1.3. Sampling and analyses of ground-based vapor
In the context of isotope analysis instrumentation, there are two well-known options, namely, (1) mass spectrometry and (2) laser absorption-based spectrometry for measuring isotopic composition of water and vapor (Costinel, Grecu, Vremera, & Cuna, 2009; Gupta, Noone, Galewsky, Sweeney, & Vaughn, 2009). Among these two instruments, the CRDS (Cavity Ring Down Spectrometer) is field deployable and portable instrument, which can measure both concentration and isotopic composition of ambient water vapor online. However, it is not cheap enough to be deployed at large number of locations for wider spatial coverage with high sampling density. In the context of sampling method, cryogenic trapping of water vapor from ambient air stream pumped at low flow rate is the standard method of sampling vapor for isotopic analyses (R. D. Deshpande et al., 2013; Purushothaman et al., 2014). However, pre-requisite of lower flow rate to avoid fractionation during sampling necessitates longer sampling duration (typically 4-5 hours) for collecting about 15 ml of liquified vapor, and availability of electric power. The sampling duration is even longer in arid regions where absolute moisture content in the air is very low. Similarly, pre-requisite of optimally lower temperature (~ -78 °C) to ensure complete cryogenic trapping necessitates availability of cooling options such as liquid nitrogen-alcohol mixture or dry ice. Thus, power supply, liquid nitrogen/dry ice and the long time are the essential pre-requisites for conventional sampling procedures which restrict the sampling of atmospheric water vapor only where these facilities can be made available.
To overcome these instrumentation and sampling limitations, it was felt necessary particularly in the developing countries, to develop a novel method of sampling ambient water vapor for its isotopic composition which can be replicated at numerous locations across different geographical regions with minimal logistic requirement.
Towards this, a novel, simple and cost-effective sampling method was developed as part of a National Programme for Isotope Fingerprinting of Waters of India (IWIN National Programme) (R. Deshpande & Gupta, 2008, 2012). In this method, ambient water vapor in condensed rapidly on an ice-cooled metallic surface (R. D. Deshpande et al., 2013). However, condensation of water vapor from ambient air on metallic surface at 0°C involves kinetic fractionation due to differential diffusivity of isotopic molecular species through supersaturated boundary layer. Therefore, the liquid condensate from ambient vapor is isotopically different from vapor. The isotopic difference between the vapor and liquid condensate has been explained by R. D. Deshpande et al. (2013) in terms of kinetic fractionation under supersaturated environment, similar to that explained by Jouzel and Merlivat (1984) for solid condensation. However, the isotopic composition of vapor cannot be computed from the liquid (or solid) condensate using these models because of uncertainties in estimating effective degree of supersaturation and diffusion coefficient for isotopic molecular species. Ignoring these limitations and in spite of the fact that isotopic composition of liquid condensate does not reflect true isotopic composition of ambient vapor some of the recent studies have used isotopic composition of liquid condensate to infer hydrological processes (Krishan, Rao, & Kumar, 2012; Saranya et al., 2018). Some studies have used simple linear regression to estimate isotopic composition of vapor from that of liquid condensate, though with large uncertainty (Purushothaman et al., 2014).
With a view to advance the usability of isotopic composition of liquid condensation as a close approximation of isotopic composition of ambient vapor, we present a new non-linear regression approach to compute the true δ18O value of ground level vapor using measured δ18O value of liquid condensate. Significance of this approach lies in the fact that, in contemporary hydrology, it is very important to monitor ground level water vapor in order to track the transport of water mass from surface or subsurface to atmosphere in different geographic areas differing in their weather, water resources, lithology, ecology and microclimate. This approach can be easily employed to monitor ground level water vapor in any remote area with minimal resources and infrastructure. This approach will therefore be very useful for coordinated network programs in developing countries.