La Plata city is located in a mid-latitude region in Argentina, near sea level (Latitude 34.90\({^\circ}\)S, Longitude 57.92\({^\circ}\)W and 25 m asl) and the local time is GMT-3 h. Data acquisitions were obtained in two opposite conditions, clear sky and cloudy days. The experimental period was between Spring equinox and Summer solstice. From August \(29^{th}\) to December \(4^{th}\), 2013, the solar zenital angle at noon was gradually reduced from 44.02\({^\circ}\) to 11.68\({^\circ}\). The solar noon times at the beginning and end of the period were at 12:52 and 12:42, respectively. In one hour around noon, \(E^{S}_{\lambda}\) measured was practically constant as well as the ambient temperature. The average temperature at noon for all experimental days was (19.2\(\pm\)5.4)\({^\circ}\)C, while the humidity and pressure were (48.2\(\pm\)7.7)% and (1021.5\(\pm\)11.7)hPa respectively.
The TUV model (Subsection \ref{TUV}) was applied to estimate \(E^{S}_{\lambda}\). We took into account the surrounding topography and also several parameters provided by satellites and previous studies \citep{Luccini2006,Cabrera2012}. However, we estimated a higher aerosol value than recorded by AERONET because an industrial area and an oil refinery are located just outside the La Plata city. Additionally, the mean surface reflectivity value of 0.05 for typical grass soil was considered \cite{Koelemeijer2003}. Lastly, a single scattering albedo of 0.95 \cite{Luccini2006} was included.
The Avantes spectrometer (Subsection \ref{hrspectrometer}), mounted in a black flat platform on grass, allowed us to easily measure the \(E^{S}_{\lambda}\) (composed by a Sun-Earth direct line of sight incident beam plus a diffuse one) coming from the whole celestial sphere (Figure \ref{irrallday}). As soon as the parameters were according to the environment, multiple comparisons were performed between \(E^{S}_{\lambda}\) values calculated applying the TUV model and those recorded with the spectrometer. Both sets of spectra showed a good agreement, resulting a mean relative difference of about 14% among them. As an example, in Figure \ref{irrallday}a the \(E^{S}_{\lambda}\) registered by the spectrometer in a clear sky day is compared to that calculated with the TUV model under the same environmental conditions.
Further, the solar UV irradiance at Earth’s surface (\(E^{S}_{UV}\)) was obtained by integrating \(E^{S}_{\lambda}\) from \(\lambda_{1}\)= 250 nm to \(\lambda_{2}\) = 400nm:
\[\label{eq:enp} E^{S}_{UV}=\int_{\lambda_{1}}^{\lambda_{2}}E^{S}_{\lambda}d\lambda\]
The hourly \(E^{S}_{UV}\) values, calculated from \(E^{S}_{\lambda}\) values measured at different times during the day using the spectrometer, showed a typical gaussian behavior and matched very well with the estimations obtained by the TUV model (Figure \ref{irrallday}b).
To find out if the solar radiation is enough to photolyse pterin derivatives of biological relevance, the values of \(E^{S}_{UV}\) (Figure \ref{irrallday}), obtained in our experiments under outdoor conditions, were compared to the intensity of a source used in previous studies \citep{Vignoni2009,Vignoni2010}. In this way, a Rayonet RPR lamp was employed as an artificial source and its corresponding \(q^{0,V}_{n,p}\) values was measured by means of an actinometer (detailed in Subsection \ref{aber}) \cite{Terazima}, for a sample placed in front of the lamp, as described in literature \citep{Vignoni2009,Vignoni2010}.
Taking into account that the lamp emits quasi-monochromatic radiation, \(q^{0,V}_{n,p}\) values were converted into the UV irradiance of the lamp (\(E^{L}_{UV}\)) with the following equation:
\[\label{eq:eL} E^{L}_{UV}= q^{0,V}_{n,p}\cdot N_{A} \cdot h\nu \left(\frac{V}{S}\right)\]
where \(N_{A}\)h\(\nu\) is the energy of a mol of photons emitted by the lamp (\(\lambda_{em}\) = 350 nm) and V and S are, respectively, the volume and the area exposed to irradiation of the cell used. A value of 41 \(W\cdot m^{–2}\) was obtained for \(E^{L}_{UV}\), which is of the same order of magnitude as those determined for \(E^{S}_{UV}\) measured in La Plata city (Figure \ref{irrallday}b). These results provide evidence that the photooxidation of pterins such as Bip and H\(_{2}\)Bip should take place at a significant rate under solar exposure.
In addition, as a control, the Avantes spectrometer was employed to measure \(E^{L}_{UV}\), using an equation equivalent to Equation \ref{eq:enp}, and a value of 36 \(W\cdot m^{–2}\) was obtained, which is in good agreement with that calculated using the chemical actinometer. Finally, this comparison confirms the proper operation of the fiber optic spectrometer.