Stable Isotopes Anomalies of Storm Event in April 2019 in Southwestern Iran: A Method to Characterize the Origin of the Atmospheric Masses


 To determine the source of April 2019 storm event in south-western Iran, stable isotopic technique (2H and 18O) of precipitation were used. The unique event, with a mean of 230 mm/year, i.e. 46% of the annual rainfall over four days, caused large dams on Dez and Karkheh rivers to overflow and a great deal of casualties and financial losses. To determine the characteristics of stable isotopes 2H and 18O of this event, 43 samples of rain water were collected in the area about 41,000 (Km)2 in south-western Iran with altitudes between zero to 2814 m (m.a.s.l). The local meteoric water line (LMWL) was prepared during the storm event and compared with the global, Mediterranean and other parts of Iran as well as neighbouring countries. According to the LMWL equation (δ2H = 6.5996 δ 18O + 7.561) and d-excess (with average 7.85 ‰), the source of precipitation was the integration of the meteoric masses of Mediterranean and Red Sea. There was a low correlation between the altitude increase and the 18O decrease (0.13 per 100 meters), which is a characteristic of this storm event. This isotopic study has proven the impact of Sudanese mass on the occurrence of storm event in south-western Iran for the first time isotopically.


Introduction
Precipitation is the last event of vapour transport and phase change (Zhang et al. 2019;Cai etal.2016). The spatial distribution of stable isotopes of precipitation depends on different isotopic effects including continental effect, amount effect, and altitude effect (Dansgaard 1964;Kong et al. 2019). Craig (1961) described the relationship between these parameters in freshwater for the first time, known as the Global Meteoric Water Line (GMWL). It is expressed as equation 1 in equilibrium conditions and temperature of 25°C. This line is derived from worldwide isotopic precipitation data and averaged over numerous local meteoric water lines (LMWLs) that each has a different slope and intercept (equation 1).
High values of stable δ 2 H and δ 18 O isotopes are in areas close to where the moisture originates and lower values are far from source moisture (Kong etal. 2019). The linear relationship between δ 2 H and δ 18 O is one means to assess the cumulative effects of fractionation caused by evaporation and condensation in samples of precipitation (local meteoric water line, LMWL) and surface/shallow groundwater (local water line, LWL) as compared to the global equilibrium between evaporation and condensation (global meteoric water line, GMWL) defined by the relation 1 (Craig 1961;Ambach 1968): LWLs may deviate from the GMWL during any non-equilibrium event, such as evaporation, mixing, or additional input of marine moisture. For example, LWLs with slopes less than or greater than 8 can indicate systems dominated by evaporation and recharge/recycled moisture, respectively (Craig 1961). During the evaporation process, isotopic fractionation causes a difference in the relationship between δ 2 H and δ 18 O, so that the δ 2 H values in the vapour decrease faster than those of δ 18 O and therefore the samples are positioned above the GMWL. In contrast, the residual water in the source will be less than the GMWL. Dansgaard (1964) called this difference d-excess d-excess = δ 2 H -8.0* δ 18 O Which has an average of 10 ‰ on a global scale, revealing changes in sources of moisture for precipitation. D-excess in precipitation is due to kinetic subtraction during the evaporation process, which is mainly influenced by the relative humidity and temperature of moisture source region (Merlivat et al. 1979;Jouzel 1997) and remains relatively constant during the Rayleigh rainout event (Gat 1996); therefore, it can detect changes in the source of moisture which precipitation originated from it (Florea et al. 2017). Dexcess values deviation from GMWL (higher or lower than 10 ‰) shows the effect of air masses source, humidity, temperature, and evaporation (Gat 1996;Lee et al. 2003).
By analyzing δ 18 O, δ 2 H and d-excess in regional precipitation we can determine the source of precipitation moisture mass (Sodemann et al. 2008;Jouzel et al. 2013;Wang et al. 2016 a, b, c;Lawrence et al. 1982) , monitored patterns of atmospheric masses rotation (Hoffman et al. 2000;Birks et al. 2002), estimate water evaporation (Wang et al. 2016c;Uemura et al. 2012;Jasechko et al. 2014Jasechko et al. , 2013, restore the past climate conditions (Thompson et al. 2013;Mariani et al. 2014), and identify the surface and groundwater recharge area through the precipitation process and their correlations (Telmer et al. 2000;Gibson et al. 2002;Yonge et al. 1989;Longinelli and Selmo 2003). However, the isotopic composition of precipitation varies temporally and spatially as a result of depletion due to temperature, precipitation, distance from the coastline, altitude effect and climatic conditions (seasonal effect) (Siegenthaler and Oeschger 1980;Windhorst et al. 2013;Gat 1969;Dansgaard 1964). Precipitation isotopic composition changes with altitude, which is called altitude effect and is a major criterion for studying spatial variability, especially in mountainous basins with high variations in elevations. And this has been investigated by many researchers (Cortes et al. 1997;Gonfiantini et al. 2001;Kattan et al. 2006;McGuire et al. 2005;Peng et al. 2010;Siegenthaler and Oeschger 1980;Yurtsever and Gat 1981;Vogel et al. 1975). Most researchers have reported δ 18 O and δ 2 H changes from −0.5 to −4 % per 100 m and −0.1 to −0.6 % per 100 m, respectively.
Iran is in undesirable situation compared to the average world due to being located in arid and semi-arid region. Although in the southwest of Iran, including Khuzestan province, major rivers such as Karkheh, Dez, Karun, and Zohreh rivers supply a large part of the water requirement, but many parts of this province have not been in the path of these surface waters and supply needs to come from groundwater resource. Therefore, changes in these resources and recent droughts have a direct effect on the lives of people in these areas. The occurrence of drought and its continuance certainly affect the quantity and quality of groundwater resources. Due to the drought occurring in the last 10 years, along with the reduction in surface water, discharge of groundwater resources has been progressive drastically reducing these resources, especially in the northern parts.
The occurrence of 10-year droughts from 2007 to 2016 in the southwest Iran and the beginning of the water scarcity crisis, prediction the source of precipitation in the region is becoming increasingly necessary using precipitation data. This study was conducted to identify and trace massive pulse in Khuzestan province (located in southwest Iran) water resources by determining the isotopic characteristics of unprecedented storm event in southwestern Iran in April 2019 causing a great deal of casualties and financial losses. In southwestern Iran, the air masses originate from the North Atlantic and Mediterranean Sea and the collision of these masses with the Sudanese tropical masses originate from the Red Sea and the Persian Gulf cause heavy rainfall. Most of the area's rainfall occurs in late fall and early spring. However, at first, this storm event has isotopically been associated with the re-formation of the low-pressure Sudanese or Mediterranean atmospheric masses. Given that no such unprecedented rainfall has ever occurred in southwestern Iran, it is important to determine its isotopic characteristics in order to know its origin, storm forecasting based on mass formation and movement, and water resources management.
In this paper stable isotopic technique ( 2 H and 18 O) was used to determine the variation of isotopic composition of precipitation with altitude (altitude effect), to define the local meteoric water line in southwestern Iran (storm event at April 2019) and to investigate the spatial distribution of stable isotopes (Zagros Mountain Range) and deuterium excess.
Since the isotopic patterns depend on the origin of the moisture mass, the local meteoric water line was determined and compared with the global meteoric water line and other parts of Iran and neighboring (western) countries. The relationship between the isotopic values and amount of precipitation (amount effect) has been investigated too. Finally, the origin of the storm event was determined (April 2019). This study on the origin of heavy storms has been performed for the first time in the Middle East.

Study area
The study area with an area of 41,000 (Km) 2 is located in south-western Iran between latitude 30°30′ to 33° 30′ N and between longitude 48° to 51° E. Most of the study area is located in Khuzestan province and parts of itborder with Lorestan and Chaharmahal and Bakhtiari provinces.  and ±0.8‰, respectively.
where δ sample is the isotope ratio of the samples to V-SMOW, R sample is the ratio of 2 H/H  Table 1.
[ Figure 2 is about here] In order to delineate the spatial distribution maps of precipitation and the values of δ 2 H and δ 18 O as well as deuterium excess, the Kriging interpolation method was used. Hatvani et al. (2017) believe that, in this method, the distance and degree of variation between the known points are considered and then the unknown points are estimated using the weighted average of adjacent points. Among the Kriging interpolation methods, the ordinary method was chosen because it provides the best linear distribution for a suitable regional view. [

Spatial distribution of stable isotopes
According to the results summarized in Table 1 [ Figure 3 is about here]

Deuterium excess (d) and determination of moisture source
Deuterium excess (d) varied from -9.3 to 15.6 ‰ in 43 rainwater samples, of which 15 samples were more than 10 ‰ and 27 samples less than 10 ‰. The average was 7.85, which is lower than the global d-excess (10). This indicates that most of the precipitation originate from areas with high evaporation and that the rain is due to the mixture of water vapour from the Mediterranean and the Red Sea atmospheric masses. In the near vicinity of the dam reservoirs, the d-excess is near 10 ‰ and more.
In the study area, from south and southwest to higher altitudes (north and northeast) the d-excess values increase (Figure 3 c). Considering that d-excess decreases with the enrichment of the isotopic composition and increases with the depletion of the isotopic composition, it can be concluded that the isotopic composition is depleted towards the altitudes of the study area (north and northeast).

Isotope effects due to precipitation amount (amount effect)
Figure 4 depicts of precipitation map of storm event in April 2019. The amount of precipitation has increased from south to north and northeast increasing to more than 520 mm in the northern elevations of the study area. With an increase in rainfall, δ 18 O and δ 2 H values have decreased. The decrease in δ 2 H is more pronounced than that in δ 18 O, so that the dispersion of δ 2 H values is high. Therefore, with an increase in rainfall, the δ 18 O and δ 2 H isotopes are depleted. The correlation between δ 18 O and precipitation amount (R 2 = 0.0957) is stronger than the relationship between δ 2 H and precipitation (R 2 = 0.0708), which is probably due to the storm event and mixing of Sudanese and Mediterranean atmospheric masses at higher altitudes; therefore, the effect of precipitation amount (amount effect) is disrupted.
[ Figure 4 is about here]

Relationship between the δ 18 O and δ 2 H values and altitude (altitude effect)
The altitude in the southern and margins of the Persian Gulf is low and close to zero and increases to the north and northeast. The maximum elevation in the northeast of study area is 4030 meters. Figure 5 illustrates the relationship between the values of δ 2 H, δ 18 O, and d-excess with altitude.
[ Figure 5 is about here] The equation for the increase in δ 18 O with altitude in this study is as follows: [ Figure 7 is about here] [ Figure 8 is about here] [ Figure 9 is about here]

GMWL, LMWL and neighboring countries of Iran
Precipitation isotopic data (Table 1) are used to provide the local meteoric water line (LMWL) of the study area. LMWL is plotted from the relationship between δ 18 O and δ 2 H as seen in Figure 10. The equation obtained for the LMWL is as follows: δ 2 H = 6.5996 *δ 18 O + 7.561 ‰ (7) [ Figure 10 is about here] It is clear that the slope of southwestern Iran LMWL is slightly slower than that of the GMWL, as the movement of the air masses toward the precipitation site produces rain that evaporates in the lower parts of the cloud. Figure 11 shows CMMWL; δ 2 H=6.5 * δ 18 O+5.28 ‰ According to the study of Ali et al. (Ali et al. 2015) Iraq meteoric water line is as follows: δ 2 H = 7.573 δ 18 O + 13.82 ‰ (10) The source of Iraq rainfall is a mixture of Mediterranean, Persian Gulf, and Atlantic waters (Ali et al. 2015). According to the results of Kattan (2006)  [ Figure 11 is about here] Considering the situation of sampling points in the study area, it is observed that some points are located between the global and Mediterranean meteoric line and with respect to the slope of the line less than the GMWL and the mean of d-excess (less than 10 ‰) it can be concluded that the source of the storm event is a mixture of the Mediterranean Sea and the Sudanese atmospheric masses which cross the Arabian Sea and the Persian Gulf.

Meteoric water line of other parts of Iran
According to studies in some areas of Iran, local meteoric water line of north (Shahrood and Gorgan), south (Shiraz), west (Kandand-Zahab-Karkheh basin), east (Rafsanjan), and central Iran (Tehran and Isfahan) have been provided. Table 2 and Figure 12 illustrate the equation and the LMWL for these areas.
[ Table 2 is about here] The average rainfall in the surveyed areas (299 mm) is similar to the average annual precipitation in the whole of Iran (250 mm). Given that the amount of precipitation is one of the important factors controlling precipitation isotopic properties, and because the average rainfall in the studied areas except Gorgan is close to the average rainfall of Iran, it can be concluded that the average line of these regions according to equation 13 represents the Iranian meteoric water line in dry regions. The slope of LMWL is between 5 and 8 (Shamsi and Kazemi 2014). In fact, the slope of Iranian LMWL is lower than that of the GMWL, which describes the prevailing dry and semi-arid climate in Iran. In such a climate condition, secondary evaporation of raindrops enriches heavy isotopes: [ Figure 12 is about here] Isotopically heavier water falls first during a precipitation event; thus, δ 18 O and δ 2 H values decrease during progressive precipitation from adiabatic cooling along an altitude gradient-an altitude lapse rate (Dansregard 1964;Clark and Fritz 1997). According to Clark and Fritz (1997) altitude rate is (-0.5%) to (-0.15%) per 100 m elevation. The relationship between isotopic values and altitude is shown in Figure 5. With increasing elevation, the isotopic values have decreased. Reduced rates were -0.13 and -0.91 per 100 m for δ 18 O and δ 2 H, respectively, based on the equations 4 and 5.
The term amount effect was presented by Dansgaard (1964), which showed a negative relationship between δ 18 O and precipitation. This effect, which primarily depends on the meteorological conditions at the time of rain fall (temperature and relative humidity), is usually due to several factors: (i) a decrease in the isotopic composition of the condensate in a cloud as cooling and raining proceed; (ii) quick equilibrium of small raindrops with ambient water vapour and temperature conditions compared to larger raindrops; and (iii) fast evaporation of small raindrops during their falling on the land surface compared to larger raindrops (Rozanski et al. 1993;Ali et al. 2015).
In the study area, with increasing rainfall, the decreasing of δ 2 H is steeper than that of δ 18 O. So the correlation between depletion of δ 18 O and precipitation (R = 0.0913) is stronger than the relationship between δ 2 H depletion (R = 0.0692), which is probably due to the storm event and the mixing of the Sudanese (Red Sea) and the Mediterranean Sea atmospheric masses at high altitudes and the disruption of the amount effect.
The typically low d-excess is due to re-evaporation in the cloud, which also results in the enrichment of heavy isotopes in precipitation. Lower d-excess values are always associated with low precipitation and high evaporation . Due to the effect of recirculation of moisture from local surface water such as dams and soil water, d-excess values increase (Froehlich et al. 2008;Warrier et al. 2010).
D-excess is an index of evaporation effect on the physico-chemical properties of water, so that if water evaporates, the d-excess is reduced (Tsujimura et al. 2007). The range of variation of d-excess of precipitation is -9.3 to 15.6 with an average of 7.85, which is lower than the global amount. This indicates that much of the region's rainfall (storm event) originates from areas with high evaporation.

Conclusion
In this study, the values of δ 18 O, δ 2 H, d-excess (

Disclosure statement
No potential conflict of interest was reported by the authors.   Location of sampling points and study area in southwestern Iran. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Distribution of (a) δ18O , (b) δ2H (c) d-excess in rainwater samples throughout the study area during the storm event in April 2019. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors. Map of iso-precipitation of the storm event (April 2019). Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 5
Correlation between altitude and isotopic values Digital Elevation Model (DEM) of study area and cross sections. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.  Local meteoric water line (LMWL) of some parts of Iran in present study