with smaller biases, was selected. Measured erythemal irradiance at seven stations of the
Argentine Ultraviolet Monitoring Network was used to validate the satellite-derived
climatologies, as well as to estimate aerosol parameters, surface albedo corrections, and
UV cloud transmittance for the calculations. Annual average biases of the monthly mean
satellite-derived UVI and erythemal daily dose with respect to the ground measurements
are mostly within ±10%. The strong longitudinal gradient of UV levels toward the
Andes Mountains is emphasized. Very high UVI and erythemal daily dose values are
registered in the northwestern tropical high-elevation Andean plateau, with extreme
monthly means above 18 and 10 kJ/m2 respectively, in December–January. Even northern
low-elevation regions show averages over 12 and 7 kJ/m2, respectively. On average,
clouds attenuate the clear-sky erythemal irradiance by less than 20% for most of the
continental region during all months. UV levels are considerably higher than those for
equivalent regions in the Northern Hemisphere.
Citation: Luccini, E., A. Cede, R. Piacentini, C. Villanueva, and P. Canziani (2006), Ultraviolet climatology over Argentina,
J. Geophys. Res., 111, D17312, doi:10.1029/2005JD006580.
1. Introduction
[2] Solar UV radiation, especially the UVB range (280 –
320 nm), which is strongly influenced by the ozone content
in the atmosphere, has large impact on the life and materials
on Earth [United Nations Environment Programme
(UNEP), 2003]. In appropriate doses, it is beneficial for
several biological reactions [Wharton and Cockerell, 1998;
Grant, 2002], but many negative effects on humans, plants,
animals and materials can be caused by excessive UV
exposure [UNEP, 2003]. The progressive depletion of the
stratospheric ozone layer in the last decades caused concern
by an eventual increase in the UV radiation levels reaching
1
Instituto de Fı́sica de Rosario, Consejo Nacional de Investigaciones
Cientı́ficas y Técnicas – Universidad Nacional de Rosario, Rosario,
Argentina.
2
Science Systems & Applications Inc., Lanham, Maryland, USA.
3
Facultad de Ciencias Exactas, Ingenierı́a y Agrimensura, Universidad
Nacional de Rosario, Rosario, Argentina.
4
Servicio Meteorológico Nacional, Buenos Aires, Argentina.
5
Programa de Estudios de Procesos Atmosféricos en el Cambio Global,
Pontificia Universidad Católica – Consejo Nacional de Investigaciones
Cientı́ficas y Técnicas, Buenos Aires, Argentina.
Copyright 2006 by the American Geophysical Union.
0148-0227/06/2005JD006580$09.00
the Earth’s surface [UNEP, 2003]. Particularly, Argentina is
placed very near the Antarctic Continent and is frequently
affected by the ‘‘ozone hole’’ which develops every year in
springtime [e.g., World Meteorological Organization,
2003]. The typical UV radiation levels at surface must be
known to correlate with possible biological and chemical
effects.
[3] The effective UV irradiance acting on a specific
biological system is obtained by weighting the incoming
spectral UV irradiance with its corresponding ‘‘action spectrum’’. In particular, the erythemal irradiance is the wavelength-integrated spectral irradiance weighted by the action
spectrum of the human skin defined by McKinlay and
Diffey [1987]. The UV index (UVI) is defined as the
erythemal irradiance at horizontal plane multiplied by a
factor of 40 m2/W [World Health Organization (WHO),
2002] (available at http://www.who.int/uv/publications/
globalindex/en/) and indicates the risk of producing damage
on typical human skin exposed to the sun. UVI is currently
informed to the people from the daily maximum erythemal
irradiance [WHO, 2002], and its analysis is based on local
solar noon values. UVI information is intended to prevent
sudden damaging effects. Information on typical UV daily
dose is also important as other effects, like photoaging and
carcinogenesis, are associated to cumulative long-term UV
exposure [e.g., Amer and Metwalli, 1998].
D17312
1 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
[4] Even though there are many sites in the world where
solar UV radiation at the surface is measured, these data are
usually representative only for a local area. If a long enough
database is available, the UV radiation climatology of
averages and trends for this single place can be determined
[Weatherhead et al., 1998; Intergovernmental Panel on
Climate Change, 2001]. To establish UV climatology over
an extended region on ground-based measurements, it is
necessary to strategically arrange a large number of instruments in representative zones and then making an appropriate geographical interpolation of their long-term data.
However, for logistic and economic reasons, the number of
distributed instruments is usually small. Another frequently
used approach to establish UV climatology consists in using
a reliable radiative transfer model to calculate the clear-sky
ground-level UV irradiance and applying attenuation factors
due to cloudiness. The input parameters for the radiative
transfer calculations in the UV range include the elevation
of the place, total ozone column, aerosol parameters, surface
UV albedo and cloud optical parameters. These input data
are often obtained from satellite databases, which cover the
whole Planet at variable geographical resolution. The calculations are then validated by comparison with the measured UV irradiance at sites in the region.
[5] Several UV radiation global climatologies have been
established in this way: Lubin et al. [1998] (at a geographical resolution of 2.5 latitude by 2.5 longitude), Mayer
and Madronich [1998] (1 1.25), Sabziparvar et al.
[1999] (5 5), Madronich et al. [1999] (1 1.25),
Herman et al. [1999] (approximately 1 1), Pubu and Li
[2003] (1 1.25). Verdebout [2000] uses a similar
approach to generate high-resolution maps over Europe
(0.05 0.05). Fioletov et al. [2003] determined UV maps
for Canada by applying an interpolation method to the
calculated values at specific sites through correlations
between the UV and the total solar radiation (300 –
3000 nm). Fioletov et al. [2004] used an improved version
of this algorithm, together with the Total Ozone Mapping
Spectrometer (TOMS) satellite method, to generate the UV
index climatology over the United States and Canada (1 1.25). Meloni et al. [2000] interpolate the calculated values
at 69 sites to determine UV maps in Italy at a resolution of
0.5 0.5.
[6] The Argentine Ultraviolet Monitoring Network (UV
Net) consists currently of 9 stations, where the erythemal
irradiance is measured. The analysis of their measurements
after an extensive calibration campaign [Cede et al., 2002a]
gave important results on characterization of the UV irradiance [Cede et al., 2002b], effects of clouds on the UV
irradiance [Cede et al., 2002c], comparison with UV TOMS
satellite method and corrections to the surface UV albedo in
zones with high variability [Cede et al., 2004], and estimation of aerosol parameters [Cede, 2001] at the stations. The
small number of stations in such an extended area prevents
the application of rigorous interpolation methods such as
kriging, inverse square distance, etc. [e.g., Shepard, 1968;
Caruso and Quarta, 1998]. Instead, the information retrieved from these sites is used in this work to construct
climatological UV maps from model calculations of the
clear-sky erythemal irradiance, corrected for mean attenuation by cloudiness. Even though the time span of available
data is not extensive enough, the term ‘‘UV climatology’’
D17312
will be used as a characterization of long-term average UV
values in the region under study.
[7] In this paper we present the climatologies of solar
erythemal UV irradiance at noon and erythemal daily dose
on a horizontal plane for Argentina, both calculated at a
geographical resolution of 0.5 0.5. We noted that
working at a geographical resolution of 0.5 0.5
supplies important local-scale details in regions with large
gradients in the UV irradiance levels, with respect to lowerresolution calculations. The base of this paper was developed by Luccini [2003] (available at http://www.ifir.edu.ar/
~solar/luccini/tesis/) using TOMS Version 7 ozone data and
Surface meteorology and Solar Energy (SSE) release 4
cloud data. Recently, TOMS Version 8 (available at http://
toms.gsfc.nasa.gov/) and SSE release 5.1 (available at
http://eosweb.larc.nasa.gov/sse/) data sets were published
in August and December 2004 respectively. Nevertheless,
new versions imply changes that are not substantial to the
present work [Wellemeyer et al., 2004; P. W. Stackhouse,
NASA Langley Atmospheric Sciences Data Center, personal
communication, 2005]. Given that a new method was
applied to determine the UV climatologies, Luccini [2003]
explored some additional steps like the fitting of several
parameters in search of improving the approach, which were
not applied in this version.
[8] The erythemal UV measurements used to validate
and estimate the uncertainty of the present climatologies
comprise the period 1996– 2000 and correspond to the
following UV Net stations: 1, La Quiaca (22.11S,
65.57W, 3459 m a.s.l.); 2, Pilar (31.66S, 63.88W,
338 m a.s.l.); 3, Rosario (32.96S, 60.62W, 25 m
a.s.l.); 4, Buenos Aires (34.61S, 58.41W, 25 m a.s.l.);
5, Comodoro Rivadavia (45.78S, 67.50W, 46 m a.s.l.);
6, San Julián (49.32S, 67.75W, 62 m a.s.l.); and 7,
Ushuaia (54.80S, 68.27W, 14 m a.s.l.). Station Mendoza
(32.88S, 68.87W, 704 m a.s.l.) started its measurements
during the calibration campaign and was not included, and
station Marambio is placed in Antarctica being then out of
the area studied in this work. Details on the stations,
measurements, calibration and characterization of erythemal irradiance at the UV Net are given by Cede et al.
[2002a, 2002b].
[9] Maps of the average UV attenuation by clouds in the
whole region are also determined, and the hourly course of
average cloud attenuation is studied by comparison of the
modeled attenuation with that resulting from the measurements at the UV Net stations. The climatological maps for
January, April, July and October are presented in this paper.
Climatological data and maps for all months of both UV
index and erythemal daily dose are available online at
http://www.ifir.edu.ar/~luccini/uvclimatology/.
2. Data
[10] Both calculations of the clear-sky UV irradiance and
cloudiness attenuation use monthly mean values of the input
parameters. UV index was determined from noon irradiance
calculations for each day, while hourly calculations within
each day were made to determine the erythemal daily dose.
[11] Calculations of the clear-sky irradiance were made
with the one-dimensional discrete ordinates radiative transfer model described in Table 1, at each grid point with a
2 of 15
Surface albedo
Rayleigh scattering
Ozone absorption
Total ozone column
Ozone vertical profile
Aerosol parameters
Continental AOD340
Marine AOD340
Single-scattering albedo
Asymmetry factor
Wavelength dependence of AOD
Continental a
Marine a
Aerosol vertical profile
Temperature and pressure vertical profiles
Number of streams
Wavelength grid
Extraterrestrial flux
Layers
Description/Value
DISORT algorithm [Stamnes et al., 1988], pseudospherical version based on the Tropospheric Ultraviolet and Visible (TUV) radiation model
(available at http://cprm.acd.ucar.edu/Models/TUV/)
16
121 wavelengths from 280 to 400 nm, step 1 nm
SUSIM-Atlas 3 spectrum (available at ftp://susim.nrl.navy.mil/pub/atlas3)
58 layers: 0.5 km width from 0 to 2 km, 1 km width from 2 to 50 km, and 5 km width from 50 to 80 km; 0 km means the elevation of the
station above sea level
formula from Bodhaine et al. [1999]
cross sections by Bass and Paur [1985]
TOMS Version 7 Level 3 pixel-by-pixel climatology 1978 – 2000 with trend
TOMS Version 7 climatology of ozone profiles for corresponding latitude and total ozone column [McPeters et al., 1998]
Model type
Table 1. Radiative Transfer Model Used for the Clear-Sky Calculations
latitudinal interpolation with altitudinal correction from the estimated monthly mean values at the UV Net stations [Cede, 2001]
0.2 (mean value from the measurements of the AVHRR satellite instrument)
0.95 [Hess et al., 1998]
0.73 [Hess et al., 1998]
Angström formula AOD(l) = bla
1.11 [Hess et al., 1998]
0.75 (mean value from AVHRR)
rural type from Shettle and Fenn [1979], combined with the ‘‘Warneck’’ vertical profile [Erlick and Frederick, 1998]
COSPAR International Reference Atmosphere (1986) (0 km to 120 km) for corresponding latitude and time (available at http://nssdc.gsfc.nasa.gov/
space/model/atmos/cospar1.html)
climatology from Herman and Celarier [1997], corrected for ‘‘mean’’ albedo toward the southernmost region as suggested by Cede et al. [2004]
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
D17312
geographical resolution of 0.5 0.5 (approximately 55 km
by 55 km) between latitudes 21.25S and 55.25S and
longitudes 53.25W and 73.75W, thus covering the whole
continental region of Argentina, part of its neighboring
countries and of the South Atlantic Ocean. Input parameters for the clear-sky calculations are the surface mean
elevation, total ozone column, surface UV albedo and
aerosol parameters. Finally, satellite cloud cover data were
used to determine the average attenuation of the UV
irradiance by cloudiness in the same grid.
2.1. Elevation
[12] The altitude increase effect on the UV erythemal
irradiance may be over 20%/km under certain conditions
[Pfeifer et al., 2006]. Then the surface elevation grid,
representing the average height of the surface for a given
geographical resolution, is a key parameter in the calculations. For this work, the elevation grid was obtained
from the terrain digital model of the region [see, e.g.,
Cancoro, 2005] (available at http://www.teses.usp.br/teses/
disponiveis/3/3138/tde-10102005-104155/) at a geographical resolution of 0.5 0.5. It is shown in Figure 1
(upper left panel). Note that at this resolution, many pixels
corresponding to the northwestern tropical high-elevation
Andean plateau (NAP) have elevations above 4000 m, and
some ones even above 5000 m (the highest-elevation pixel
is 5610 m a.s.l.). Taking into account that the highest peak
of the region is Mount Aconcagua (32.65S, 69.98W)
with 6960 m a.s.l., the selected grid is very appropriate for
the calculations.
2.2. Ozone
[13] A monthly pixel-by-pixel climatology of total ozone
vertical column was built by analyzing TOMS Version 7/
Level 3 data from the whole period of Nimbus 7 satellite
(1978 – 1993) and the first years of EarthProbe satellite
(1996– 2000). Data from EarthProbe were merged to the
Nimbus 7 data by using their differences to the groundbased Dobson network. This climatology comprises monthly
averages, standard deviations and tendencies of the total
ozone over the studied region. To derive the tendencies, the
monthly means for each pixel of 1 latitude by 1.25
longitude were first corrected for the seasonal and solar
cycles, El Niño effect and the quasi-biennial oscillation,
and then a linear regression was applied on the residues. In
order to change from the original resolution of 1 1.25 to
the new resolution of 0.5 0.5, the ozone amount
below the pixel’s elevation was estimated using the TOMS
Version 7 ozone vertical profiles and added to the monthly
means to get the mean total ozone at sea level. After this,
data were bilinearly interpolated to the 0.5 by 0.5 grid and
finally the ozone amount below the elevation of these new
pixels was resubtracted. UV calculations were made for
year 2000. As an example, the ozone map for January 15,
2000 is also shown in Figure 1 (upper right panel).
2.3. Albedo
[14] The surface UV albedo from the TOMS monthly
climatology [Herman and Celarier, 1997] was used as a
base. Given that it was determined from an average of the
smallest albedo values at each point, this climatology is
appropriate for zones and seasons without large variations
3 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
Figure 1. Maps of the surface elevation, monthly mean total ozone column, surface UV albedo, and
aerosol optical depth at 340 nm, used as input parameters in the daily calculations of the clear-sky
erythemal irradiance in Argentina. Maps for each month were built at a geographical resolution of 0.5 0.5 of the calculations. Maps for January are shown as an example.
in the albedo. However, in regions with large variability,
e.g., due to frequent snowfall, it is more appropriate to
replace it by a climatology of ‘‘mean’’ albedo [Cede et al.,
2004]. Fioletov et al. [2004] use a snow flag method to
estimate the presence of snow in daily UV TOMS estimates.
The work by Cede et al. [2004] showed that the TOMS
climatology represents rather well the surface UV albedo for
stations 1 (La Quiaca) to 6 (San Julián), providing also the
4 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
monthly average corrections to the albedo at station 7
(Ushuaia), where snowfall is frequent. On the basis of those
results, in the present work the albedo from the latitude of
San Julián station (49.32S) was increased linearly over the
continental region toward the corrected albedo of Ushuaia
station (54.80S) for each month. Maximum correction
increased the albedo at Ushuaia from the climatological
TOMS value of 0.07 up to 0.25 in July. As an example, the
UV albedo map for January is also shown in Figure 1 (lower
left panel).
2.4. Aerosols
[15] The aerosol parameters used for the calculations
are the total vertical aerosol optical depth at 340 nm
(AOD340), the Angström exponent a, the single-scattering
albedo and the asymmetry factor. The Angström formula
AOD(l) = bla for the spectral dependence of the aerosol
optical depth was used, where b is the Angström turbidity
coefficient. The monthly means of AOD340 in the continental region were obtained by geographical interpolation
of the average values estimated at the stations from UV
measurements under clear-sky conditions [Cede, 2001].
Initially, a linear interpolation in latitudinal direction was
made for AOD340. In addition we applied a correction for
elevations above 400 m, reducing the AOD340 linearly
down to the AOD340 value of station 1 at 3459 m (average
AOD340 0.17 at station 1 [Cede, 2001]). Then, other
linear interpolation above 3459 m was made between the
AOD340 value of station 1 and an estimated very clean
atmosphere AOD340 of 0.01 at 5700 m. The elevation of
400 m was selected, as it is just above all the low-elevation
stations of the UV Net (stations 2 – 7, see section 1).
[16] Of course, the AOD340 maps are just rough estimates
within the limited available data. For example, the AOD340
at the latitude of the urban station 4 (Buenos Aires) is
extrapolated to every site below 400 m at the same latitude
in the continent, including many rural sites on that latitude.
However, this is only a conceptual observation as the
estimated AOD340 has rather similar values for all the
central stations: Pilar (rural), Rosario (river coastal-urban)
and Buenos Aires (big city) [Cede, 2001]. As pointed out by
Cede et al. [2004], the aerosol correction is important as it
reduces significantly the systematic differences between
calculations and measurements. The future incorporation
of data sets from the NASA Aerosol Robotic Network sites
in the region (AERONET, http://aeronet.gsfc.nasa.gov/) and
NASA Moderate Resolution Imaging Spectroradiometer
(MODIS, http://modis-atmos.gsfc.nasa.gov/) satellite data
will give new insights on this subject. Continental a was
taken constant at the value 1.11 [Hess et al., 1998]. In the
marine region, the mean values of AOD340 and a were
determined from the measurements of the Advanced Very
High Resolution Radiometer (AVHRR) satellite instrument
[Mishchenko et al., 1999]. As both parameters exhibited no
important seasonal variation, constant values of AOD340 =
0.2 and a = 0.75 were used throughout the year in the
marine region. Also, given the small differences, a constant
single-scattering albedo of 0.95 and asymmetry factor of
0.73 were used in both continental and marine regions [from
Hess et al., 1998]. With these parameters, maximum reduction of erythemal irradiance throughout the region due to
aerosols, with respect to the aerosol-free case, amounts
D17312
12% for AOD340 0.5 in the central continental region
[Cede et al., 2004]. As an example, the AOD340 map for
January is also shown in Figure 1 (lower right panel).
2.5. Clouds
[17] Cloud data were taken from the 10-year (1983 – 1993)
global climatology given by the SSE release 4 data set [see
Whitlock et al., 2000], at a geographical resolution of 1 1
obtained from the original (280 km equal-area grid) International Satellite Cloud Climatology Project (ISCCP) D1 database. For comparison, SSE Release 5.1 consists of true 1 1 resolution cloud data obtained from the raw ISCCP DX
database, implying improvements in grid boxes bordering
topographic features like the Andes Mountains (P. W. Stackhouse, NASA Langley Atmospheric Sciences Data Center,
personal communication, 2005). Nevertheless, given that the
major cause of the large UV gradient in these zones, i.e., the
surface elevation grid, is taken at true 0.5 0.5 resolution,
cloudiness play a modulating role that reduces the influence
of such differences. Two cases using SSE data were tested.
Case 1 consists in monthly average total cloud cover percentage every 3 hours for each month (at 0000, 0300, 0600, 0900,
1200, 1500, 1800 and 2100 UT). Case 2 supplies the
percentage of time within the month that the sky is in each
of the following conditions: clear skies from 0% to 10% cloud
cover, broken-cloud skies from 10% to 70%, and nearovercast skies from 70% to 100%, every 3 hours (at 0000,
0300, 0600, 0900, 1200, 1500, 1800 and 2100 UT). No
distinction in the type of clouds or in the frequency distribution within each case of cloud cover is given. A bilinear
geographical interpolation and a linear time interpolation
were made to obtain grids of 0.5 0.5 every 1 hour in
each case.
[18] As will be seen in section 3.3, case 2 produced the
smaller biases between the calculations and ground measurements, and all the following analysis is centered on SSE
case 2 data. As an example, the maps for each range of SSE
case 2 cloud cover at 1600 UT (around local solar noon) in
January are shown in Figure 2.
3. Methods and Results
3.1. UV Index Climatology
[19] With the input data introduced in section 2, calculations of the clear-sky erythemal irradiance at noon for
every day in the year were made for each point of the 0.5 0.5 grid. Monthly means were determined, obtaining in this
way the ‘‘clear-sky climatology’’ of noon erythemal irradiance. Then, the average attenuation by clouds was incorporated. For this purpose, the measured average erythemal
effective transmittance factors for each case of SSE cloud
cover data were obtained from a previous work on UV
measurements and cloud cover observations at the stations
of the Argentine UV Net [Cede et al., 2002c]. In that work,
major differences were noted for large cloud cover percentage between the high-elevation station (1, La Quiaca) and
the low-elevation ones. In the NAP, high clouds with
smaller optical depth are more frequent, yielding a larger
transmittance of erythemal irradiance [Cede et al., 2002c].
On the basis of the results of that paper, the average
erythemal transmittance curve as a function of the cloud
cover percentage was used for SSE case 1. In turn, the
5 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
Figure 2. Maps of the percentage of time within the month with cloud cover between 0 and 10%, 10
and 70%, and 70 and 100% in Argentina, from NASA SSE release 4. Data are used to apply the cloud
attenuation to the calculated clear-sky UV irradiance. Maps for January at 1600 UT (around local noon)
are shown as an example.
erythemal transmittance curve was averaged within each
range of SSE cloud cover for case 2: the mean erythemal
transmittance factors for the cloud cover ranges 0 – 10%
(T0 – 10) and 10– 70% (T10 – 70) were set at the values 1
and 0.96 respectively, for the complete region. While, the
mean erythemal transmittance in the cloud cover range
70– 100% (T70 – 100) was set to 0.78 for the high-elevation
station and 0.67 for all the other (low-elevation) stations.
Finally, a map was constructed for T70 – 100 making 0.67
the transmittance for every continental or marine region,
except toward the NAP were a linear altitudinal interpolation was made for elevations between 1 km (T70 – 100 =
0.67) and 3 km (T70 – 100 = 0.78), setting T70 – 100 = 0.78
for elevations above 3 km. A small systematic difference
appears here. As several works showed [Malberg, 1973;
Schreiner et al., 1993; McKenzie et al., 1998], satelliteestimated cloud cover fraction is generally smaller than
that determined from ground observations. This is to be
expected, as the first one is representative of the area
covered while the latter is the fraction of the sky
hemisphere covered. From the effective cloud transmittance curve, determined as a function of the ground
observations of cloud cover [Cede et al., 2002c], this
implies that the effective transmittance of corresponding
SSE cloud cover interval is slightly overestimated (SSE
cloud cover range 70 – 100% implies ground-observed
>70 – 100%). The higher satellite-derived UV cloud transmittance with respect to that measured from the ground,
particularly for large cloud cover, was stated by Matthijsen
et al. [2000] and Williams et al. [2004] on ISCCP D1
cloud data. The difference is more notorious when observed
on daily averages than on monthly averages, given that
large cloud cover percentages are more frequent for shorter
time periods. Then, differences by this fact are negligible
for the present work.
d,noon
is the modeled clear-sky erythemal irradi[20] If Ics,mod
ance at noon at each grid point on day d of month M having
N days, then the satellite-derived monthly mean erythemal
irradiance including the attenuation by cloudiness for SSE
case 2 is given by
ISM;noon ¼
N
1 X
M ;noon
M ;noon
I d;noon ðT010 F010
þ T1070 F1070
N d¼1 cs;S
M ;noon
þ T70100 F70100
Þ
M ;noon
M ;noon
M ;noon T010 F010 þ T1070 F1070
þ T70100 F70100
¼
N
N
X
d;noon
Ics;S
ð1Þ
d¼1
M,noon
M,noon
where FM,noon
0 – 10 , F10 – 70 and F70 – 100 are the fractions of
time of month M around noon with the sky in the three
ranges of cloud cover (0 – 10%, 10– 70% and 70– 100%)
respectively, provided by the SSE database (see Figure 2).
The small time offset between the noon calculations at
each grid point and the cloud cover data at 1600 UT
(ranging from 27 min on the east to +55 min on the west
of the region under study) is completely concealed by the
monthly average of cloud data. Of course, the condition
M,h
M,h
FM,h
0 – 10 + F10 – 70 + F70 – 100 = 1 is satisfied for every hour
h within month M.
[21] Thus climatological maps of monthly mean erythemal irradiance at noon, converted to UVI units, were
obtained for each month in the year. Figure 3 shows,
particularly, the monthly mean UVI for January, April, July
and October.
[22] UVI for the low-elevation zones (see Figure 1)
shows the characteristic latitudinal dependence, responding
to the behavior of the stratospheric ozone and the increasing
6 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
Figure 3. Maps of the satellite-derived monthly mean UV index for January, April, July, and October in
Argentina, calculated at a geographical resolution of 0.5 0.5.
7 of 15
D17312
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
Figure 4. Maps of the satellite-derived monthly mean erythemal daily dose for January, April, July, and
October in Argentina, calculated at a geographical resolution of 0.5 0.5.
noon solar zenith angle (SZA) toward the south. Monthly
mean values for the low-elevation regions range between
about 12 in the north and 6 in the south in summer, and 4
and less than 1 respectively in winter. Simultaneously, a
strong longitudinal gradient of the UVI at fixed latitudes is
observed toward the Andean mountains in the west of the
country, with extreme averages above 18 around January in
the NAP. The high elevation, the small noon SZA and the
relatively thin and stable total ozone column at low-latitude
regions [e.g., Herman et al., 1991, 1993] cause the charac-
8 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
teristically high UVI values observed throughout the year
over this tropical and subtropical area, which are within the
highest UVI levels for densely populated regions in the
world. Values are even higher than those registered at
Lhasa (29.7N, 91.1E, 3648 m a.s.l.) on the Tibetan
plateau, with monthly mean UVI over 16 in July [Pubu
et al., 1999]. It is important to remark that the mean UVI
values observed in the southernmost region during springtime may be strongly exceeded in cases when the Antarctic
‘‘ozone hole’’ passes over the continent [Dı́az et al., 2001;
Cede et al., 2002b].
[23] The geographical resolution of the calculations,
strongly dominated by the resolution of the elevation grid,
is critical to represent local-scale details of the UV irradiance levels within a given region. A comparison of the
present 0.5 0.5 UVI climatology with the global UV
radiation climatology (GUVRC) by Lubin et al. [1998],
which has a geographical resolution of 2.5 2.5 and is
valid for the period 1985 – 1989, shows minimum differences in the low-elevation zones for every month. Those
differences are within the uncertainty of the calculations for
the different epochs they represent. On the contrary, in
regions with large altitudinal gradient like the Andes Mountains, GUVRC strongly underestimates the irradiance values
with respect to the present climatology (over 30% in the
NAP), mainly because of the smaller resolution of its
elevation grid.
[24] Comparison of the climatology in this area of the
Southern Hemisphere (SH) with equivalent ones in the
Northern Hemisphere (NH) will be restricted to lowelevation zones, so as to avoid differences that may be
attributable only to the geographical resolution of the calculations. For example, mean noon erythemal irradiance for
coastal zones in January from Figure 3 is systematically
higher by about 20– 25% from latitudes 30 to 50S than
the equivalent TOMS climatology in North America from
30 to 50N in July [Fioletov et al., 2004]. A similar result
is observed from the global TOMS climatology by Herman
et al. [1999], where NH and SH are analyzed with the same
TOMS algorithm. About 7% of this difference is inherent to
the smaller Earth-Sun distance in the SH summer solstice
with respect to the NH summer solstice [e.g., Liou, 2002]. To
study the contribution of ozone to that difference, we
determined the time average of latitudinal ozone bands from
the TOMS climatology (complete years 1979 – 1992 for
Nimbus 7 and 1997 – 2005 EarthProbe satellites). Summer
average total ozone column for band 30 –50S is about 9%
smaller than for band 30– 50N. Taking into account that
the radiation amplification factor for erythemal UV irradiance, relating the percentage change of erythemal irradiance
to a given percentage change of ozone, is approximately 1.1
[Madronich et al., 1998], then about 10% of the north-south
difference is explained by the ozone layer. On the other hand,
a test was made to analyze the NH-SH differences in cloud
attenuation using the SSE database from the low-elevation
regions 30 – 50S/50– 70W of the present work and 30 –
50N/70– 90W of North America, under the assumption
that the same cloud transmittance factors T detailed previously are valid for both regions. Average differences between
the total cloud attenuation in both NH and SH regions are
smaller than 2% in summer and 5% in winter. In summary, of
the total 20 –25% higher summer UVI along the 30 – 50S
D17312
low-elevation band in Argentina with respect to the summer
UVI along the 30– 50N low-elevation band in North
America, about 10% is attributable to the ozone layer, 7%
to the Earth-Sun distance, and less than 8% to the sum of
clouds, aerosols and surface albedo.
3.2. Erythemal Daily Dose Climatology
[25] To determine the climatology of erythemal daily
dose, hourly calculations of the clear-sky erythemal irradiance were made in the same geographic grid of 0.5 0.5
within each day in the month. Then the hourly clear-sky
maps were corrected for clouds attenuation using the hourly
interpolated data for both cases of SSE cloud cover mentioned in section 2.5. The obtained hourly erythemal irradiance including cloud attenuation was integrated in each
grid point within the corresponding day to obtain the
erythemal daily dose. Finally, the daily dose maps were
averaged for each month.
[26] The satellite-derived monthly mean erythemal daily
dose using SSE case 2 data for month M is given by
DM
S ¼
N
1 X
Dd;M
N d¼1 S
ð2Þ
where
Dd;M
¼
S
24
X
d;h
M ;h
M ;h
M;h
Ics;S
T010 F010
þ T1070 F1070
þ T70100 F70100
h¼1
ð3Þ
is the calculated daily dose including the attenuation by
cloudiness for SSE case 2 on day d (the time interval factor
and integration only for daylight time are implicit). The
same transmittance factors T than for the case at noon were
applied (section 3.1), supported by the fact that null or very
small dependence with the SZA was found in the effective
UV transmittance as a function of the cloud cover
percentage for all sky conditions [Kuchinke and Nunez,
1999; Josefsson and Landelius, 2000; Cede et al., 2002c;
Alados-Arboledas et al., 2003; Foyo-Moreno et al., 2003].
[27] Figure 4 shows the climatological maps of monthly
mean erythemal daily dose for January, April, July and
October. A behavior similar to that of the noon erythemal
irradiance is noted: a pronounced latitudinal dependence
with a strong longitudinal gradient toward the high-elevation
Andes Mountains at fixed latitude. Low-elevation regions
present typical daily dose values between about 7 kJ/m2 in
the north and 4 kJ/m2 in the south in summer, and 2 kJ/m2
and 0.2 kJ/m2 respectively in winter. The NAP presents
extreme monthly mean values of erythemal daily dose over
10 kJ/m2 in December – January. For comparable latitude
and elevation, values are considerably higher than those
registered at Lhasa on the Tibetan plateau with monthly
mean daily dose of about 7.5 kJ/m2 in July [Pubu et al.,
1999].
[28] A comparison between climatologies for the SH and
NH is made again for low-elevation zones only. Mean
erythemal daily dose at 50S in January from Figure 4 is
about 4.1 kJ/m2, i.e., approximately 30% higher than the
equivalent for 50N over Canada in July [Fioletov et al.,
9 of 15
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
D17312
[30] A small systematic difference appears at this point
given that the measurements correspond to the average
over 30 min, while calculations are for the instantaneous
local solar noon giving systematically higher values.
This difference, verified for the clear-sky case, is of
less than 1% and therefore negligible within the involved
uncertainties.
[31] Similarly, the percentage difference of the satelliteM
derived (DM
S ) with respect to the ground-measured (DG )
monthly mean erythemal daily dose at a given station for
month M is
M
DS DM
G
DSGM
¼
100x
D
DM
G
Figure 5. (top) Percentage difference of the satellitederived monthly mean UV index for SSE case 2, with
respect to that measured from the ground at the stations,
numbered as in section 1, throughout the year. (bottom)
Annual mean percentage difference for SSE case 1 (gray
symbols) and for SSE case 2 (black symbols) at the stations
as a function of latitude and variability at the level of 2
standard deviations.
2003]. As mentioned, part of this difference is due to the
smaller Earth-Sun distance around the SH summer solstice.
Also in the daily dose the geographical resolution of
calculations yields significant differences to climatologies
with lower resolution. As an example, the global climatology by Pubu and Li [2003], at a resolution of 1 1.25,
underestimates the monthly mean values of erythemal daily
dose by above 20% in the NAP.
3.3. Validation and Uncertainty Analysis
[29] Monthly means of the UV measurements at the
stations were taken as a reference to validate the satellitederived climatologies and estimate its uncertainty. Comparison was made with the calculated values at the closest grid
point to each station. The percentage difference of the
) with respect to the groundsatellite-derived (IM,noon
S
) monthly mean UVI at a given station
measured (IM,noon
G
for month M is given by
DSGM
I
¼ 100x
ISM ;noon IGM ;noon
IGM ;noon
!
ð4Þ
ð5Þ
[32] Tests of uncertainty were made for the two cases of
SSE cloud cover data. SSE case 2 showed the smaller
uncertainties and is emphasized. In Figure 5 the uncertainty
in the calculation of UVI is analyzed. Figure 5 (top) shows
DSGM
I for SSE case 2 at the stations during the year. A
slight tendency is noted in the calculated values to underestimate the measured monthly means, with the largest
difference of 23% in April at station 6 (San Julián). This
is more evident in Figure 5 (bottom), where the annual
average of DSGM
I at the stations for SSE cases 1 and 2 is
shown as a function of latitude and SSE case 2 is highlighted. Variability is shown at the level of 2 standard
deviations (2s). The largest annual mean biases for SSE
case 2 are noted at station 1 (6 ± 6%), station 5 (11 ±
8%) and station 6 (12 ± 12%). In turn, the largest
variability is observed at station 7 (1 ± 14%). It must be
noted that Patagonian stations 5 and 6 show an important
negative bias with respect to central stations 2, 3 and 4 and
southern station 7. As pointed out by Cede et al. [2002a], a
possible underestimation of the TOMS albedo climatology
may also be considered for this region, which could reduce
this difference by about 3 to 6%. Biases are just a little
larger than those found by Cede et al. [2004] for the same
stations from daily TOMS Level 2 – derived erythemal
irradiance at overpass time. This is expected given that
Level 2 ozone and reflectivity overpass data have higher
precision for TOMS calculations, and the retrieved information represents accurately the instant of measurement.
Biases for summer months are of similar magnitude but
generally of opposite sign than those found from comparison between daily TOMS Version 8 – derived and Brewermeasured UVI for summer months over the United States
and Canada [Fioletov et al., 2004].
[33] In Figure 6 the uncertainty in the calculation of the
erythemal daily dose is analyzed. Upper panel presents
DSGM
D for SSE case 2 at the stations during the year, and
bottom panel the annual average of DSGM
D as a function of
latitude, where SSE case 2 is highlighted over SSE case 1
which showed a systematic overestimation of the satellitederived daily dose and also a larger dispersion. The highelevation tropical station 1 (La Quiaca) presents a very
small annual mean bias and variability of +1 ± 5%. On the
other hand, annual mean bias and variability are largest at
the southernmost station 7 (Ushuaia), reaching +10 ± 20%.
The DSGM
D obtained by the present method are in close
agreement with those found by Cede et al. [2004] from
daily TOMS Level 2 – derived erythemal daily dose for
10 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
3.4. Effect of Clouds on the Erythemal Irradiance
[36] The calculation given by
100x
M ;noon
Ics;S
ISM ;noon
M ;noon
Ics;S
!
M ;noon
¼ 100x 1 T010 F010
M ;noon
M;noon
þ T1070 F1070
þ T70100 F70100
ð6Þ
Figure 6. (top) Percentage difference of the satellitederived monthly mean erythemal daily dose for SSE case 2,
with respect to that measured from the ground at the
stations, numbered as in section 1, throughout the year.
(bottom) Annual mean percentage difference for SSE case 1
(gray symbols) and for SSE case 2 (black symbols) at the
stations as a function of latitude and variability at the level
of 2 standard deviations.
stations 1 to 6, and it is noticeable that the systematic
negative differences observed by Cede et al. [2004] at
station 7 due to the underestimation of surface UV albedo
by the TOMS albedo climatology were solved here through
the ‘‘mean’’ albedo correction detailed in section 2.3. Biases
for station 7 are now similar to those for the central lowalbedo stations.
[34] Calibrated ground measurements from other stations
like Mendoza (32.88S, 68.87W, 704 m a.s.l.) [see Cede et
al., 2002a] and from the neighboring countries are currently
being collected to extend the validation of the determined
climatologies.
[35] It is important to emphasize that this method
applies directly to the determination of monthly mean
UVI and erythemal daily doses. Therefore the deviations
found between calculated and measured monthly mean
values at the stations are related to those from other
satellite methods based on daily retrievals (like TOMS
UV reflectivity method) in a direct form, if simplifying
conditions like normal distributions are assumed [e.g.,
Baird, 1995].
where IM,noon
is the monthly mean clear-sky erythemal
cs,S
irradiance at noon, represents the monthly mean percentage attenuation of noon erythemal irradiance by clouds.
Figure 7 shows the maps of the percentage of the clear-sky
noon erythemal irradiance that is attenuated by the
cloudiness throughout the region in January, April, July
and October.
[37] Maps of cloud attenuation of the erythemal daily
dose are in close qualitative and quantitative agreement with
those for noon erythemal irradiance. This is expected given
that most of contribution to the daily dose is given in hours
around the local solar noon. It is interesting to note that the
attenuation by clouds of the clear-sky noon erythemal
irradiance and erythemal daily dose is always below 30%
throughout the whole region. Even more, for most of the
continental and densely populated region the attenuation is
generally of less than 20%. In terms of the UV index, this
implies that cloudiness attenuates, on an average, less than 3
UVI units in the whole region during the year. The
maximum attenuation (over 25%) occurs in the Andean
Southwest, Patagonia and Atlantic Southeast regions, where
predominant southwesterly winds generate a larger cloudiness percentage (see Figure 2). On the other extreme, the
smallest attenuation (below 15%) occurs on the NAP, where
the partial cloud cover predominates [Cede et al., 2002c]
(see Figure 2). These results are in very close agreement
with the monthly mean cloud attenuation of UVB daily
doses from the global UV climatology by Sabziparvar et al.
[1999] for this region. From that work, it can be observed
that the attenuation of less than 15% over the NAP is within
the smaller cloud effect in the SH, similar to the values over
Australia.
[38] In order to study the cloud effect behavior through
daylight hours, the average effective cloud transmittance
M,h
M,h
given by T0 – 10FM,h
0 – 10 + T10 – 70F10 – 70 + T70 – 100F70 – 100 for
every hour h within each month M was calculated for day 1
of each month from the SSE database and compared with
that resulting from the statistics of all 30 min of irradiance
measurements at the stations. For this purpose, the ratio of
measured (under all-sky conditions) to clear-sky erythemal
irradiance was established for each data. Measurements for
SZA < 70 were selected, as uncertainty increases substantially for larger SZA [Cede et al., 2002a]. The reference
clear-sky erythemal irradiance was calculated using the
TOMS EarthProbe daily total ozone column, and the
monthly mean albedo and aerosol parameters at the stations
detailed in sections 2.3 and 2.4, respectively. Results are
shown in Figure 8, where time around 1300 local time (LT)
corresponds to local solar noon. It is interesting to note that
in every station the SSE-derived average cloud transmittance does not vary more than about 10% throughout the
year at all daylight hours. The SSE-derived and the mea-
11 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
Figure 7. Maps of the monthly mean percentage attenuation of the noon erythemal irradiance by clouds
for January, April, July, and October in Argentina, from the satellite-derived calculations at a
geographical resolution of 0.5 0.5.
sured average cloud transmittance at the stations show a
significant agreement. Average cloud transmittance is rather
homogeneous throughout the year for every hour, confirming the fact that similar cloud attenuation was found for
noon irradiance and for daily dose. Even though quantita-
tive SSE-derived and measured values agree for all the
stations, some major qualitative difference in the hourly
course of average transmittance is observed at station 1 (La
Quiaca). As mentioned in section 3.1, in the NAP high
clouds with smaller optical depth predominate and then the
12 of 15
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
D17312
D17312
Figure 8. Average cloud UV transmittance as a function of local time during daylight for all months
from SSE cloud data at the stations (gray lines), compared with the measured average cloud transmittance
from 30-min measurements for SZA < 70 throughout the year (black lines). The maximum and
minimum numbers of measurements averaged at their corresponding hours for each station are given in
parentheses. Local solar noon is around 1300 LT.
measured transmittance overestimates that of SSE. On the
other hand, most of medium-altitude clouds in the NAP are
broken cumulus, which may have a major influence in the
UV attenuation for large SZA, effect that is evident in the
measurements on firsts and lasts daylight hours at La
Quiaca, but is not observed from the satellite.
4. Conclusions
[39] The climatologies of UV index and erythemal daily
dose were established for Argentina on the base of clear-sky
UV calculations with monthly average input parameters on
a radiative transfer model and correction for the mean
attenuation by cloudiness. The studied region includes also
part of Uruguay, south of Brazil, Paraguay, Bolivia, Chile
and the South Atlantic Ocean. TOMS and SSE climatologies provided key parameters for the calculations. Two
cases of SSE cloud cover data were tested. Case 1 consists
of monthly averages of total cloud cover, while case 2
consists of monthly averages for three ranges of cloud cover
percentage (0 – 10%, 10 – 70% and 70 – 100%), every 3
hours (at 0000, 0300, 0600, 0900, 1200, 1500, 1800 and
2100 UT). Case 2 was selected as it showed the smaller
biases with respect to the ground measurements, making
evident that counting with more cloud information improves
13 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
substantially the precision of calculations. Maps of monthly
mean UVI and erythemal daily dose were obtained at a
0.5 0.5 high geographical resolution, which allows
observation of important local-scale details in areas with
large gradients in the radiation values, as is the case for the
Andes Mountains. Monthly mean levels as high as UVI
over 18 and erythemal daily dose over 10 kJ/m2 were
determined around January in the NAP, highlighting the
need to take rigorous protective measures against the
excessive sun exposure for the dense population in this
region. Even low-elevation northern areas have monthly
mean UVI and erythemal daily doses above 12 and 7 kJ/m2,
respectively. Satellite-derived climatologies were validated
against the erythemal irradiance measured at seven stations
of the Argentine UV Net. Annual mean biases between
satellite-derived and ground-measured UVI and erythemal
daily dose were mostly within ±10%. Average attenuation
of both the clear-sky erythemal irradiance at noon and
daily dose by clouds is below 20% for most of the
continental and densely populated zones throughout the
year. Hourly course of the SSE-derived average cloud
attenuation during daylight is very homogeneous, with
variations of less than 10% throughout the year. It agrees
rather well with the average cloud attenuation derived from
all-sky measurements at the stations of the UV Net.
[40] It is interesting to underline that unlike other satellite
methods based on daily UV retrievals to calculate the
monthly averages (see, e.g., Arola et al. [2002] and satellite
algorithms cited in section 1), this new method determines
directly the monthly mean UV irradiance and daily dose,
using only long-term monthly averaged input parameters in
both the clear-sky (ozone, aerosols and albedo) and cloud
attenuation calculations. The geographical resolution of
calculations is defined by the surface elevation grid. We
have demonstrated in this study that working with a surface
elevation grid of 0.5 0.5, calculations based on interpolated TOMS and SSE input data produced reliable results
over an extended area of South America. The only additional parameter necessary to apply this method is the
average effective cloud transmittance for all-sky case as a
function of the cloud coverage. There are large networks in
the world where the average effective cloud transmittance
has already been determined, for example, by Long et al.
[1996] at 21 sites in the United States, and also in other
smaller networks and individual sites [see Calbó et al.,
2005]. At these sites the applicability of this method is
immediate. In turn, given that complementary maps of input
parameters are currently available (e.g., ozone and albedo
from TOMS, aerosols from MODIS, and clouds from SSE),
accurate maps for large regions may be established if a
surface elevation grid of high geographical resolution is
available.
[41] This work provides a detailed reference on the
typical UV irradiance and doses in the region, to be used
in multidisciplinary studies on the effects of solar UV
radiation on public health, biological systems, environment
and materials.
[42] Acknowledgments. This work was supported by the Argentine
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET),
Ministerio de Educación Proyecto FOMEC 142, Servicio Meteorológico
Nacional and Fundación del Cáncer de Piel, and by the World Meteorological Organization. Special thanks are given to Dan Lubin for providing
D17312
the data files of the ‘‘global surface ultraviolet radiation climatology from
TOMS and ERBE data,’’ to Fran McLemore and Kathleen Morris of the
NASA Langley Research Center Atmospheric Sciences Data Center for
their advice in the use of the SSE database, to Cristina Pacino of the
Facultad de Ciencias Exactas, Ingenierı́a y Agrimensura, Universidad
Nacional de Rosario (in agreement with the Instituto Geográfico Militar
de Argentina), for providing the elevation grid data, and to Pablo Garcı́a of
the Observatorio Astronómico de Rosario and Miguel Parodi of IFIR-UN
Rosario for their collaboration in this work.
References
Alados-Arboledas, L., I. Alados, I. Foyo-Moreno, F. J. Olmo, and
A. Alcántara (2003), The influence of clouds on surface UV erythemal
irradiance, Atmos. Res., 66, 273 – 290.
Amer, M., and M. Metwalli (1998), Environmental dermatology: An overview, Clin. Dermatol., 16, 23 – 25.
Arola, A., et al. (2002), Assessment of four methods to estimate surface UV
radiation using satellite data, by comparison with ground measurements
from four stations in Europe, J. Geophys. Res., 107(D16), 4310,
doi:10.1029/2001JD000462.
Baird, D. C. (1995), Experimentation: An Introduction to Measurement
Theory and Experiment Design, 3rd ed., Prentice-Hall, Upper Saddle
River, N. J.
Bass, A. M., and R. J. Paur (1985), The ultraviolet cross-section of ozone,
I, Results and temperature dependence, in Atmospheric Ozone. Proceedings of the Quadrennial Ozone Symposium, edited by C. S. Zerefos and
A. Ghazi, pp. 606 – 610, Springer, New York.
Bodhaine, B. A., N. B. Wood, E. G. Dutton, and J. R. Slusser (1999), On
Rayleigh optical depth calculations, J. Atmos. Oceanic Technol., 16,
1854 – 1861.
Calbó, J., D. Pagès, and J. González (2005), Empirical studies of cloud
effects on UV radiation: A review, Rev. Geophys., 43, RG2002,
doi:10.1029/2004RG000155.
Cancoro, A. C. (2005), Implementaçao de modelos digitais de terreno para
aplicacoes na area de geodesia e geofı́sica na América do Sul, Ph.D.
thesis, Univ. de São Paulo, São Paulo, Brazil.
Caruso, C., and F. Quarta (1998), Interpolation methods comparison, Comput. Math. Appl., 35(12), 109 – 126.
Cede, A. (2001), Erythemal UV in Argentina: Measurements, climatological interpretation and comparison with satellite-derived data, Ph.D. thesis,
Univ. of Innsbruck, Innsbruck, Austria.
Cede, A., E. Luccini, L. Nuñez, R. D. Piacentini, and M. Blumthaler
(2002a), Calibration and uncertainty estimation of erythemal radiometers
in the Argentine Ultraviolet Monitoring Network, Appl. Opt., 41(30),
6341 – 6350.
Cede, A., E. Luccini, L. Nuñez, R. D. Piacentini, and M. Blumthaler
(2002b), Monitoring of erythemal irradiance in the Argentine Ultraviolet
Network, J. Geophys. Res., 107(D13), 4165, doi:10.1029/
2001JD001206.
Cede, A., E. Luccini, L. Nuñez, R. D. Piacentini, and M. Blumthaler
(2002c), Effects of clouds on erythemal and total irradiance as derived
from data of the Argentine Network, Geophys. Res. Lett., 29(24), 2223,
doi:10.1029/2002GL015708.
Cede, A., E. Luccini, L. Nuñez, R. D. Piacentini, M. Blumthaler, and
J. Herman (2004), TOMS-derived erythemal irradiance versus measurements at the stations of the Argentine UV Monitoring Network,
J. Geophys. Res., 109, D08109, doi:10.1029/2004JD004519.
Dı́az, S., G. Deferrari, C. R. Booth, D. Martinioni, and A. Oberto (2001),
Solar irradiances over Ushuaia (54.49S, 68.19W) and San Diego
(32.45N, 117.11W) geographical and seasonal variation, J. Atmos.
Sol. Terr. Phys., 63, 309 – 320.
Erlick, C., and J. E. Frederick (1998), The effects of aerosols on the
wavelength dependence of atmospheric transmission in the ultraviolet
and visible, part II: Typical continental and urban aerosols in clear skies,
J. Geophys. Res., 103, 23,275 – 23,285.
Fioletov, V. E., J. B. Kerr, L. J. B. McArthur, D. I. Wardle, and T. W.
Mathews (2003), Estimating UV index climatology over Canada, J. Appl.
Meteorol., 42(3), 417 – 433.
Fioletov, V. E., M. G. Kimlin, N. Krotkov, L. J. B. McArthur, J. B. Kerr,
D. I. Wardle, J. R. Herman, R. Meltzer, T. W. Mathews, and J. Kaurola
(2004), UV index climatology over the United States and Canada from
ground-based and satellite estimates, J. Geophys. Res., 109, D22308,
doi:10.1029/2004JD004820.
Foyo-Moreno, I., I. Alados, F. J. Olmo, and L. Alados-Arboledas (2003),
The influence of cloudiness on UV global irradiance (295 – 385 nm),
Agric. For. Meteorol., 120, 101 – 111.
Grant, W. B. (2002), Health benefits of solar UV-B radiation: Cancer
risk reduction, in Ultraviolet Ground- and Space-Based Measurements, Models, and Effects, Proc. SPIE Int. Soc. Opt. Eng., 4482,
324 – 334.
14 of 15
D17312
LUCCINI ET AL.: ULTRAVIOLET CLIMATOLOGY OVER ARGENTINA
Herman, J. R., and E. Celarier (1997), Earth surface reflectivity climatology
at 340 nm to 380 nm from TOMS data, J. Geophys. Res., 102, 28,003 –
28,011.
Herman, J. R., R. McPeters, R. Stolarski, D. Larko, and R. Hudson (1991),
Global average ozone change from November 1978 to May 1990,
J. Geophys. Res., 96, 17,279 – 17,305.
Herman, J. R., R. McPeters, and D. Larko (1993), Ozone depletion at
northern and southern latitudes derived from January 1979 to December
1991 Total Ozone Mapping Spectrometer data, J. Geophys. Res., 98,
12,783 – 12,793.
Herman, J. R., N. Krotkov, E. Celarier, D. Larko, and G. Labow (1999),
Distribution of UV radiation at the Earth’s surface from TOMS-measured
UV-backscattered radiances, J. Geophys. Res., 104, 12,059 – 12,076.
Hess, M., P. Koepke, and I. Schult (1998), Optical properties of aerosols
and clouds: The software package OPAC, Bull. Am. Meteorol. Soc., 79,
831 – 844.
Intergovernmental Panel on Climate Change (2001), Climate Change 2001:
The Scientific Basis: Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change,
edited by J. T. Houghton et al., 881 pp., Cambridge Univ. Press, New
York.
Josefsson, W., and T. Landelius (2000), Effect of clouds on UV irradiance:
As estimated from cloud amount, cloud type, precipitation, global radiation and sunshine duration, J. Geophys. Res., 105, 4927 – 4935.
Kuchinke, C., and M. Nunez (1999), Cloud transmission estimates of UV-B
erythemal irradiance, Theor. Appl. Climatol., 63, 149 – 161.
Liou, K. N. (2002), An Introduction to Atmospheric Radiation, 2nd ed.,
Elsevier, New York.
Long, C. S., A. J. Miller, H.-T. Lee, J. D. Wild, R. C. Przywarty, and
D. Hufford (1996), Ultraviolet index forecasts issued by the National
Weather Service, Bull. Am. Meteorol. Soc., 77, 729 – 748.
Lubin, D., E. H. Jensen, and H. P. Gies (1998), Global surface ultraviolet
radiation climatology from TOMS and ERBE data, J. Geophys. Res.,
103, 26,061 – 26,091.
Luccini, E. A. (2003), Radiación solar UV en Argentina: Climatologı́a
espacio-temporal: Caracterización de la radiación solar en regiones de
Sudamérica y Antártida, Ph.D. thesis, Univ. Nac. de Rosario, Rosario,
Argentina.
Madronich, S., R. L. McKenzie, L. O. Björn, and M. M. Caldwell (1998),
Changes in biologically-active ultraviolet radiation reaching the Earth’s
surface, J. Photochem. Photobiol. B., 46, 5 – 19.
Madronich, S., B. Mayer, and C. A. Fischer (1999), A climatology of
erythemal ultraviolet radiation, 1979 – 1992, paper presented at NATO
Advanced Study Institute (ASI) Chemistry and Radiation Changes in
the Ozone Layer, Kolympari, Crete, Greece.
Malberg, H. (1973), Comparison of mean cloud cover obtained by satellite
photographs and ground based observations over Europe and the Atlantic, Mon. Weather Rev., 101, 893 – 897.
Matthijsen, J., H. Slaper, H. A. Reinen, and G. J. Velders (2000), Reduction
of solar UV by clouds: A comparison between satellite-derived cloud
effects and ground-based radiation measurements, J. Geophys. Res.,
105, 5069 – 5080.
Mayer, B., and S. Madronich (1998), Calculation of ultraviolet radiation
quantities using TOMS ozone and ISCCP cloud data, Eos Trans. AGU,
79(45), Fall Meet. Suppl., F170.
McKenzie, R. L., K. J. Paulin, G. E. Bodeker, J. B. Lilley, and A. P.
Sturman (1998), Cloud cover measured by satellite and from the ground:
Relationship to UV radiation at the surface, Int. J. Remote Sens., 19(15),
2969 – 2985.
McKinlay, A. F., and B. L. Diffey (1987), A reference action spectra for
ultraviolet introduced erythema in human skin, in Human Exposure to
Ultraviolet Radiation: Risks and Regulations, edited by W. R. Passchier
and B. M. F. Bosnajakovich, pp. 83 – 87, Elsevier, New York.
McPeters, R. D., et al. (1998), Earth Probe Total Ozone Mapping Spectrometer (TOMS), Data Products User’s Guide, NASA Goddard Space
Flight Cent., Greenbelt, Md.
Meloni, D., G. R. Casale, A. M. Siani, S. Palmieri, and F. Cappellani
(2000), Solar UV dose patterns in Italy, Photochem. Photobiol., 71,
681 – 690, doi:10.1562/0031-8655.
Mishchenko, M. I., I. V. Geogdzhayev, B. Cairns, W. B. Rossow, and A. A.
Lacis (1999), Aerosol retrievals over the ocean by use of channels 1 and 2
D17312
AVHRR data: Sensitivity analysis and preliminary results, Appl. Opt., 38,
7325 – 7341.
Pfeifer, M. T., P. Koepke, and J. Reuder (2006), Effects of altitude and
aerosol on UV radiation, J. Geophys. Res., 111, D01203, doi:10.1029/
2005JD006444.
Pubu, C., and Z. Li (2003), Long-term global Earth surface ultraviolet
radiation exposure derived from ISCCP and TOMS satellite measurements, Agric. For. Meteorol., 120(1 – 4), 51 – 68.
Pubu, C., Y. Gjessing, and F. Sigernes (1999), Measurements of solar
ultraviolet radiation on the Tibetian Plateau and comparisons with discrete ordinate method simulations, J. Atmos. Terr. Phys., 61, 425 – 446.
Sabziparvar, A. A., K. P. Shine, and P. M. F. de Forster (1999), A modelderived global climatology of ultraviolet irradiation at the Earth’s surface,
Photochem. Photobiol., 69, 193 – 202.
Schreiner, A. J., D. A. Unger, W. P. Menzel, G. P. Ellrod, K. I. Strabala, and
J. L. Pellett (1993), A comparison of ground and satellite observations of
cloud cover, Bull. Am. Meteorol. Soc., 74, 1851 – 1861.
Shepard, D. (1968), A two-dimensional interpolation function for irregularly-spaced data, in Proceedings of the 1968 23rd ACM National Conference, pp. 517 – 524, Assoc. for Comput. Mach. Press, New York.
Shettle, E. P., and R. W. Fenn (1979), Models for the aerosols of the lower
atmosphere and the effects of humidity variations on their optical properties, Opt. Phys. Div., Air Force Geophys. Lab., Air Force Syst. Command, U. S. Air Force, Hanscom Air Force Base, Mass.
Stamnes, K., S. C. Tsay, W. Wiscombe, and K. Jayaweera (1988), A numerically stable algorithm for discrete-ordinate-method radiative transfer
in multiple scattering and emitting layered media, Appl. Opt., 27, 2502 –
2509.
United Nations Environment Programme (UNEP) (2003), Environmental
Effects of Ozone Depletion and Its Interactions With Climate Change:
2002 Assessment, Secr. for the Vienna Conv. for the Prot. of the Ozone
Layer and the Montreal Protoc. on Substances That Deplete the Ozone
Layer, Nairobi.
Verdebout, J. (2000), A method to generate surface UV radiation maps over
Europe using GOME, Meteosat, and ancillary geophysical data, J. Geophys. Res., 105, 5049 – 5058.
Weatherhead, E. C., et al. (1998), Factors affecting the detection of trends:
Statistical considerations and applications to environmental data, J. Geophys. Res., 103, 17,149 – 17,161.
Wellemeyer, C. G., et al. (2004), A new release of data from the Total
Ozone Mapping Spectrometer (TOMS), SPARC Newsl. 22, pp. 37 – 38,
Stratospheric Processes and Their Role in Clim., World Clim. Res. Programme, Geneva.
Wharton, J. R., and C. J. Cockerell (1998), The Sun: A friend and enemy,
Clin. Dermatol., 16, 415 – 419.
Whitlock, C. H., et al. (2000), Release 3 NASA surface meteorology and
solar energy data set for renewable energy use, paper presented at 26th
Annual Conference of the Solar Energy Society of Canada Inc. and Solar
Nova Scotia, Halifax, N. S., Canada.
Williams, J. E., P. N. den Outer, H. Slaper, J. Matthijsen, and G. Kelfkens
(2004), Cloud induced reduction of solar UV-radiation: A comparison of
ground-based and satellite based approaches, Geophys. Res. Lett., 31,
L03104, doi:10.1029/2003GL018242.
World Health Organization (WHO) (2002), Global solar UV index: A practical guide, WHO/SDE/OEH/02.2, Geneva.
World Meteorological Organization (2003), Scientific assessment of ozone
depletion: 2002, Global Ozone Res. Monit. Proj. Rep. 47, 498 pp.,
Geneva.
P. Canziani, Programa de Estudios de Procesos Atmosféricos en el
Cambio Global, Pontificia Universidad Católica, Ramon Freire 183,
(C1426AVC) Capital Federal, Buenos Aires, Argentina.
A. Cede, NASA Goddard Space Flight Center, Greenbelt, MD 20771,
USA.
E. Luccini and R. Piacentini, Instituto de Fı́sica de Rosario, Universidad
Nacional de Rosario, 27 de Febrero 210 bis, 2000 Rosario, Argentina.
(luccini@ifir.edu.ar)
C. Villanueva, Servicio Meteorológico Nacional, 25 de mayo 658,
Buenos Aires, Argentina.
15 of 15