Geoscience Reference
In-Depth Information
1 Introduction
In the climate system, the hydrological cycle is a key component of many processes
important for the radiative balance and heat transport related to cloudiness (e.g.,
Tang et al. 2012 ; Gagen et al. 2011 ) and the exchange of sensible and latent heat
with the Earth
s surface (e.g., Seneviratne et al. 2010 ).
Water supply for instance in the Mediterranean region and agriculture in semi-
arid regions is the major climatic factor that impact human societies and ecosys-
tems. For longer time scales paleoclimatology offers the possibility to study
changes in the past hydrological cycle before meteorological measurements are
available.
A previous study (Pauling et al. 2006 ) reconstructed spatial precipitation
'
elds
over Europe applying linear methods (Principal Component Regression or simple
ordinary least square regression). This project aimed at reconstructing the variations
of spatially resolved European precipitation
fields over the past two millennia by
implementing novel statistical reconstruction approaches [Bayesian Hierarchical
modeling (BHM) and analog method] using multiproxy data and regional climate
modeling (dynamical downscaling). The ultimate goals were to analyze the spatio-
temporal variability of past precipitation changes over continental Europe
(including parts of the Mediterranean) and to assess the associated uncertainties on
decadal-to-centennial time scales. A 2000-year spatio-temporal highly resolved
reconstruction represents the backdrop to frame present and projected anthropo-
genic climate changes.
The application of regional modeling combined with the BHM (Tingley and
Huybers 2010 ; Werner et al. 2013 ) and analog method (Franke et al. 2011 ) allow
better estimations of past precipitation variations. The output of the regional model
offers the possibility to test statistical methods such as the BHM and analog method
for the generation of pseudoproxies with the aim to reconstruct spatially resolved
precipitation patterns. In the following, examples for two European regions, the
larger Alpine area and central Europe representing different levels of spatial het-
erogeneity, will be used to characterize the skill of the statistical and precipitation
and its evolution over the last 2,000 years.
2 Materials and Methods
The resolution of global climate models (GCMs) used within the paleoclimatic
context is still very coarse with grid spacings in the order of a few hundreds of
kilometers. To analyze precipitation variability, the European climate over the past
two millennia is therefore downscaled with high resolution regional climate models
(RCMs). The
finer horizontal resolution in the order of 50 km with a model time
step of typically three minutes allows a better representation of precipitation
processes, for instance related to the formulation of convective precipitation.
Search WWH ::




Custom Search