Summary and Conclusion

Isotope tagging of ground level water vapor is very important to understand hydrological processes in different geographical regions having varied eco-hydrological, agro-climatic and water resource situations.
There are limitations and constraints in obtaining reliable isotopic signatures of ground level water vapor, related to infrastructure and analytical capabilities for ground based or remotely sensed observations. In this study a simple, cost efficient and novel methodology is discussed in which ambient water vapor is collected by liquid condensation on ice-cooled metallic surface, and using a non-linear regression model true isotopic composition of ground level water vapor is computed from measured isotopic composition of liquid condensate.
The non-linear regression model in this study is based on a known kinetic fractionation process during solid and liquid condensation under supersaturated environment, discussed respectively by Jouzel and Merlivat (1984) and R. D. Deshpande et al. (2013). This non-linear regression model is necessary because the physical parameters involved during kinetic fractionation, namely, effective degree of supersaturation, actual condensation temperature and diffusivity coefficients for heavier and lighter isotopologues cannot be determined precisely for the liquid condensation. Therefore, without the proposed non-linear regression model it is not possible to correctly estimate the true isotopic composition of ambient water vapor from that of liquid condensate, using the known theory of kinetic fractionation under supersaturated environment.
The non-linear regression model can estimate the isotopic composition of ground level water vapor far more accurately (±1.8308‰ for δ18O at Ahmedabad) compared to the best available remotely sensed data. Moreover, the method of non-linear regression is successful for samples collected from three different climate zones assuring that it is geographically invariant. The ground level water vapor sampling method of liquid condensation on ice-cooled surface can be easily adopted anywhere and at a very low cost. This paves the way for a cheap and yet reliable method of sampling ground level water vapor which can be extensively carried out at any location with limited infrastructure and resources. The data obtained can be further used to calibrate and strengthen the remotely sensed data as well as to verify the input and output from isotope enabled atmospheric general circulation models. This method can be particularly useful for developing nations with limited infrastructure and resources for fundamental research.