Hanno Ohvril

The atmospheric integral transparency coefficient and the Forbes effect

The Forbes effect (known from 1842) expresses the fact that a total (broadband) direct solar beam becomes increasingly penetrating on its way through the atmosphere – because of changes in its spectral content and loss of less penetrating spectral components. The effect restricts the use of the Bouguer-Lambert law for calculating the characteristics of transparency (or turbidity) of the air. Efforts to overcome this effect and to transform the coefficients of transparency to a given solar elevation have finally met with success, and in this study one of such methods (developed by Murk and Ohvril) is presented in detail. This method is compared with those of Evnevich and Savikovskij. For solar elevations 20°≤h≤80° all three techniques are compatible, securing accuracy with an uncertainty below 2.5% with regard to standard values of the coefficient of transparency. Simple transition to the Linke turbidity factor is available. Time courses of mean annual values of the coefficient of atmospheric transparency at solar elevation h=30° and the Linke factor for the last 40 years at an Estonian station at Tiirikoja are given as examples. In the average the multiannual time courses demonstrate an increase of atmospheric turbidity as well as bear evidence of sensitivity to great volcanic eruptions.

Column water vapour: an intertechnique comparison of estimation methods in Estonia

Despite different techniques for the estimation of column integrated water vapour (precipitable water, PW) no method has yet been identified as the most accurate or the reference one. In this work we report intercomparisons between four PW estimation methods - radiosonde, Aerosol Robotic Network (AERONET), Global Positioning System (GPS), and High Resolu- tion Limited Area Model (HIRLAM). Two intensive observation periods at Toravere, Estonia, were used: 22 June−6 November 2008 and 9−12 August 2010. During the longer campaign, only observations by GPS, AERONET, and HIRLAM were performed. An agreement with average difference less than 2.2% among all three methods was established. However, compared to HIRLAM and GPS, the AERONET method overestimated PW by 5-9% at PW 25 mm. In addition, the consistency test applied indicated that previously reported uncertainty in AERONET-measured PW is too high. During the shorter but more complex campaign, data obtained with all four methods were available. Although the average differences between PW from radiosonde and three other methods were < 5%, the discrepancy between single measurements reached 33%. Relatively low temporal and spatial resolution of the HIRLAM grid as well as launching sparseness of radiosondes caused higher scatter from the other methods. The study suggests that besides radiosonde, as a traditional meteorological tool, the most reliable PW estimation can be made by GPS.