Master Thesis Defense by Tobias Nørkjær Holmgaard

Title: Atmospheric modes of variability as a driver for European drought conditions

Abstract:
In a changing climate, it is important to understand drought occurrences for adaptation to the impacts of droughts. For this purpose, better long-term weather forecasting of droughts is highly desired, and calls for a better understanding of the mechanisms behind drought occurrences and whether this can reconstruct previous droughts and their drivers of atmospheric variability. This thesis’ focus is to analyze the relationship between drought conditions with atmospheric modes of variability in the North Atlantic region. Standardized Precipitation Index (SPI) is used as drought index on the timescales 1, 3, 6, 12 and 24 months. Data on the Fraction of Absorbed Photosynthetic Active Radiation Anomaly (FAPARA), from the MODIS sensor on the Terra satellite has been used to verify the SPI as drought index in Europe. SPI-3 is best for detecting vegetation impact drought south of 55◦N, outside of high mountain ranges and the Iberian peninsula. In the cold regions north of 55◦N and high elevation, anticorrelations are found rather than correlations with FAPARA. SPI-6 and SPI-12 have the best correlations at the Iberian peninsula. An EOF analysis has been performed, finding the four most present modes of variability for the extended winter season (NDJFM), the extended summer season (MJJAS), annual modes (all months), and the spring (MAM) at MSLP and the 500 hPa level. The summer modes have a northwards shift compared to the winter modes, with lower explained variance. The PC of the EOF modes has been correlated with the SPI values, where SPI-1 NDJFM has the strongest correlations with the North Atlantic Oscillation (NAO) and the Scandinavian pattern. SPI[1]1 MJJAS is not strongly affected by the NAO, but highly by the Scandinavian pattern (EOF3) and the East Atlantic/West Russian pattern (EOF4), but only the Scandinavian pattern gives robust correlations for SPI-3 MJJAS. SPI-6 MJJAS is in central Europe strongly affected by the spring EOF2 at 500 hPa level. SPI-12 MJJAS is highly controlled by the extended winter season (NDJFM) NAO (EOF1) for Scandinavia and the Mediterranean. The Scandinavian pattern (EOF3) NDJFM is highly related to MJJAS SPI-12 in northwestern Europe. These results have been tested for spatial autocorrelation with Moran’s I and all results are globally highly clustered and are significant at a 99%- significance level, showing large-scale circulation is to a large extent driver of the drought conditions. These results can be used when linking drought conditions from climate proxies such as pollen and tree rings to atmospheric modes of variability. Long-term forecasting of droughts can benefit from these results, with more work needed.

Supervisor: Jens Hesselbjerg Christensen
Censor: Louise Grøndahl