Behavior of stable isotopes (D – 18O) of H2O in atmospheric processes

Thumbnail Image



Σαλαμαλίκης, Βασίλειος

Journal Title

Journal ISSN

Volume Title



The use of environmental tracers, tracers that naturally exist in the Earth system not due to any anthropogenic activity, may offer direct insights into the mechanisms and interactions taking place within the components of the global hydrological cycle. Among environmental tracers, stable water isotopic species due to their conservative essence have been extensively used in order to seek information about several meteorological, hydrological, climatic, etc. disciplines at different spatial and temporal scales. The isotopic applications may extend from meteorological studies as the recognition of the main moisture transport pathways of the air parcels responsible for precipitation over a certain area to anthropological and archaeological applications as the reconstruction of past human dietary patterns and the migration of past populations. The isotopic applications increased nowadays since the isotope derived information from climate archives such speleothems, tree rings, lake sediments, corals, human and animal bones etc. are excellent proxies for the past climate reconstructions. Thus isotopic tracers can be applied in order to obtain information about past climate patterns and to further improve our thoughts for planning and studying the current climate behavior and also the future shifts. Tracing the water cycle requires knowledge of the isotopic patterns for various water reservoirs such as precipitation, lakes, rivers and groundwater. Precipitation is the key component of the global hydrological cycle since it links the atmosphere with the soil-plant continuum. Thus the accurate understanding of the isotopic patterns of precipitation and water vapor is the principal step for water resources studies and planning. This thesis focuses on the behavior of the stable water isotopologues (18O and 2H) in various atmospheric processes under the view of water vapor and precipitation patterns. Chapter 1 provides an overview of the basic isotopic methodology, including the theory of the different fractionation processes that control the distribution of the stable water species in phase changes, the isotopic models for the theoretical simulation of the stable isotopic content of water vapor and precipitation; finally the main isotopic effects that modulate the global isotope precipitation patterns are briefly discussed. Chapter 2 focuses on the isotopic variability of water vapor in a specific location of the Eastern Mediterranean, namely Patras, Greece and its dependence to different air mass trajectories. This study is divided in two parts. The first part examines the temporal variability and the ‘seasonal’ and ‘temperature’ effects of the water vapor isotopes while in the second part a moisture source analysis combined with isotopic tracers is presented. The temperature dependence is more distinct in the mean monthly δv for temperatures below 18 oC, resulting to temperature gradients equal to 0.43‰/oC and 2.6‰/oC for 18O and deuterium respectively. The mean monthly isotopic distribution of water vapor in Patras follows the isotopic variability of a typical coastal station where the maximum of the isotopic composition is reached in June. The Vapor Evaporation Line (VEL) in Patras is δ2H = 6.8*δ18Ο-3.5, indicating evaporation at high humidity conditions (90 - 95%). The isotopic composition is strongly affected by westerly air masses but also by air masses coming from Eastern Europe. More depleted isotopic signatures are obtained for Medium Moving trajectories while the deuterium excess especially for air masses with terrestrial source history becomes greater than 18‰. In general higher d-excess values are observed for air parcels having crossed the middle of the Mediterranean basin. Chapter 3 examines the sub-cloud evaporation effect on precipitation in terms of its isotopic composition using three different isotopic approaches namely that of Meteoric Water Lines (MWLs), an isotope-evaporation model and a numerical isotope-evaporation model. The overall analysis is performed using isotopic data for various stations located in the Northern Hemisphere. The lowest MWL slopes are obtained for Madrid, Spain (αLMWL = 7.33) and Rehovot, Israel (αLMWL = 7.35) while in the greatest deviation from the equilibrium state is detected in Rehovot. Then an isotope-evaporation model is evaluated for the calculation of the isotopic enrichment in the falling raindrops induced by the sub-cloud evaporation effect. Both δ18O and δ2H show a linear relationship with the evaporated fraction with slopes ranging between 0.25‰/% and 0.28‰/% for 18O while for 2H the Δδ/EF values are around 0.9‰/%. The isotopic enrichment is highly correlated with the raindrop diameter and relative humidity conditions where Δδ increases with decreasing relative humidity and decreasing raindrop size. Temperature also contributes to the sub-cloud evaporation effect leading to higher isotopic enrichments for higher temperatures. However the temperature dependence is more distinguishable for smaller raindrops. Then, the vertical variation of the isotopic composition of falling raindrops and the corresponding isotopic enrichment is investigated through a numerical isotope-evaporation model. The isotopic composition becomes enriched in comparison to the initial isotopic composition at the cloud base level when the raindrops travel through drier and warm atmospheres, leading to Δδ18Ο up to 20‰ depending on the raindrop size and the initial meteorological conditions. For near saturated atmospheric conditions (RH = 95%) the isotopic composition does not vary significantly, indicating the absence of sub-cloud evaporation. Chapter 4 discusses the periodic patterns and the ‘temperature effect’ of δ18O in precipitation for various stations located around Central Europe, using periodic and wavelet models. The seasonal distribution of δ18O follows the temporal variability of air temperature providing seasonal amplitudes ranging from 0.94 to 4.47‰; the monthly isotopic maximum is observed in July. The Morlet Wavelet Transform reveals the main periodicity at 12-months where the wavelet power is mainly concentrated. Stations with limited seasonal isotopic effect depict a complex isotopic fingerprint that cannot be examined solely by the seasonality effect. In the case of the temperature dependencies on δ18O the isotope-temperature slope ranges from 0.11 to 0.47‰/oC with steeper values observed at the southernmost stations of the study area. Apart from the linear regression analysis the isotope-temperature slope is investigated under a spectral point of view. High coherencies are detected at the periodicity of 12-months. Generally the slope fluctuates around a mean value but in certain cases (sites with low seasonal effect) abrupt slope changes are derived and the slope becomes strongly unstable. The time-frequency slope is calculated at the annual periodicity mode ranging from 0.45 to 0.83‰/oC with higher values at stations that show a more distinguishable seasonal isotopic behavior. Vienna, Austria, depicts the most complicated coherency spectrum. Apart from the 12-month periodicity mode, inter-annual (bi-annual and decadal) periodicities are also represented with a decadal slope (1.9‰/oC) almost stable over the entire time period. In Chapter 5 the spatial patterns of the isotopic composition of precipitation in a global scale are briefly discussed. Global gridded isotopic (δ18O and δ2H) surfaces of precipitation through the Köppen-Geiger climate partitioning system (Köppen-Geiger, K-G precipitation ‘isoscapes’) are generated using a hybrid regression/geostatistical approach. The K-G ‘isoscapes’ capture the main isotopic effects as revealed by the Rayleigh distillation process with depleted values at higher latitudes (i.e. North America), at more inland locations (i.e. Siberia) and at high elevated areas (Andes, Alps and the Tibetan Plateau). However the isotopic grids fail to reproduce the isotopic depletion when moving from lower mid-latitudes to the equatorial regions since both the isotopic models and the spatially resolved residuals are unable to explain the isotopic composition in those regions where the high precipitation amounts and the strongly humid conditions are responsible for the observed isotopic depletions. The spatial isotopic uncertainty is quantified in terms of regression and interpolation uncertainties; high uncertainties are due to the isotopic regression models. The mean uncertainty values are equal to ±1.53‰ and ±13.1‰ for δ18O and δ2H respectively. Low uncertainties are associated to adequately sampled regions while high values are obtained for high-elevated regions and in areas with absence of isotopic measurements. The high correlations between the K-G ‘isoscapes’ and GNIP observation reveal the successful isotopic representation of the gridding procedure. Furthermore the generated isotopic grids are compared against the isotopic products from iso-GCMs. Significant spatial isotopic differences among the K-G ‘isoscapes’ and ECHAM, MIROC free models are investigated with negative differences in the Northern-Western and the Central-Eastern parts of Africa. On the other side positive differences are depicted in southern Iran, Pakistan and north India for both isotopic species and iso-GCMs.



Stable isotopes, Atmospheric processes, Precipitation, Water vapor, Sub-cloud evaporation, Periodicity analysis, Isoscapes, Koppen-Geiger classification system