3.3. Field experiment

The field site for the study was located near the University of Oulu (65ºN) (Fig. 1). A lamp bank consisting of three lamps (Philips TL 12/40 W) was mounted 50 cm above a selected branch of each Scots pine. The pines were about 40 years old, and were allocated to ambient (7 trees) and supplemental UV-B (7 trees) treatments. The lamps were filtered either with 0.1 mm cellulose-diacetate film (transmission down to 290 nm) for supplemental UV-B or with 0.1 mm polyester film (absorbs almost all radiation below 316 nm) for ambient treatment. The filters were changed once in a month. The lamps were on for 6 hours per day centered at solar noon on 7 days a week. The UV-B doses were adjusted to follow the normal UV-B doses, which vary during the day and the growing season (Jokela et al. 1995). The daily UV-B doses were regulated by a potentiometer to give different UV-B fluence rates without causing spectral displacement. Spectral irradiances under the lamps were measured at least once a week with a spectroradiometer (Model 754, Optronics, Orlando, FL, USA) at 1 nm intervals. The absolute spectral UV irradiance of the Philips lamp was weighted with the generalized plant response action spectrum (Caldwell 1971) and normalized to 300 nm to obtain the daily biologically effective irradiance (UV-B_BE). UV-B doses varied in the ranges of 0.5–2.4 kJ m⁻² d⁻¹ (ambient) and 0.7–5.1 kJ m⁻² d⁻¹ (supplemental UV-B) (Table 2). Solar radiation (305–1500 nm) was measured with a pyranometer CM-3 (Kipp & Zonen). In the first year, 1996, the exposure was started on June (III) and in the other two years at the beginning of April (IV, V, VI). In each year, exposure was discontinued at the end of September.