1.1. Satellite Remote Sensing and Radiosonde
The Tropospheric Emission Spectrometer (TES) on board the Aura spacecraft of NASA has provided data which provides useful insight into the mechanism governing the recycling of water vapor (Beer, Glavich, & Rider, 2001; John Worden et al., 2006). The Atmospheric Trace Molecular Spectroscopy (ATMOS) experiment, a NASA, JPL instrument flew in a space shuttle and monitored the stratospheric and upper tropospheric isotopic composition of vapor providing evidence of large-scale transport, condensation and convective processes (Rinsland et al., 1991). It has been observed that around 20% of rainfall typically gets evaporated and fed back to the source whereas the value can go up to 50% in case of convective clouds (J. Worden, Noone, Bowman, Tropospheric Emission Spectrometer Science, & Data, 2007). We do not yet completely understand the variation of factors such as relative humidity and its corresponding effect on the isotopic composition of ground level water vapor (Bosilovich, Schubert, & Walker, 2005; Kevin E Trenberth, 2002; Kevin E. Trenberth, Dai, Rasmussen, & Parsons, 2003).
However, the satellite-based methods have certain shortcomings. The major limitation in the airborne measurement technique comes from the fact that the isotope values are calculated based on the average values of the 550-800 hPa mean layer. The isotopic values in these layers are susceptible to the maximum variations and are extremely sensitive and the estimated δD values have a precision of around 10‰ in the tropics and 24‰ in case of the poles (John Worden et al., 2006). Moreover, the isotopic abundance ratio HDO/H2O suffers due to spectral interference of H2O on HDO due to pressure broadening. To add to this the space borne interferometers and spectrometers suffer from serious limitations in resolving thermal emissions in the short-wave infrared range thus failing to provide much insight into the composition of lower tropospheric vapor where much of the dynamic processes within the hydrological cycle take place.
In the past three decades water isotopes have been incorporated into the Atmospheric General Circulation Models(AGCM) and has provided valuable insight into the underlying processes involved in surface evaporation, condensation, super saturation, vertical distribution of isotopes etc. (Xi, 2014). Thus AGCMs numerically represent and interpret the physical processes that may affect isotopic variation in vapor .But these AGCMs encounter certain limitations that stem from the General Circulation Model(GCM) such as biases in temperature or precipitation simulation (Noone & Sturm, 2010). AGCMs work based on numerical simulations using certain approximations which fail when a sharp boundary is encountered such as for example the Tibetan Plateau (Yao et al., 2013). In addition to this the AGCMs face numerical inefficiencies when simulating moisture transport and lead to a deficient simulation by large scale motions. Most AGCMs simulate water vapor transport pathways by simulating the convective updrafts and downdrafts which involves solving the cloud microphysics using numerical integration which is computationally expensive.
Vertical distribution of deuterium isotopic composition, and aspects related to interaction between rain and ambient vapor have been studied using radiosonde. Radiosonde measures temperature, humidity and sometimes wind velocity at different pressure levels which can be further converted into upper tropospheric humidity quantities. (Adeyemi & Joerg, 2012; Li, 2003; Soden & Fu, 1995). Available experimental data on the vertical deuterium distribution in atmospheric water vapor show a steep decrease of deuterium with altitude in the lower troposphere (Rozanski & Sonntag, 2016). The limitations of this approach are that temporal variation at the geo-coordinates is not possible and the ground level vapor cannot be monitored. It provides only a snapshot observation which is inadequate to understand in details a dynamic system such as the lower tropospheric water vapor.