Seasonal dynamics of albedo across European boreal forests: Analysis of MODIS albedo and structural metrics from airborne LiDAR

• Published in: Remote sensing of environment 2019-04, Vol.224, p.365-381
• Main Authors: Hovi, Aarne, Lindberg, Eva, Lang, Mait, Arumäe, Tauri, Peuhkurinen, Jussi, Sirparanta, Sanna, Pyankov, Sergey, Rautiainen, Miina
• Format: Article
• Language: English
• Subjects: Albedo; Algorithms; Annual variations; Boreal forests; Climate change; Climate effects; Cloud cover; Correlation; Deciduous forests; Dependence; Energy balance; Fjärranalysteknik; Forest effects; Forest management; Forest Science; Geographical locations; Landscape; Lidar; MODIS; Radiative forcing; Remote Sensing; Seasonality; Skogsvetenskap; Snow; Snow cover; Spatial discrimination; Spatial resolution; Taiga
• ISSN: 0034-4257; 1879-0704; 1879-0704
• DOI: 10.1016/j.rse.2019.02.001

Uncertainties in estimation of albedo-related radiative forcing cause ambiguity in evaluation of net climate effects of forests and forest management. Numerous studies have reported local relations between forest structure and albedo in the boreal zone. However, more research is needed to establish these relations for geographically extensive areas, and to examine seasonal courses of albedo to understand the effects of forest structure on mean annual shortwave energy balance. Remote sensing is a viable option for accomplishing these goals, but there are many challenges related to e.g. long periods of cloud cover and low solar elevations in high latitudes. We used the new MODIS Collection 6 (MCD43A3) daily albedo product, and analyzed MODIS albedo dependence on airborne LiDAR-based forest structure in 22 study sites in Estonia, Finland, Sweden, and Russia (57°–69° N, 12°–57° E). Wall-to-wall LiDAR data allowed us to take into account the effective spatial resolution of MODIS, which notably improved correlations between albedo and forest structure. Use of the best quality backup algorithm (magnitude inversion) together with main algorithm results in the MODIS albedo product did not reduce the correlations compared to using main algorithm only. We quantified the effects of landscape-level forest structure (forest height, canopy cover, fraction of young forest) and fraction of broadleaved deciduous forest on mean annual albedo. We showed that because the forest structure-albedo relations are the strongest in snow-covered periods, and because the snow-covered period is longest in the north, the effect of forest structure on mean annual albedo increases towards the north. On the other hand, the effect of broadleaved fraction did not show such latitudinal trend. Our results indicate that even within a single climatic zone the optimal forest management solution to mitigate climate change depends on geographic location. •Linking daily MODIS albedo and forest structure in a large geographic area•Airborne LiDAR used for deriving forest structure wall-to-wall for each MODIS pixel•Accounting for MODIS effective resolution strengthened albedo-structure relations•Albedo-forest structure relations stronger in snow covered than snow-free seasons•Effects of snow and forest structure on mean annual albedo increased towards north