Application of Isotopic Signature of Atmospheric Vapor for Identifying the Source of Air Moisture-An Example from Roorkee, Uttarakhand, India

Stable isotope hydrology received a big boost with the identification of the co-variance of hydrogen and oxygen isotopes in precipitation first by Friedman, 1953 [1] and later in the form of GMWL by Craig, 1961 [2]. A continental effect from as large as 45‰ per 100 km in δD on North coast of California to as little as 2‰ across the Great Basin in Northern Nevada was observed [3]. Values of elevation effect have been reported to be approximately 40‰ in δD per 1000m increase on the west slope of the Sierra Nevada [4]. Dansgaard, 1964 [5] reported a good world-wide correlation for coastal and polar stations between mean δD and δ¹⁸O values of precipitation and the mean annual air temperature (MAAT) with an average decrease of >5.6‰ for δD and 0.7‰ for δ¹⁸O per degree Celsius. These observed variations are consistent with temperature dependence of equilibrium fractionation factor again indicating that condensation is an equilibrium process. The precipitation progressively decrease from 18O values between -2 ‰ to -4 ‰ [6].

The monsoon is a large scale wind circulation phenomenon that brings large amount of moisture on the Indian sub-continent during the period mid June to September. Earlier studies on monsoon dynamics mainly dealt with deviation of total precipitation from the normal, spatial variability in precipitation, extreme events and timing of onset of monsoon [7-9]. Most of these studies relied on conventional meteorological parameters such as temperature (Continental and sea surface), relative humidity, wind speed, wind direction, cloud pattern etc. However, sufficient accuracy in onset of monsoon, its dynamics on the continent and the timing of its withdrawal is not well understood. The complexity of the phenomenon is due to link between mixing of sub-regional moisture with continental scale moisture arising directly from Arabian Sea and/or Bay of Bengal and/or Indian Ocean. Using the isotopic signatures of water molecules, it is possible to track the movement of air moisture. This is due to fact that with the onward movement of the air moisture heavy isotopes get systematically depleted and the recycling of moisture due to condensation and subsequent evaporation cause a difference in relative enrichment in deuterium and oxygen isotopic composition.

In India, the Southwest (SW) monsoon begins from Lakshadweep, Minicoy and Kerala [10,11] and it enters North India from Arabian Sea and Bay of Bengal. During the winter and springs, Northeast (NE) monsoon is prevalent in the Southeast (SE) parts of India, while the north and Northwest (NW) part of the country receives rains due to the Western Disturbances [10-14]. Thus the sources of precipitation/atmospheric vapor in India are mainly Arabian Sea and Bay of Bengal in Monsoon season. But during non-monsoon seasons the sources of precipitation/atmospheric vapor are local/regional derived vapors.

The stable isotopes of oxygen and hydrogen (¹⁸O and ²H) can be used for tracking the sources of atmospheric vapor/precipitation due to the systematic mass-dependant partitioning of the three isotopomers ¹H¹H¹⁶O, ¹H²H¹⁶O and ¹H¹H¹⁸O among solid, liquid and gaseous phases as water passes through the hydrological cycle.

The studies [15-21] were carried out in different parts of the World on the isotope data of surface vapor have a limited data and is not sufficient to adequately explain the isotopic composition of atmospheric water vapor. No attempt has been made so far to identify the source of atmospheric vapor/precipitation at Roorkee, Uttarakhand, India using isotopic signature.

Therefore, the present study is carried out for identifying the sources of atmospheric vapor/precipitation at Roorkee using the isotopic signatures and the isotopic data has also been correlated with the wind trajectory and satellite data obtained from India Meteorological Department, New Delhi to establish the similarities among these.

Study area

The studies were carried out in the National Institute of Hydrology, Roorkee (Uttarakhand), India. The study area falls under latitude 29°52’, longitude 77°53’ and altitude 268 m. The meteorological parameters of Roorkee are: normal rainfall is 1156.4 mm per annum; the monthly average maximum temperature is 28.8°C; monthly average minimum temperature is 15.5°C; average relative humidity is 78%.

Air moisture sample collection

The condensed atmospheric vapour/ground level vapour (GLV) samples for isotopic analysis are collected (2008-09) daily from 9.30 am to 11.00 am at Roorkee, Uttarakhand, India by condensation method. The parts used in this methods are shown in Figure 1 [22-23].

Figure 1: Condensation device.

Rain sample collection

For rainwater a specially designed sampling system developed by the International Atomic Energy Agency (IAEA) is used to collect rain water samples on event basis. The system ensures amount weighing and negligible post-precipitation evaporation of the collected sample during the interval between rainfall and collection. The date and time, and temperature and relative humidity are recorded [24].

Sample analysis

The isotopes in GLV were analyzed using GV-Isoprime Dual Inlet Isotope Ratio Mass Spectrometer (Figure 2) [25].

Figure 2: Dual Inlet Isotope Ratio Mass Spectrometer.

Identifying dominant sources of seasonal vapour influx to India is vital to improve understanding of (i) continental-scale hydrological circulation and (ii) land- ocean-atmosphere interaction as Indian monsoon is an important component of global atmosphere circulation and source of all important water resources in India.

Isotope characterization of source of GLV

The interpretation of the isotope data of GLV for the years 2008- 09 indicates that the seasonality effect divides the entire data spectrum into three sub-units as: monsoon water vapour (δ¹⁸O<-25, δD<-126) of Bay of Bengal (BoB)/oceanic origin are the most depleted; the most enriched vapours (δ¹⁸O>-5, δD>-6) of local origin and the vapours originating due to Western Disturbances (WD) have values (δ¹⁸O~- 15, δD~-53) between these two sources. The minor depletion in the isotopic composition of water vapour in the February-May period is due to insurgent of moist air as a result of Western Disturbances [10]. However, this disappears after May, with the arrival and masking by strong monsoon winds from BoB. Between July to September, the atmosphere is pre-dominantly occupied by the monsoon vapours. After September, with the departure of monsoon, the isotopic composition of atmosphere gradually regains its local moisture source value (δ¹⁸O>-5, δD>-6). The cycle then repeats again for the next year. The depletion in the isotopic composition of air moisture provides date of onset of monsoon. Any time during this cycle, the relative proportion of local moisture to regional inflow moisture can be estimated from the relative contribution from the end member source composition (Monsoonal: (δ¹⁸O~-25, δD~-126); Local (δ¹⁸O~-5, δD~-6). The isotopic characteristic of end member sources and the general isotopic characteristic due to admixture of these two sources are shown in the Figure 3.

Figure 3: Isotopic characteristic and the average isotopic composition of monsoon vapour, local vapour and the mixed vapour at Roorkee.

Monsoon vapour: δD=6.29δ¹⁸O+32.80; R²=0.88

Local vapour: δD=4.83δ¹⁸O+28.17; R²=0.80

W.D. vapour: δD=5.38δ¹⁸O+30.38; R²=0.73

Isotope characterization of precipitation

Uttarakhand receives 81% of its rain from the SW monsoon, spread over the period from mid of June to mid of October. Of the total 19% remaining, about 15% is received in almost equal distribution in the months January, February, March and May and remaining 4% as occasional rains during December and April. The premonsoon rains (March to 1^st week of June) are considered to be due to Western Disturbances. Depending on the source of moisture and the meteorological parameters it is expected that these rains may differ in their isotopic characteristic. In order to evaluate this, the isotopic data on rainfall collected over the period 2008-2009 are divided and separated into three periods as; monsoon (mid of June to mid of October, Winter rains (December to February) and pre-monsoon showers (March to 1^st Week of May). The isotopic data spread over these three periods and their systematics is shown in the Figure 4.

Figure 4: Isotopic characteristic of precipitation during monsoon (mid of June to mid of October), winter (December to February) and pre-monsoon season (March to 1st week of June).

From the Figure 4, it can be seen that the monsoon showers are isotopically the most depleted with the highest slope (7.78) of its characteristic line followed by winter rains which are marginally enriched (Table 1). The pre-monsoon showers are the most enriched rains in the year. With these systematic it can be easily differentiated between pre-monsoon and monsoon rains and in identifying the true monsoon rains or onset of monsoon. The isotopic results on the atmospheric moisture and rains also indicate that the delay in monsoon or drought can be observed from the enriched isotopic composition of rain and atmospheric moisture and that the system comes down the depleted levels due to monsoon showers [6].

Table 1: Isotopic characterization of air moisture and rain.

Isotope and wind trajectory

The wind trajectory data can be used for interpreting the isotopic data of the GLV. The wind trajectory was selected on the basis of the highest depletion and the enrichment in the isotopic composition during a time spectrum analysis. Due to the Westerlies wind source (Figures 5a, 5c and 5f), the isotopic data of GLV starts depleting from δ¹⁸O= -8.39‰ (Figure 5a) to δ¹⁸O=-12.20‰ (Figure 5c) and it further depletes to a maximum of δ¹⁸O= -14.71‰ (Figure 5f) from June to September. Similarly, the same trend is observed in the winds coming from the Arabian Sea, where the isotopic valuesof δ¹⁸O depletes from -24.67‰ (Figure 5b) to -7.79 ‰ (Figure 5e). This is due to the Seasonal effect [10].

Figure 5: Use of Wind Trajectory Data for interpreting the isotopic data of GLV.

Isotope, wind trajectory and water vapour data

The isotope data was correlated with the wind trajectory and water vapour data (Figure 6). The atmospheric vapour/air moisture from the local/regional source is enriched (δ¹⁸O= -11.18‰; Figure 6a) as compared to atmospheric vapour/air moisture originating from Arabian Sea & Bay of Bengal (δ¹⁸O= -18.94‰; Figure 6b). This is due to the Continental effect [10].

Figure 6: Correlation of isotopic data with water vapour data and wind trajectory.

The variations in the isotopic values of δ¹⁸O of atmospheric air moisture during the different seasons of the year indicate the different sources of atmospheric air moisture. The interpretation of the isotope data of GLV for the years 2008-09 indicates that the seasonality effect divides the entire data spectrum into three sub-units as: monsoon water vapour (δ¹⁸O<-25, δD<-126) of Bay of Bengal (BoB)/oceanic origin are the most depleted; the most enriched vapours (δ¹⁸O>-5, δD>-6) of local origin and the vapours originating due to Western Disturbances (WD) have values (δ¹⁸O~-15, δD~-53) between these two sources. The minor depletion in the isotopic composition of water vapour in the February-May period is due to insurgent of moist air as a result of Western Disturbances. The isotopic results on the atmospheric moisture and rains also indicate that the delay in monsoon or drought can be observed from the enriched isotopic composition of rain and atmospheric moisture and that the system comes down the depleted levels due to monsoon showers. However, the sources of the vapour can be more clearly understood by using the wind trajectory data or the Kalpana satellite water vapour data.

Isotopic signatures of GLV can be used to distinguish the source of atmospheric moisture as well as to estimate the contribution of the monsoonal and local vapours. With further analysis in future, semi-quantitative estimation of air-moisture flux transport can be provided. The proposed isotopic investigations are specially designed to understand such land-ocean-atmosphere interactions a Department of Science and Technology, New Delhi sponsored national programme entitled “National Programme on Isotope Fingerprinting of Waters of India (IWIN)” is being carried out by 14 institutes/organizations including National Institute of Hydrology, Roorkee with the main objectives, to study the components of the local and regional hydrological cycles, their seasonal evolution, interactions between the various components and controls exercised by geographic factors and climatic forcing. This programme is considered vital for predicting the consequences of the impending engineered modification of the hydrological cycle of the country necessitated by increase in population and the need to rapidly develop and exploit the water resources of the country on a massive scale.

While this study will generate considerable new knowledge/ information/data, its importance lies in how best this new data could be advantageously used to facilitate water resource development and management of the country more efficiently particularly for irrigation. The increasing water vapor content of the atmosphere, particularly during non-monsoon months will also affect the hydrological cycle. In this context, the data collected or generated during the study could form a baseline against which it will be possible to monitor the impacts of anthropogenic interventions on local/regional hydrological cycle at later date.

The work was carried out under the project “National Programme on Isotope Fingerprinting of Waters of India” and authors are thankful to DST-SERC (Funded by DST vide IR/54/ESF/05-2004 dated July 17, 2007) for sponsoring the study. Authors thank Director, National Institute of Hydrology for constant support and encouragement. Authors are also thankful to the unknown reviewers for their valuable and constructive comments. The wind trajectory data received from NHAC-NWP, IMD-Delhi and satellite data from Satellite Meteorology Department, IMD, New Delhi is also duly acknowledged.