Climate – Tree-Growth Relationships in Central Sweden : An Evaluation of the Palmer Drought Severity Index as a Tool for Reconstructing Moisture Variability

A tree-ring width chronology from Scots pine (Pinus sylvestris L.) was constructed from a xeric site in Stockholm to investigate the relationships between climate and tree growth and to reconstruct past moisture variability. The measure of moisture conditions employed here is a self-calibrating Palmer Drought Severity Index (PDSI). The index is derived from temperature, precipitation, and available water capacity of the soil, and assesses the intensity and duration of drought. It is widely used in tree-ring based climate reconstructions, a method which has never before been tested in the Nordic countries. The comparison of the Stockholm tree-ring chronology with monthly temperature and precipitation data from a nearby meteorological station shows that tree growth is reduced by high summer temperatures, whereas high precipitation at the beginning of the growing season favours growth. The comparison with a PDSI calculated from this meteorological data shows that negative PDSI values are associated with narrow rings. Although tree growth in the humid climate of central Sweden is generally not limited by precipitation, the trees sampled for this study prove to be sensitive to changes in water supply. Their rings thus provide a record of past moisture variability and enable the reconstruction of precipitation and drought. The transfer function models for the reconstructions are calibrated using linear regression. A detailed verification of the results using the more than 200-year long meteorological record from Stockholm affirms the good model performance. May–June precipitation sums and the July PDSI could be reconstructed back to 1625. The Palmer Drought Severity Index is found to be a useful tool in a tree-ring based reconstruction of past moisture variability, approximating the fraction of rainfall which is actually available to the tree, by including soil moisture storage, runoff, and the influence of temperature on evapotranspiration. It cannot completely account for the combined temperature and precipitation forcing of tree growth, and the use of the index does not improve the reconstruction compared to using precipitation alone. However, a reconstruction of both precipitation and the PDSI is possible when selecting an adequate sample site