A tree-ring field reconstruction of Fennoscandian summer hydroclimate variability for the last millennium
- First Online:
- Cite this article as:
- Seftigen, K., Björklund, J., Cook, E.R. et al. Clim Dyn (2015) 44: 3141. doi:10.1007/s00382-014-2191-8
Hydroclimatological extremes, such as droughts and floods, are expected to increase in frequency and intensity with global climate change. An improved knowledge of its natural variability and the underlying physical mechanisms for changes in the hydrological cycle will help understand the response of extreme hydroclimatic events to climate warming. This study presents the first gridded hydroclimatic reconstruction (0.5° × 0.5° grid resolution), as expressed by the warm season Standardized Precipitation Evapotranspiration Index (SPEI), for most of Fennoscandia. A point-by-point regression approach is used to develop the reconstruction from a network of moisture sensitive tree-ring chronologies spanning over the past millennium. The reconstruction gives a unique opportunity to examine the frequency, severity, persistence, and spatial characteristics of Fennoscandian hydroclimatic variability in the context of the last 1,000 years. The full SPEI reconstruction highlights the seventeenth century as a period of frequent severe and widespread hydroclimatic anomalies. Although some severe extremes have occurred locally throughout the domain over the fifteenth and sixteenth centuries, the period is surprisingly free from any spatially extensive anomalies. The twentieth century is not anomalous in terms of the number of severe and spatially extensive hydro climatic extremes in the context of the last millennium. Principle component analysis reveals that there are two dominant modes of spatial moisture variability across Fennoscandia. The same patterns are evident in the observational record and in the reconstructed dataset over the instrumental era and two paleoperiods. The 500 mb pressure patterns associated with the two modes suggests the importance of the summer North Atlantic Oscillation.