Analysis of Compound Climate Indices across Europe: from Trends to Risk Zones

With climate change impacts increasing evident across European continent, understanding and accurately characterizing climatic events and their compound interactions is important for effective climate risk assessment and mitigation strategies. Climate indices is an important tool for monitoring changes in the frequency, intensity, and duration of climate events, yet detailed continental scale assessments remains limited. The focus of the dissertation (five interrelated studies) is to provide a comprehensive framework for analyzing climate trends, identifying compound risk patterns, and improvising precipitation data across Europe through advanced statistical techniques and high-resolution data analysis.

The first study determines the spatial and temporal trends of 74 climate indices from 1950 to 2021 at a high resolution of 0.1° on a monthly scale. Using Mann-Kendall and Sen’s slope estimator, significant increase in growing degree days and decrease in heating degree days are identified, along with changes in other climate indices such as universal thermal climate index, relative humidity, wind chill index, global radiation, and potential evapotranspiration. Country-specific zoning reveals southeastern Europe experiences highest number of warmer days, while eastern Europe faces lower colder days.

The second part focuses on compound climate indices, employing 27 bivariate and 10 trivariate high correlated climate indices pairs of copula models to identify regions with risks. Northern and eastern Europe countries are vulnerable to compounded risks involving global radiation, temperature, evapotranspiration, and bioclimatic indices. The exposure of agricultural and coastal areas to these risks, need further integrated risk assessment approaches.

In the third study, probabilistic approach used to derive high exposure zones across Europe based on 56 climate indices. Precipitation and temperature based indices are important in identifying high-risk regions for southern and southeastern Europe, while northern and western Europe are more affected by precipitation indices. This study also tracks the exposure of dominant land use types and population density to these risks.

To improvise the accuracy of the precipitation based climate indices, the fourth study addresses the challenge of wet antenna attenuation in commercial microwave links used for rainfall estimations. By developing frequency dependent models calibrated with rain gauge data, the study improves reliability of rainfall intensity estimates, particularly for high-frequency bands.

The final study evaluates the bias-correction technique for downscaling precipitation data from CMIP6 models across Europe. The Empirical Quantile Mapping (EQM) technique outperforms other methods, and when combined with Random Forest machine learning models, provides more accurate regional precipitation predictions, especially for low rainfall intensities. The approach reduces the overestimation in central and eastern Europe and is also a part of enhancement technique of precipitation data.

Collectively the study advances the climate indices impacts in Europe and provides methodological framework for improving the accuracy of precipitation data. By integrating detailed trend analysis, compound event assessment, high-risk exposure zones, and advance downscaling techniques, this dissertation contributes to the field of climate science especially in developing effective strategies to mitigate the climate risks in Europe.