Objective
1.General background
The harmful effects of intense UV-B radiation (280 to 315 nm) on various forms of life on earth are well documented. In particular it is well-known that excessive exposure of the human skin to UV-B radiation will increase the risk of developing skin cancers. In recent years a worldwide increase in the incidence of skin cancers has been observed among the white population. High UV-B levels may also cause eye diseases and may effect the human immune system. Therefore a life-style with increased exposure to intense sunlight implies a risk to public health.
The long term reduction of column ozone amounts which has been observed during the last two decades over middle and high latitudes is also a matter of concern because it is believed to cause an increase in the UV-B irradiance on Earth. But even after a complete recovery of the ozone layer, other factors which contribute to the extinction of UV in the atmosphere may be equally important for the variability on different time-scales of UV-B irradiance near the surface. Therefore long term programmes of reliable UV-B observations and public information should be set up for reasons similar to those concerning other hazardous atmospheric phenomena.
In the "Environment and Climate" Community research programme much work is devoted to a European UV-B measuring network. Some steps taken within the World Meteorological Organization (WMO) Global Atmospheric Watch (GAW) also aim at a quality improvement of UV measurements and the establishment of a global UV monitoring network. Furthermore, a number of countries in the northern as well as in the southern hemisphere have established operational UV-B forecasts as part of a public awareness programme. Various empirical models or radiative transfer models for the calculation of UV radiation near the surface are in use. These models take into account the solar zenith angle, the ozone column density and Rayleigh scattering. But other important parameters such as various types of cloudiness, aerosol loading in the stratosphere and the troposphere, vertical distribution of ozone, surface albedo and column densities of NO2 and SO2 may be included.
An important aspect in UV-B forecasting for public awareness is the way in which the information is presented to the public. Many countries use an index to indicate the intensity of the erythemally weighted UV-B irradiance whereas others use the time required to obtain a sunburn for a specific human skin type. At a WMO meeting on UV-B measurements, data quality and standardization of UV indices (Les Diablerets, Switzerland, July 1994), it was recommended to use a UV index similar to the index used in Canada. Many European countries started (or will start in the near future) an operational UV-B forecasting programme. From a survey which has been carried out by the Danish Meteorological Institute it appears that in these countries various ways are used to present the UV forecasts to the public.
Differences also exist among the European countries that already have implemented or plan to implement an operational UV-B forecast in the radiative transfer code used to calculate the UV-B irradiance at ground level and the methods used to obtain the various input parameters to this code. One of the
important input parameters is the total column ozone amount which is mostly predicted from its correlation with one or several meteorological parameters such as the temperature at a certain pressure level, the altitude of the tropopause, the height of a certain pressure level, or the vertically integrated potential vorticity. Again, the choice of how the column ozone amount is predicted varies among countries.
The development and implementation of UV-B forecasting programmes in different European countries lead to an overlap of efforts and a lack of harmonization. A coordination between the various existing and planned programmes of UV-B forecasting is therefore highly desirable. This can be achieved in the framework of a COST Action.
2.Objectives of the action
The final goal is to establish tested and standardized UV-B radiation forecasts for public information in Europe. To achieve this goal a number of steps have to be taken. They can be summarized as follows:
2.1To provide the necessary input data for the UV forecasting with a quality relevant for public information.
2.2To test and to optimize models for the calculation of UV radiation near the surface.
2.3To implement a validation of forecasted ozone amounts by high quality ground-based ozone measurements and/or by satellite data.
2.4To improve the methods to include cloudiness in UV forecasts.
2.5To implement a validation of the final product for public information.
2.6To follow the WMO/WHO (World Health Organization) recommendations concerning issuing the final product for public information.
3.Scientific content of the action
3.1Input data for public UV-forecast
Various meteorological variables as well as time and site dependent variables are needed to make UV forecasts. In order of importance the main variables are:
-solar elevation: has a strong influence on the daily and seasonal course of UV radiation at the surface. This variable can be readily calculated using astronomical formulae.
-Clouds: mostly decrease ground level UV irradiance, although under special cloud conditions an increase of the irradiance may also be observed. Cloud forecasts are provided with variable accuracy; also satellite pictures can be helpful.
-Ozone: daily variations in the total column ozone (which may amount to 30% in Europe) and in the vertical distribution of ozone cause variations in ground level UV radiation. Ozone forecasts are based on the correlation between its column amounts and certain atmospheric variables. The forecasts require reliable ground based observations of total column amount of ozone but the vertical distribution may be desirable also; on a larger scale the forecasts may be improved by using satellite ozone data.
-Aerosol: its effect on ground level UV irradiance may amount to 15%, depending on local and regional aerosol sources. The quantification of this effect could be improved by means of backward trajectory studies.
-Albedo: becomes significant for UV-B radiation only at values higher than 10% (e.g. at snow-covered surfaces). Input data in models may be improved by using satellite data.
Research and sensitivity studies are needed to assess the impact of these variables on ground level UV irradiance. For this purpose the Management Committee of the Action will:
(1)implement sensitivity studies to assess the relative importance of each of these variables;
(2)implement studies to improve the quality of the input data with emphasis on aspects relevant to UV forecasts issued to the public and paying proper attention to the outcome of sensitivity studies.
3.2Test/optimize models for computing UV radiation
UV-B forecasting models that are currently in use can be divided into three types:
-Empirical models: models to compute UV irradiances based on fits of several years of UV observations; the input variables are usually solar elevation and ozone.
-Simple parameterized models, e.g. the well known Green's model; ozone, solar elevation, and sometimes aerosol are used as input.
-Sophisticated radiative transfer models: among these are the so-called doubling/adding method and the discrete ordinate method. They have in common the fact that the atmosphere is characterized in as detailed a way as possible (by profiles of temperature, humidity, ozone, sulphur dioxide, aerosol, scattering and absorbing properties of atmospheric constituents, etc).
UV-B forecasts from the participating countries should be made consistent to some extent. For this purpose the Management Committee of the Action will:
(1)implement studies to test the models, compare computed UV irradiances to each other and to observed UV irradiances, and make improvements if needed;
(2)implement studies to optimize the models by considering operational (computer time) and scientific (input data quality) aspects.
3.3Implementation of validation of ozone forecasts
To make a UV forecast, it is necessary to make a forecast of the amount of ozone in the atmosphere first. Various methods are used, e.g. based on the height of certain temperature levels or the tropopause, or using column integrated potential
vorticity; all these methods need to be validated. For this purpose the Management Committee of the Action will:
(1)make an inventory of the methods that are currently used to make a forecast of the amount of ozone and implement a validation scheme;
(2)examine the feasibility of developing chemical-dynamical models for ozone forecasts;
(3)take the necessary steps to exchange high quality ground-based total ozone measurements among the participating countries, for the validation of ozone forecasts;
(4)employ large-scale ozone fields measured with satellite instruments for validating forecasted ozone fields. Currently total ozone amounts from TOVS (TIROS Operational Vertical Sounder) are available. In the first half of 1995 the ERS-2 satellite will be launched with the GOME (Global Ozone Monitoring Experiment) which will allow for the first time to make accurate ozone measurements from a European satellite. In the future ozone measurements from other European instruments (such as SCIAMACHY) will be available.
3.4Cloudiness in UV forecasts
The largest short-term variations of the spectral UV radiation at the surface are caused by changes of cloudiness. Therefore it is important to forecast the influence of the expected cloud conditions on the UV index. Unfortunately there are still many problems to overcome in the assessment of the influence of clouds:
-The spectral attenuation factor for spectral global radiation by cloudiness is not well-known.
-The forecast of cloudiness according to the type and amount of clouds is a difficult task.
-A proper forecast of the time of cloud appearance (at least within 4 hours around noon) is critical since the daily course of biologically effective UV radiation at the surface has - compared to visible radiation - a relatively pronounced maximum at local noon.
The National Meteorological Services that are engaged in an operational UV forecast take the effect of cloudiness into account by a number of different methods (e.g. by forecasting a UV-B index for clear sky conditions and by applying simplified attenuation factors for different classes of cloud cover).
The problems related to cloudiness in UV forecasts need to be tackled. For this purpose the Management Committee of the Action will:
(1)implement studies to improve the accuracy of cloudiness attenuation factors;
(2)implement studies to improve the forecast of cloudiness and local cloudiness effects;
(3)implement the development of "last hour" forecast procedures using now-cast cloudiness, for public information at sunbathing sites;
(4)make recommendations for worst case assumptions for public information, if the cloudiness forecast is very uncertain.
3.5Implementation of validation of final products
As it is a recognized practice in the meteorological community to predict only quantities that can be experimentally verified, a number of measures are unavoidable before UV forecasts on a continental scale become meaningful. For this purpose the Management Committee of the Action will:
(1)support and make use of the efforts within WMO and Community research programmes, to determine calibration procedures for UV instruments and to improve the quality of UV measurements;
(2)prepare proposals and support efforts on the exchange of reliable UV radiation observations with a short time delay among the participating countries, the minimum exchangeable information being the UV index at local solar noon as recommended by WMO (see point 3.6);
(3)select a suitable algorithm for the quantitative assessment of the quality of UV forecasts;
(4)take measures so that the validation of UV forecasts can be continued after this COST Action has expired;
(5)provide experimental data - preferably from locations where the seasonal variation of the surface reflectance is large - on possibilities to forecast UV irradiances falling on tilted surfaces;
(6)assess the possible role of ECMWF in UV-B forecasts.
3.6WMO/WHO recommendations concerning UV forecasts
At a WMO meeting on UV-B (Les Diablerets, Switzerland, July 1994) the following recommendation regarding a forecasted UV index was formulated:
"Create a standard UV index based on the following criteria:
Utilization of the C.I.E. (1987) action spectrum normalized to 1,0 at 297 nm.
A minimum requirement to report irradiance values at local solar noon.
The index is expressed by multiplying the weighted irradiance in W/m2 by 40,0 (this will lead to an open-ended index, which is normally between 0 and 16)".
To facilitate the understanding of UV-B forecasts by the public when travelling from one country to another, it is desirable that all European countries issue (at least) the same type of index to indicate the expected amount of erythemally effective UV-B radiation at the surface, which is currently not the case. For this purpose the Management Committee of the Action will take the necessary steps to adopt the WMO/WHO recommendations concerning a common UV index.
4.Timetable
The Action will last for four years. Since the improvement of radiative transfer models and the validation of forecasted parameters will be a continuous and interactive process, it is likely that most of the Action areas specified in section 3 will be investigated in parallel.
5.Organization management and responsibilities
A Management Committee will be set up following the appropriate number of signatories to the Action Memorandum of Understanding. A Chairman will be elected and tasked to draft an outline project plan based upon the items specified in section 3. This plan will be discussed and approved by the Management Committee and submitted to the Technical Committee for Meteorology.
In order to carry out the work a limited number of working groups will be formed with the responsibility to carry out specific tasks.
The Management Committee will meet at least twice per year. Progress reports will be produced on an annual basis and a detailed final report will be published at the end of the Action. It is expected that at least one workshop will be arranged during the Action. The Management Committee will report formally on an annual basis to the COST Senior Officials Committee through the Technical Committee for Meteorology, and will also provide short reports at each meeting of the Technical Committee either verbally or in written form.
6.Economic dimension of the action.
It is expected that about 12 countries will participate and that each country will allocate up to 4 man-years (mainly by scientists) to the Action, which means that the total Action cost will amount up to about ECU 3,2 million including overheads.
Current status
To reach these objectives, it has been decided to create three Working Groups (WG) :
WG 1 : Data
This WG is in charge of two projects :
Input data for public UV-forecasts
Various meteorological variables as well as time and site dependent variables are needed to make UV forecasts. In order of importance the main
variables are :
Solar elevation,
Clouds,
Ozone,
Aerosol,
Albedo.
Research and sensitivity studies will be carried out to assess the impact of these variables on ground level UV irradiance.
Cloudiness in UV forecasts
The largest short-term variations of the spectral UV radiation at the surface are caused by changes of cloudiness. There are still many problems to overcome in the assessment of the influence of clouds :
To tackle the problems related to cloudiness the WG 1 is in charge of :
implementing studies to improve the accuracy of cloudiness attenuation factors ;
implementing studies to improve the forecast of cloudiness and local cloudiness effects ;
implementing the development of forecast procedures using now-cast cloudiness ;
making recommendations for worst-case assumptions for public information, if the cloudiness forecast is very uncertain.
WG 2 : Models
This WG is responsible for testing/optimising models for computing UV radiation.
UV-B forecasting models that are currently in use can be divided in 3 types :
Empirical models,
Simple parameterised models,
Sophisticated radiative transfer models.
They have in common that the atmosphere is characterised in as much detail as possible (by profiles of temperature, humidity, ozone, sulphur dioxide, aerosol, scattering and absorbing properties of atmospheric constituents, etc.).
UV-B forecasts from the participating countries should be made consistent to some extent. For this purpose WG 1 is in charge of :
implementing studies to test the models ;
implementing studies to optimise the models.
WG.3 : Validation
This WG is responsible for two projects :
Implementation of validation of ozone forecasts.
To make a UV forecast, it is necessary to make a forecast of the amount of ozone in the atmosphere first. Various methods are used which shall be validated.
For this purpose the WG 1 will :
make an inventory of the methods ;
examine the feasibility of developing chemical-dynamic models for ozone forecasts ;
among the participating countries ;
ozone fields.
Implementation of validation of final products
For UV forecasts on a continental scale becoming meaningful the WG 1 has to :
support and make uses of the efforts within WMO and EC to improve the quality of UV measurements ;
prepare proposals and support efforts on the exchange of reliable UV radiation observations ;
select a suitable algorithm for the quantitative assessment of the quality of UV forecasts ;
take measures so that the validation of UV forecasts can be continued after this COST Action has expired ;
provide experimental data on the possibilities of forecasting UV irradiances falling on tilted surfaces ;
assess the possible role of ECMWF in UV-B forecasts.
Programme(s)
Topic(s)
Call for proposal
Data not availableFunding Scheme
Data not availableCoordinator
1049 Brussels
Belgium