Temperature and precipitation effects on the isotopic composition of global precipitation reveal long-term climate dynamics.

Earth's climate history is traced through the long-term covariance between the isotopic (δ 18O) composition of archived meteoric waters (groundwater, ice cores) with air temperature (T) and amount of precipitation (P). To assess recent multi-decadal climatic changes, we analysed δ18O, T and P, and the relationships between these parameters at 20 stations having 60 years of continuous monthly isotopic records. Using nonparametric regressions and time series modelling we found significant linear and non-linear relationships for δ18O with T and P and showed that the δ18O dependency on these two parameters varied over decadal scales, thereby revealing complex relationships related to recycled moisture, large-scale convective processes and atmospheric-oceanic oscillations. Due to multiple factors controlling the δ 18O composition of precipitation including P and T effects, we found that time-varying relationships between δ18O in precipitation P and T were better explained using the non-linear regressions. Our results affirmed that δ18O distributions in global precipitation are integrative indicators of climate dynamics whose patterns can be applied to better understand region-specific climatic changes in the present, past, and future.

A critical assessment of widely used techniques for nitrate source apportionment in arid and semi-arid regions.

The assessment of nitrate pollution origin using stable isotope techniques is a fundamental prerequisite for the application of sustainable groundwater management plans. Although nitrate pollution is a worldwide groundwater quality problem, existing knowledge on the origin of nitrate pollution in arid and semi-arid regions is still scarce. Using the example of the Grombalia aquifer (NE Tunisia), this work summarizes the main strengths and constraints of multi-isotope techniques targeting at nitrate source identification and apportionment The results highlighted that, even in the case of well-established methodologies, like those of isotope hydrogeochemistry (δ15NNO3, δ 18ONO3 and δ 11B) and mixing modelling for source apportionment, it is fundamental to take into account regional and local end-members to avoid biased data interpretation and to fully exploit the potential of such accurate tools.