Background & Summary

Groundwater currently provides more than 36% of the world’s drinking water and 42% of irrigation water use1. Additionally, more than two billion people worldwide rely on groundwater for drinking, sanitation, and irrigation2. This dependence is exacerbated in arid and semi-arid regions, such as Iran, where groundwater supplies around 63% of the country’s drinking water and serves as the sole source of water in certain major cities and many rural areas3,4,5,6. Therefore, a clear understanding of the hydrogeological processes in national aquifers is required for conserving groundwater resources and limiting potential contamination pathways. In this context, water stable isotope-based techniques can be used to quantify and elucidate the key hydrogeological processes underlying the changes in both groundwater quantity and quality7.

Stable oxygen (δ18O) and hydrogen (δ2H) isotopes serve as environmental tracers in hydrogeology studies, allowing researchers to investigate groundwater recharge processes8, water flow paths9, and residence times10, as well as to quantify the source and fate of pollution11 and soil moisture12. Furthermore, the ratios of deuterium (i.e., δ2H) to δ18O can reveal the source of water and precipitation, as well as evaporation rates13. These isotopes can also be used in groundwater models via data assimilation to effectively manage groundwater supplies and undertake hydrological studies14.

Unlike international standards and the successful experiences of countries with similar (semi)arid climate—such as some Middle Eastern countries, North African nations (e.g., Morocco and Egypt), parts of China, and the United States—Iran lacks a regular, scheduled, and extensive groundwater sampling network for stable isotopes to provide homogeneous and reliable isotopic data. Globally, organizations such as the International Atomic Energy Agency (IAEA) and the Global Network of Isotopes in Precipitation (GNIP) have established precise standards for isotopic sampling, analysis, and data exchange, which are applied in countries with similar climatic conditions. These frameworks include carefully locating stations across various regions, conducting regular sampling (typically monthly or seasonal), and employing advanced analytical technologies to produce continuous, high-quality data that are crucial for long-term hydrological analyses and sustainable water resources management. As a result, groundwater isotopic studies are mostly sporadic and confined to specific geographic areas in Iran15,16,17,18,19,20,21,22,23,24,25, resulting in insufficient and inconsistent data. Additionally, the groundwater stable isotope data used in the related conducted studies are not publicly accessible to international researchers.

In this research, we aimed to provide a dataset of δ2H and δ18O in 10 main aquifers in Iran. Moreover, the isotopic lines associated with these 10 aquifers were calculated and compared with the global meteoric water line (GMWL)26. Additionally, we calculated the deuterium excess (D-excess) and line-conditioned excess (LC-excess). D-excess and LC-excess give insights into the evaporation conditions at the vapor source and the evaporation signal in groundwater wells normalized to precipitation, respectively27,28. To our best knowledge, this openly accessible dataset is the first of its kind in Iran to include stable isotopes (δ¹⁸O, δ²H), D-excess, and LC-excess across main aquifers, supporting integrated analysis of groundwater, atmospheric inputs, and anthropogenic effects.

Methods

Site description

Iran is located in southwest Asia, between 24° and 40° N in latitude and 44° and 64° E in longitude. The region has an extremely arid, arid, or semi-arid climate29, but there are some subtropical regions near the Caspian coast in the north end29. The country has more than 90 million inhabitants overall, with over 20 million of them residing in the provinces of Tehran and Alborz.

In this study, a comprehensive sampling campaign was performed to measure water-stable isotopes in 10 main aquifers in Iran, namely, the Tehran-Karaj, Varamin, Shiraz, Darab, Isfahan, Kashan, Tabriz, Marand, Fareman, and Mashhad aquifers (Fig. 1). According to the classification of Iran’s climate zone by Najafi and Alizade30, the designated aquifers are located in five distinct climate zones: warm and semi-arid (Mashhad and Fariman aquifers), cold and temperate semi-arid (Tabriz and Marand aquifers), temperate and humid (Tehran-Karaj and Varamin aquifers), hot and extremely dry (Isfahan and Kashan aquifers), and cold and dry (Shiraz and Darab aquifers). Additionally, these aquifers are located in the five populated provinces of Tehran, East Azarbaijan, Razavi Khorasan, Isfahan, and Fars, and, as the country’s most significant aquifers, supply the water needs for over half of Iran’s population. As a result, the findings of this study could be crucial in laying the groundwork for the long-term viability, well-being, and public health of the end-users of groundwater resources. Details on the studied aquifers, including their type (confined or unconfined), lithology, recharge area, annual recharge rate, and extraction pressure, are summarized in Table 1.

Fig. 1
Fig. 1
Full size image

Location of the studied aquifers in Iran, showing the names of each aquifer and the spatial distribution of sampling points within them.

Table 1 Hydrogeological characteristics of the studied aquifers, including type (confined or unconfined), lithology, recharge area, annual recharge rate, and extraction pressure.
Full size table

Sample collection

The number of samples taken in the aquifers varied from 12 (in Shiraz aquifer) to 35 (in Tehran-Karaj aquifer), determined based on several key factors: population density, aquifer area, groundwater flow paths, location of agricultural and industrial zones, accessibility of transportation routes, and socioeconomic considerations. In total, we collected 202 samples from the 10 designated aquifers from different groundwater extraction points, namely, semi-deep wells (<25 m), deep wells (>25 m), qanats, and springs. Sampling was performed in October 2023, which is the dry season for groundwater in Iran31,32, to ensure that the collected groundwater samples reflect the post-recharge isotopic composition with minimal interference from recent rainfall events. The sampling points in each aquifer were chosen to be evenly distributed throughout the aquifer. Detailed information about the number of samples and the spatial distribution of the sampling points are listed in Table 2 and Fig. 1. Geographic coordinate, elevation, and the area of each aquifer were obtained from the Iran Water Resources Management Company (IWRMC)33.

Table 2 Number of groundwater samples and the corresponding surface area (in hectares) of each designated aquifer.
Full size table

In this study, 100 cm3 gray glass containers were used to collect the groundwater samples. Before the sampling campaign, we sterilized the glass containers using sulfuric acid ~98% manufactured by Merck Company, Germany (CAS-No.7664-93-9). Groundwater samples were collected using different methods depending on the source type. Specifically, samples from wells were collected using electric pumps, whereas samples from qanats and springs were gathered directly at the locations where water naturally flows to the Earth’s surface. Sampling was feasible in most of the designated points; however, in some wells where the pump was inactive and could not be turned on, attempts were made to collect samples from the nearest adjacent well. Additionally, for the wells where the pump motor was inactive, the motor was first turned on, and eventually, once the flow was clear and free of sediment, sampling was conducted. Finally, we stored all the collected samples at a temperature below 4 °C in a cool box and transferred them to the authorized laboratory for further investigations.

δ2H and δ18O measurements

Determining the isotopic composition of liquid water through conventional methods, such as isotope ratio mass spectrometry (IRMS), is time-consuming and costly. However, advancements in laser-based isotope analyzers such as off-axis integrated cavity output spectroscopy (OA-ICOS) have expedited the analysis of stable water isotopes, making it significantly cost-effective, accurate (±0.08‰ for δ18O and ±0.1‰ for δ2H), and accessible34,35. Here, we utilized the laser-based Los Gatos Research (LGR) Liquid Water Isotope Analyzer (LWIA) to analyze the groundwater samples. The LGR provides measurements of δ18O and δ2H of liquid water and separates vapor samples with unsurpassed performance. For the best stability and accuracy, the LGR uses the LWIA in the accumulated enhanced performance package36. In analyzing using the LGR (here, LGR DLT-100 instrument), we followed the International Atomic Energy Agency (IAEA) standards37. Accordingly, five LGR working standards (LGR#1–LGR#5) with the three IAEA standards (i.e., Standard Light Antarctic Precipitation-SLAP26, Greenland Ice Sheet Precipitation-GISP38, and Vienna Standard Mean Ocean Water-VSMOW39) were used for LGR DLT-100 calibration (Table 3).

Table 3 Measurement accuracy of deuterium (δ2H) and oxygen isotope (δ18O) for two standards of International Atomic Energy Agency (IAEA) (i.e., Vienna Standard Mean Ocean Water—VSMOW30 and Standard Light Antarctic Precipitation—SLAP17) and five commercial working standards available in the Los Gatos Research (LGR).
Full size table

Calculation of D-excess and LC-excess

By analyzing 400 water stable isotope samples, Dansgaard27 reported the relative relationship between δ2H and δ18O as

$${{\rm{\delta }}}^{2}{\rm{H}}=8{{\rm{\delta }}}^{18}{\rm{O}}+10$$
(1)

In Eq. (1), the intercept (i.e., 10‰) is D-excess. This equation treats the average relative humidity of the world’s oceans as 85%. As D-excess is challenging to utilize for evaporation enrichment, we calculated the LC-excess from Eq. (2)40:

$${\rm{LC}}-{\rm{excess}}=[{{\rm{\delta }}}^{2}{\rm{H}}-a{{\rm{\delta }}}^{18}{\rm{O}}-b]/S$$
(2)

where,

$${\rm{S}}={[{({{{\rm{\delta }}}^{2}{\rm{H}}}_{{\rm{analytical\; error}}})}^{2}+{({a{{\rm{\delta }}}^{18}{\rm{O}}}_{{\rm{analytical\; error}}})}^{2}]}^{0.5}$$
(3)

In Eqs. (2, 3), a and b are the slope and intercept of the local meteoric water line (LMWL) for each aquifer, and S is a standard deviation measurement uncertainty for both δ18O and δ2H3940. Moreover, the analytical errors comprise the uncertainties in the analyzer for δ18O and δ2H. Here, we assumed conservative measurement uncertainties for the S calculation as ±0.2‰ for δ18O and ±1.0‰ for δ2H41. When LC-excess in groundwater samples is less than zero, the feeding water on the earth surface has evaporated. The average LC-excess of the global groundwater samples was reported as –1.8‰41.

Unit adjustment

LGR DLT-100 gives δ2H and δ18O concentrations as units per million (‰). Meanwhile, the δ2H and δ18O concentrations in water resources studies are given as δ(‰). To make our results comparable with other related conducted studies, we change ‰ to δ(‰) using Eq. (4).

$${{\rm{\delta }}}_{{\rm{sample}}}\left({\rm{\textperthousand }}\right)=\left(\frac{{{\rm{R}}}_{{\rm{sample}}}-{{\rm{R}}}_{{\rm{standard}}}}{{{\rm{R}}}_{{\rm{standard}}}}\right)\times 1000$$
(4)

where, Rsample represents δ2H and δ18O concentrations in our groundwater samples and Rstandard indicates the δ2H and δ18O concentrations in standard values42.

Local meteoric water line (LMWL)

Precipitation is classified into rain caused by nonevaporative condensation and precipitation that evaporates43. The global precipitation isotopic line includes both forms of precipitation, whereas the LMWL only includes nonevaporative precipitation and represents D-excess in the condensation process. Considering that designated aquifers are located in different climate zones, we calculated the equation of LMWLs for each aquifer. Additionally, we calculated the LMWL equation using the 202 samples, representing the LMWL for Iran’s aquifers.

Data Records

We archived the raw isotope data measured in 10 main aquifers in a single table containing 203 rows and 13 columns. This dataset is freely accessible through https://doi.org/10.5281/zenodo.1687065844. In this dataset, the sampling points are represented by the rows, and isotopic measurements are represented by the columns, which include δ2H, δ18O, D-excess, and LC-excess. The columns also represent the geographic characteristics of the sampling points (i.e., latitude, longitude, and elevation).

δ2H in the designated aquifers varied from –79.21 to –27.81 (‰), with a standard deviation of 11.54 (‰). The values of δ18O ranged from –11.95 to –4.68 (‰), with a standard deviation of 1.48 (‰). D-excess and LC-excess varied from –5.99 to 24.26 (‰) and –6.12 to 11.20 (‰), respectively. The standard deviation values for D-excess and LC-excess were 5.52 and 2.66 (‰), respectively (Table 4). As shown in Table 4, the maximum values of δ2H and δ18O were observed in the Isfahan aquifer. The lowest LC-excess values were observed in the Mashhad aquifer. Detailed statistical characteristics of the δ2H, δ18O, D-excess, and LC-excess for each aquifer are given in Table 4.

Table 4 Descriptive statistics for deuterium (δ2H), oxygen isotope (δ18O), deuterium excess (D-excess), and the water isotope data normalized to precipitation feeds by line-conditioned excess (LC-excess) in 10 Iranian main aquifers.
Full size table

LMWL for the studied aquifers

The calculated equations of LMWLs for the studied aquifer are given in Table 5. The relative position of the LMWL for each aquifer to the GMWL is also shown in Fig. 2. Our results revealed that the most similar LMWL to the GMWL corresponds to the Isfahan aquifer whereas the maximum difference between the LMWL and the GMWL is from the Fariman aquifer. Equation (5) is the LMWL equation using the 202 samples, representing the LMWL for Iran’s aquifers. This line is closely aligned with the GMWL equation, namely Eq. (5) (Fig. 2):

$${\Delta }^{2}{\rm{H}}=6.92{{\rm{\delta }}}^{2}{\rm{O}}+2.69\left({{\rm{R}}}^{2}=0.79\right)$$
(5)
Fig. 2
Fig. 2
Full size image

Isotopic values of δ²H vs δ18O in groundwater samples for Tehran-Karaj aquifer, Varamin aquifer, Isfahan aquifer, Shiraz aquifer, Darab aquifer, Kashan aquifer, Tabriz aquifer, Marand aquifer, Mashhad aquifer, Fariman aquifer, and all 202 groundwater samples.

Table 5 Calculated equations of local meteoric water lines (LMWLs) for 10 Iranian main aquifers.
Full size table

Technical Validation

We used the five working standards of LGR with known isotopic compositions (i.e., LGR#1–LGR#5 in Table 3) to ensure the quality control/assurance of the analysis of our groundwater samples. Accordingly, δ2H and δ18O were varied from –161.3 to –10.5‰ and –20.72 to –3‰, respectively (Table 3). We also normalized all isotope ratios using three IAEA standards (i.e., SLAP, VSMOW, and GISP). Our findings showed that the analytical precision for δ2H and δ18O in groundwater samples are ±0.08‰ and ±0.05‰, respectively, both of which are more accurate than those specified in the related standards (see Table 2). Furthermore, we compared the precisions of δ2H and δ18O measurements for our groundwater samples with those from related studies conducted in Iran. Our results were consistent with the findings of Heydarizad et al.19, Daneshian et al.20, and Osati et al.21.

Usage Notes

The compiled stable water isotope dataset in Iranian aquifers has been saved in CSV format to facilitate easy access from a wide range of computer software. This format allows users to analyze and process the data easily without requiring intensive computational resources.

Our compiled dataset represents a significant advancement in hydrogeological studies for the main aquifers in Iran. The effectiveness of these data depends on their application and can elucidate the physical mechanisms driving the changes in the δ2H and δ18O compositions of oxygen and hydrogen. Although the dataset is highly useful for tracing atmospheric and hydrological processes, a limitation warrants acknowledgment to ensure its appropriate interpretation and application. The current dataset represents a snapshot in time and therefore has limitations for long-term temporal modeling. Groundwater isotopes often reflect multi-seasonal or long-term processes. Therefore, a single-timepoint dataset may not sufficiently capture seasonal dynamics, and recharge temporal variability. However, even with this limited temporal scope, the isotope dataset can serve as a valuable input for regional climate and hydrological models in several ways. First, the spatial distribution of δ18O and δ²H values across major aquifers provides a baseline for model initialization and spatial calibration. For example, these data can help constrain the representation of evaporation, precipitation, and moisture recycling processes in isotope-enabled regional climate models (e.g.,Weather Research and Forecasting model). Second, the isotopic signatures can be used to validate modeled isotope outputs during similar seasonal conditions, improving the representation of water source attribution and vapor transport pathways. Finally, the dataset supports hybrid approaches that combine stable isotope data with geostatistical or machine learning models to spatially estimate isotope patterns in unsampled areas. These applications demonstrate the potential of the dataset to inform both scientific investigations and evidence-based water policy in arid and semi-arid environments.