A landscape approach in the isotopic modeling of natural precipitations: two cases in Mediterranean mountain areas

Riassunto

The present paper proposes a method to simplify the very complex isotopic fractionation processes occurring during the water cycle. The method is constrained by a relatively small number of variables, with the precision needed in hydrological applications. After a theoretical introduction on the adopted interpolation criteria, two cases in the Mediterranean are presented. In both cases the evaluation of the “geometric complexity” of the systems appears to be the best tool to produce reliable isotopic models. If the complexity is low, it is apparently easier to fit different models; on the contrary the higher the complexity is, more difficult it is to find a reliable model but, at the same time, more difficult it is to find effective alternative models.
1. Introduction The isotopic signature of natural waters, expressed in terms of δ18O and δD, is an important information for understanding the complex relationships between meteoric waters and aquifers [Gonfiantini, 1978]; in particular, in hydrological applications, the comparative analysis of the average δ18O and δD values in meteoric precipitations and underground waters allows to identify the actual catchment basins that feed the aquifers. If contributions from outer hydrogeological structures, and/or re-evaporation of infiltrated waters due to high temperature/low humidity climatic regimes are negligible, the average isotopic composition of underground waters is related to the isotopic signature of precipitation by the following relation: δuwmrt (18O,D) = Σ(ss=1,mrt) δpss(18O,D) *(Pss/Pmrt); [1]
2.Theoretycal background of proposed spatial modelling procedure
where δuwmrt is the average isotopic composition of underground waters calculated on period corresponding to the mean residence time (mrt) of the considered aquifers, δpss the isotopic composition of precipitations in a single sampling session (ss); the weighing factor Pss/Pmrt is the ratio between the amount of precipitations occurred in a single sampling session and the heights of the whole mrt period. In a typical hydrogeological application mrt is at least 1 year, due to the natural frequency of variations in atmospheric phenomena, and ss is typically 1 month, due to practical and economic reasons. The main question when equation [1] is applied to hydrological models is: how great is the error in extending to the whole catchment
The isotopic signature of meteoric waters has been previously expressed both in terms of...