Large scale statistical fluctuations of precipitation in Portugal during the 20th century

The impact of changes in climate variables on society and the environment is a major issue worldwide. For example, variations in precipitation over daily, seasonal, annual, and decadal time-scales have an impact on water balances. Although large ranges of spatial and temporal scales are involved, the local impact can be affected by local conditions and regional specificities. This is particularly true and relevant for Portugal, where climate variables are very irregular in space and time. In this study we explore the statistics and temporal structure of precipitation from Portugal over wide ranges of scales using scaling and multifractal methods. In particular, spectral analysis locates the dominant frequencies, and is a useful exploratory tool to identify the upper and lower scales that bound scale-invariant regimes. Moreover, it allows us to investigate climatic fluctuations in the data, over the largest scales, as well as high frequency dynamics revealed by the small scales. The instrumental data investigated are from various locations scattered across Portugal; some of the time series date back to the 19th century. These precipitation time series are the longest available in this geographic region. The focus is on the low frequency weather regime characterized by a spectral plateau with relatively small spectral exponent defined by the power spectrum. This regime extends out to the time scale after which the spectral exponent increases again (i.e. with decreasing frequency), which is interpreted as the signature of the climate regime. The results highlight scaling properties in the data from different origins and identify scaling regimes of special interest for the understanding of climate multi-decadal variability and trends and the establishment of climate change scenarios.