UV-related web tools maintained by Ola Engelsen

FastRT - Fast and easy UV simulation tool

Simulations of ultraviolet irradiances, doses and indices at the Earth's surface at user specified UV wavelengths. A simpler version for non-experts is available at

Author: Ola Engelsen
Norwegian Institute for Air Research
N-9296 Tromsø

Copyright © 2005 Ola Engelsen

Last modified December 20th 2008.

The current model has undergone internal checking and validation. If you have any questions or comments on this service and how it could be improved for your needs, please contact the author. If you use this model for your work I would appreciate if you sent me an e-mail telling me about it.

If you use this program and publish the results, I would appreciate a lot if you cite it:
Engelsen O. and Kylling A., Fast simulation tool for ultraviolet radiation at the Earth's surface. Optical Engineering, 44 (4), 041012 (2005).
A user interface and more information about this model is available at

FastRT changelog


This program computes downward surface irradiances in the spectral range 290-400nm as a function of important radiative parameters such as solar zenith angle, ozone content, cloud and aerosol optical thicknesses, surface reflectance and cloud constellations. The surface irradiances are obtained by spline interpolation of effective transmittances stored in Look-Up Tables (LUTs). The LUTs were computed using the freely available, yet rigorous and accurate, libRadtran atmospheric radiative transfer software package []. LibRadtran is based on the multi-stream discrete ordinates radiative transfer equation solver DISORT by Stamnes et al. (1988). The computations were here done using the pseudospherical approximation (SDISORT) by Dahlback and Stamnes (1991) in order to ensure high levels of accuracy even for low solar elevations. Irradiance is here the corresponding amount of radiant energy (mJ) which flows through a horizontal surface element of unit area (m2) during a unit time (s) per unit wavelength (nm).

LUT files

The lookup tables enclosed with the code were computed using libRadtran version 0.15 assuming the following atmospheric and surface conditions:

The LUT contains effective transmittances (in this case the flux at the surface divided by the extraterrestrial solar flux). The convoluted response is computed by interpolation of the LUT entries.

UV action spectra and doserates

An action spectrum is a parameter that describes the relative effectiveness of energy at different wavelengths in producing a particular biological response. An action spectrum is used as a "weighting factor" for the UV spectrum to find the actual biologically effective doserate (BED) for a given effect. The total doserate (consistent with mW/m2) is found by integrating the product of ultraviolet irradiance and the action spectrum values over the wavelength range from 290-400 nanometers (nm). A daily dose is the doserate integrated over a 24 hour day (= 86400s). The daily dose is consistent with units J/m2 as of version 0.2. For data obtained from versions before 0.2 multiply daily dose by 3.6 to assimilate data to current results.

Note that a direct comparison of dose amounts for any induced effect can usually not be made on the basis of action spectrum alone. These spectra normally provide only an indication of the relative biological effectiveness at particular wavelengths, not the actual dose amount required to produce a biological response. In other words, the biological action spectra are usually expressed in relative sensitivities and units rather than absolute ones. The resulting doserates are thus only loosely attached to physical radiation quantities. The use of units with UV doses may thus not be justifiable. The UV doses computed with different action spectra cannot be compared to each other. However, it is meaningful to compare UV doses relative to solar, surface, and atmospheric conditions as they vary with location and time.

Error Analysis

We have compared FastRT simulations to results obtained from libRadtran at 305nm for a large number of scenarios covering the full range of the major input parameters, i.e. solar zenith angle, ozone column, visibility, surface altitude, surface albedo and cloud liquid water contents. All scenarios selected for this test are centered between the tabular entries which we expect yield the largest deviations. Note that for atmospheric scenarios represented as tabular entries within the FastRT source code, libRadtran and FastRT produce the same results.