Assessing site suitability for Scots pine using airborne and terrestrial gamma-ray measurements in Finnish Lapland

Abstract

Low-altitude airborne gamma-ray (AGR) data, obtained with the aircraft-mounted 25-l NaJ(Tl) gamma spectrometer and interpolated into a 50 × 50 m pixel size, over 1200 km² of glaciated terrain in Finnish Lapland, were applied to classify and interpret site suitability for Scots pine (Pinus sylvestris L.). The selection of the training and validation sets for the classification of AGR data was based on the forest management history and soil moisture content (θ_v) determined by dielectric (ε) measurements. The ground calibration measurements showed a significant negative correlation between the soil ε (i.e., θ_v) and terrestrial gamma-ray flux (TGR-γ) from potassium (γ_K) and thorium decay series (γ_Th), suggesting that the attenuation of gamma flux is due to soil θ_v. Both ground and airborne surveys indicated that γ_K was significantly higher in drift of Scots pine stands than in drift of Norway spruce (Picea abies (L.) Karst.) stands. Out of four tested combinations of the AGR channels, i.e., (i) potassium (K), (ii) K and thorium (Th)/K, (iii) K, Th, and Th/K, and (iv) K and Th, K alone resulted in the best overall accuracy of 80.44% (Kappa coefficient, κ = 0.609) to classify drift materials suitable for Scots pine. The present study demonstrates that the TGR-γ_K and AGR-γ_K measurements provide a basis to delineate soil θ_v patterns within the site and landscape level, thus having a significant implication for the forest management planning to assess sites suitable for Scots pine.