The overall objective of this project is to produce practical recommendations for the deployment of an integrated ultraviolet B(UVB) network throughout Europe.
The main results of these campaigns are summarised below
1. Good calibration procedures are a necessary first step for accurate and reliable measurements. Careful monitoring of the lamp current ensures consistency of the calibration source, the lamp-spectrometer geometry must be rigorously maintained, and stray light within the calibration room must be excluded.
2. The spectrometer geometry and mechanical stability must allow the calibration to remain valid between the fixed conditions of the calibration room and the different instrument orientations, spatial distributions of radiation and ambient conditions in the environment. In most cases this involves at least a temperature stabilisation of sensitive parts of the spectrometer.
3. Important characteristics of the ideal spectrometer are:
(a) wavelength specification. This could be achieved to an accuracy of about 0.1 to 0.3nm in many of the instruments and is acceptable unless very accurate work in the short wavelength UV-B region of the spectrum is required.
(b) slit function. This determines the near field stray light from adjacent wavelengths which is attributed to the nominated wavelength of a measurement. The slit function is important where the measured spectrum changes very rapidly with wavelength, as with solar UV-B. Measuring the slit function of the spectrometers allowed the effects of different slit functions on a measurement of the sun light to be calculated. Correcting for both slit function and wavelength specification improves the comparison between instruments, especially in the UV-B.
(c) far-field stray light. Radiation from wavelengths outside the region of the slit function must be rigorously excluded. if not, it provides a background measurement which limits the sensitivity of the spectroradiometer. As the solar spectrum changes by three orders of magnitude across the UV-B regions, spectrometers need stray light rejection of the same order. In general the single monochromator and diode array instruments did not have sufficient stray light rejection for measurements at the shorter UV-B wavelengths.
(d) cosine response. The fore optics of each spectrometer should have a cosine response but in practice this was often far from perfect. Asymmetry in the cosine response can lead to apparent diurnal asymmetry in the measured irradiances, and different cosine errors can add to discrepancies in response between instruments. Knowing the cosine response of an instrument allows corrections to be made, but these involve assumptions about the spectral distribution of the incident radiation.
In conclusion, a group of UV spectrometers has been identified which could act together to form a UV network. Other instruments require some improvement in design or operational procedures, but could achieve the level of performance set by the core group. The experiences of the campaigners have led to a greater understanding of instrument and operational requirements, which will aid newcomers to this area of research and point the way to future improvements in hardware and methodologies. A transportable lamp system has been designed and built to maintain an independent check of calibrations when instruments are isolated at their home sites.
The recommendations will be based on a comparison of different methods of measurement both between and within instrument categories. Suitability of instruments and the criteria for a network will be judged by their ability to meet the requirements of the data users, atmospheric modellers, photobiologists and photochemists.
In achieving the major aim, 2 other independent objectives need to be met to allow a network based on a number of different instruments. One is the development of a calibration system that can be used for all instruments, the second is a computational means of normalizing the results from instruments with different optical input geometries.
A side product of the proposal will be a data base of UVB measurements made throughout Europe during the 2 years of the proposal. While not a unified network, the investigation of measurement techniques and instruments in the proposal will provide compatibility limits between different locations, and allow data from other networks (eg SAOZ, Systeme d'Analyse par Observations Zenithales) to be used with an acknowledged degree of uncertainty in absolute spectral measurements. The SAOZ network can also provide information on the level of various atmospheric constituents which together with the measured UV irradiances should aid the understanding of UV radiative transfer.
Instrumentation, calibration, standards, sampling procedures, expansion of the core network to existing but less dedicated instruments etc will be discussed with emphasis on providing useful data to the widest possible audience of potential users. All results and experiences in the preceding tasks will be summarized and the requirements for a dedicated UVB network will be determined on the basis of these experiences. The opinions of the final meeting will be presented as the end product of the project.
Funding SchemeCSC - Cost-sharing contracts
3730 AE De Bilt
CB3 0ET Cambridge
59655 Villeneuve D'ascq
CB2 3ET Cambridge