Spatiotemporal changes in Iranian rivers’ discharge

Iranian rivers have witnessed several extensive floods/droughts in the past few decades (Modarres et al., 2016; Hooshyaripor et al., 2022). For example, destructive floods occurred during the winter and spring of 2019, causing a national emergency to be announced in 40% of the country and resulting in 76 casualties. At the other extreme, a 50% decline in precipitation in 2017 created an acute drought that negatively impacted around 90% of Iran (Darand and Sohrabi, 2018).

In recent decades, changes in Iranian rivers' discharge have been detected and attributed to climate change (Zarghami et al., 2011; Fazel et al., 2017) and, in particular, to extensive changes in land-use patterns and agricultural water-use (Abghari et al., 2013; Aadpour et al., 2020; Shirmohammadi et al., 2020; Noori et al., 2021a; Noori et al., 2021b; Sharifi et al., 2021). Water-use patterns have changed as a result of construction of large dams, with 647 dams currently in operation and 146 dams under construction. Construction of an additional 537 dams across the country is planned (Iran Water Resources Management Company [IWRMC], 2020). Therefore, Iran’s river flows are now strongly regulated/fragmented with the aim of enhancing regional economies and securing the country’s independence in food production.

The severity and frequency of extensive floods/droughts are expected to increase as a consequence of climate variability and (water)land-use changes in Iran, but the impacts are expected to be unequally distributed across the country (Noori et al., 2009; Vaghefi et al., 2019). Therefore, a good understanding of the patterns and trends in Iranian rivers’ discharge is fundamental to decrease the extensive pressures on Iran’s surface fresh-water resources and to help plan water management strategies, particularly with respect to irrigation water supply.

Some recent studies have investigated potential river flow trends in individual subbasins in Iran (Masih et al., 2011; Zarghami et al., 2011; Abghari et al., 2013; Azari et al., 2016; Fazel et al., 2017; Zamani et al., 2017). However, generalization of findings from these studies to national scale is difficult, as the studies include regional differences and cover different periods. To our knowledge, trends in river flow at national scale in Iran remain largely unclear, despite good coverage of river flow at multiple monitoring stations. To address this gap, trends in Iranian rivers’ discharge during recent decades are analyzed in our present study. Knowledge of the rate of changes (RoCs) in the country’s river flows is imperative, since surface water are suppling approximately 5 × 10⁹ cubic meters (km³) of water to about 3 × 10⁶ ha of Iranian wetlands, providing around 44% of the water needs in urban areas and the agricultural sector, and recharging up to 10 km³ of groundwater resources in Iran (Madani, 2014). The analysis used in our study is based on monthly streamflow observations in all basins in Iran (30 basins), obtained at 139 stream gauging stations that mostly have more than 30 years of records. Spatiotemporal trend analyses are performed on monthly, seasonal, and annual timescales to determine past streamflow variations across the country. In addition, the RoCs in precipitation data, as the main natural driver of the changes in river flow, measured at 96 rain-gauge stations across the country are analyzed. The overall aim of the study is to picture out changes in Iranian rivers’ discharge for sustainable management of water resources.

The study area covers all of Iran’s territory. Iran is located in the world’s dry belt, mainly covered by an (semi)arid climate, with mean annual precipitation of around 250 mm (less than one-third of the global average). The overall morphology of the country in western and northern parts consists of numerous mountains, while eastern and central parts mostly consist of plains with uniform terrain. The presence of the mountains has led to the formation of diverse climates, so that Iran experiences both an arid climate (east region) and a Mediterranean climate (coast of the Caspian Sea at the north) at the same time of year. These physical conditions have led to an uneven distribution of rainfall in the country, whereby around 70% of Iran’s surface area is arid and receives only about 31% of total rainfall (Figure S1). Annual potential evaporation ranges from 500 mm in the north-west to 3,750 mm in southern basins of the country. Mean annual temperature varies between 0°C in the north and 28°C in the south of the country. Mean annual rainfall varies from 50 mm in the center and east to 1,800 mm in the northern regions of Iran (Figure S2). Long-term mean annual runoff in Iran is about 400 km³, with approximately 270 and 130 km³ of evapotranspiration and renewable water capacity, respectively (Some’e et al., 2012).

During the last decades, surface water resource planning and management in Iran has been based on political rather than hydrological units. In hydrological terms, the country contains 6 primary hydrological basins and 30 secondary hydrological basins (hereafter referred to as “basin”) (Figure S3) equipped with more than 1,730 active gauging stations. Iran’s water resources monitoring network is relatively strong. The first hydrometric stations in Iran were established in the 1940s, in the vicinity of the capital city of Tehran and in the south-west of the country. Currently, around 20% of all stations have a service life of more than 40 years (Figure S4). Approximately 34% of the length of Iran’s land border with neighboring countries is composed of rivers (26 rivers), through which part of surface runoff enters or leaves the country.

Monthly streamflow measurements were used in this study. Also, seasonal and annual mean streamflow data were generated from the monthly mean measurements in each hydrometric station. Monthly mean streamflow data from 1,730 active hydrometric stations were obtained from the IWRMC and filtered based on the quantity and quality of measurements. Finally, 139 data-rich stations with proven data quality were selected for further analysis (Figure 1). The selected stations were distributed across different hydro-geomorphological units and cover around 97% of the country’s rivers with more than 30 years of observations. Given the smaller number of perennial rivers in the center, east and south-east of the country, a few hydrometric stations are available in these regions. Table S1 summarizes the statistical characteristics of the hydrometric stations.

We also investigated the RoCs in mean monthly, seasonal, and annual precipitation as the main natural driver of changes in river flow regime, across Iran. In this regard, information of 635 rain-gauge stations was obtained from the IWRMC. Then, the rain-gauge stations were filtered based on the quantity and quality of measurements. In general, life span of rain-gauge stations is less than that of the hydrometric stations. For example, historical data measurements in the oldest rain-gauge station dates back to 1946 while more than 54 out of 139 hydrometric stations were active even before 1946. Another criterion for selection of rain-gauge stations was their proximity to the selected 139 hydrometric stations. Given such criteria (quantity and quality of measurements, and proximity to hydrometric stations), finally 96 data-rich rain-gauge stations (Figure 1) were selected for further analysis.

The trend analysis methodology used in the present study is illustrated in Figure S5. Several statistical trend analysis techniques have been successfully used in the field of hydrology (Nourani et al., 2018; Mudelsee, 2019). Both accuracy and simplicity of the trend analysis techniques are important particularly when trend analysis is aimed to apply for big datasets in relatively large geographical scales, such as entire Iran. To select a parsimonious method, we first made a trade-off between the accuracy and simplicity of some trend analysis methods available in the literature. Then, the method that best fits the aims and scope of our study was selected. We found although the old methods (e.g., Mann, 1945; Sen, 1968; Kendall, 1975) can provide beneficial information about the possible RoCs in our datasets, but they have some disadvantages that may hinder the detection of true changes in the data. In addition, some of newly suggested hybrid trend analysis methods are so complicated (Mudelsee, 2019) that may not be simply applicable for our study that includes big datasets. Thus, we selected a relatively new method suggested by Pinhas et al. (2012) to report the possible RoCs in streamflow and precipitation time series across Iran. The method is based on a linear regression analysis followed by analysis of variance (ANOVA). The methodology has the advantage of simplicity, while providing the necessary information in terms of slope and variability, making it useful in practice for analysis of big datasets.

We determined the border of subbasins for individual hydrometric stations in the ArcSWAT environment (Figure 2). Detailed information about the 139 subbasins corresponding to hydrometric stations is given in Table S2. We normalized the streamflow data by diving by the corresponding subbasin area (mm/yr) to better interpret the possible trends across small to large rivers. To apply the used trend analysis’s methodology in our large database, a code was first assigned to each station and a number to its corresponding subbasin to better distinguish the obtained results for different hydrometric stations. The entire dataset was then categorized using the Pivot Table in Microsoft Excel software, based on the assigned codes and subbasin numbers. Linear regression and ANOVA F-test were performed using the analysis Tool Pak in Microsoft Excel. The null hypothesis in this approach is that all coefficients of each estimated regression are equal to zero. Therefore, rejection of the null hypothesis means that the coefficients are not equal to zero, that is, the estimated regressions are meaningful. This happens when the estimated significance level for the F-statistics is <0.1 (90% confidence level).

Figure 3 shows the mean, maximum, and minimum annual streamflow and indicates how their coefficient of variation (CV) varied across Iran (the greater the diameter of a circle, the higher value of the index). The south-western and western rivers flowing into the Persian Gulf and the Caspian Sea had the highest annual streamflow among all the rivers. These parts contain most of Iranian permanent rivers. Contrariwise, the lowest annual streamflow (approximately 0) was observed in the southern and eastern water basins, where the intermittent streams in the country are dominant. These intermittent streams play a key role in human life and ecosystem sustainability by recharging aquifers and nourishing lakes, wetlands, marshes, and dam reservoirs in the arid regions (Molle et al., 2009). The minimum CV was for annual river discharges in the Caspian Sea region, where a Mediterranean climate prevails. For example, in this region, stations #19051 and #18106 had CV of 21.1% and 22.3%, respectively (Table S1). The highest values of CV, representing high internal variability, were found in the east, north-east, and central parts of Iran, where the climate is classified as (semi)arid. Stations #41243 and #16079 in this zone had CV of 586.4% and 254.2%, respectively (Table S1).

Figure 4 shows the area-normalized value of trends (mm/yr/yr) computed on analyzing the streamflow data at annual timescale. The results revealed negative and positive trends at 78 (approximately 56%) and 12 (approximately 9%) stream gauging stations investigated, and null trends at 49 stations (approximately 35%) (P value < 0.1). On global scale, around 22% of the world’s rivers are reported to be showing significant decreasing trends in annual discharge and 9% are showing increasing trends (Walling and Fang, 2003). Thus, the number of rivers showing a downward trend in annual discharge is 2.5 times greater in Iran than in the rest of the world.

Figure 5A shows times series of the country’s annual mean streamflow and precipitation normalized between 0 and 1. Annual mean streamflow and precipitation were calculated based on the historical data at 139 and 96 hydrometric and rain-gauge stations, respectively. According to this figure, both streamflow and precipitation show a statistically significant decreasing trend (P value < 0.1) in Iran. However, the RoC in the streamflow is 3.2 times greater than that in precipitation. This finding denotes other factors such as man-made activities may dominantly contribute to severe decline in the country’s annual streamflow. Construction of large dams to secure water/food/energy for the expanding population largely regulates/fragments Iranian rivers’ discharge. The number of large dams and the reservoir’s storage capacity constructed by the Iran’s Ministry of Energy have increased rapidly from the late of 1970s to 2017 (Figure 5B). Many small and diversion dams have been also constructed by the Iran’s Ministry of Agriculture Jihad (the main governmental organization responsible for agricultural crops and food-security) during the last decades. In general, Iran currently has 647 dams in operation that strongly regulate/fragment the river flows with the aim of securing the country’s independence in food production and enhancing regional economies (Maghrebi et al., 2020). Meanwhile, more than 50% of the total capacity of Iran’s dams was remained unfilled during 2 last decades (Iranian Parliament Research Center, 2017). While we did not investigate the impact of other possible drivers such as evapotranspiration on the change in Iranian river flows, we hypothesise that an increased potential evapotranspiration could have contributed to a decreased streamflow in Iran, as reported in some basins in the world (Miller et al., 2011). Therefore, further research is required to show the possible impact of change in potential evapotranspiration on the decreased streamflow in Iran.

Extensive spatiotemporal variability in surface water resources availability has resulted an increasing reliance on groundwater sources in Iran (especially in southern, central, and eastern parts) and consequently, an acute drop in groundwater table up to 20 m averaged across the country during the last 5 decades (Nabavi, 2018; Noori et al., 2021c). Given the surface water–groundwater interactions, such severe decline in groundwater table can change the role of groundwater from a “recharger” to a “discharger” for surface water sources (Jasechko et al., 2021), leading to an accelerated decline in Iran’s river flows. Overall, such intensive growth in river regulation/fragmentation and groundwater overabstraction have led to drying national inland lakes, wetlands, and rivers (Aghakouchak et al., 2015; Ashraf et al., 2019; Noori et al., 2021c). According to the former Deputy Director of Planning for Water and Water Resources at the Iranian Ministry of Energy, Hedayat Fahmi, by 2018 about 20% of Iran’s perennial rivers, such as the Zayandeh-Rud river, had transformed into intermittent (seasonal) rivers and most former seasonal rivers had dried up or become narrow streams. This is a result of the nationwide unsustainable development programs that aim at progress in the economy, infrastructure, and the agriculture-sector, regardless of the country’s renewable fresh-water and environment-ecosystem resilience (Madani, 2014; Madani et al., 2016). However, there was no data accessible regarding the amount of groundwater recharge/discharge into the surface water resources in the country scale to quantify the exact share of groundwater on the decline in the river flows.

In the present analysis, the largest positive change in annual discharge was seen at station #21237 (+14.10 mm/yr/yr) and the largest decline at station #21191 (−78.27 mm/yr/yr). A downward trend in annual flow was observed in streams located in the central part of Iran, highlighting the overall decrease in the streamflow availability in this region. Man-made impacts of development, such as extensive inter/trans-basin water diversion policies, expanding the agricultural lands, and overuse of aquifers are the main reasons for the decline in river flows in central Iran (Madani, 2014). According to the literature, there was also a downward trend in annual flow of streams that are the main source of national inland lakes (Lakes Urmia, Zarivar, Parishan, Shadegan, Namak, Bakhtegan, and Maharlu), all of which are suffering from a continuous decline in their water level (Aghakouchak et al., 2015; Khazaei et al., 2019; Vaghefi et al., 2019; Nodefarahani et al., 2020; Malekmohammadi et al., 2023). Therefore, our results are consistent with fact that Iranian lakes/wetlands are drying up. This must be carefully taken by Iran’s water resources authorities into consideration. The greatest decrease in annual flow was observed in streams located in the north-west, near the Caspian Sea, and in the south-west, near the Persian Gulf, which constitute the main streamflow into these water bodies (Figure 4). This can add to environmental problems in these regions, as rivers play a pivotal role in balancing water salinity and transporting organic matter, and nutrients (Maghrebi et al., 2018). Large dams and reservoirs impounded along perennial rivers have contributed dramatically to the decrease in annual discharge in these regions (Torabi Haghighi et al., 2014). Simultaneously, climate change impacts in terms of decreasing precipitation in western Iran (Modarres and Sarhadi, 2009) have reduced annual discharge to the Caspian Sea and Persian Gulf (Akbari et al., 2020).

Most eastern and south-eastern transboundary rivers were also found to show a downward trend in annual flow (except for station #53013). In addition to the aforementioned factors, harnessing of rivers in neighboring countries, such as Afghanistan, for agriculture development and hydropower generation had also considerable influence on inflow from the transboundary rivers to Iran (Shahbazbegian et al., 2016). The decline in annual discharge in the eastern and south-eastern transboundary rivers has reduced crop yields (Siahsar et al., 2010), increased inland migration (Anvari et al., 2018), lowered the water level in Lake Hamun (Sharifikia, 2013), and more importantly, exacerbated international disputes between Iran and Afghanistan (Mianabadi et al., 2020). In north-east Iran, which is one of the country’s main agricultural hubs, a negative trend in river discharge has increased groundwater abstraction for agricultural development. This could be the reason for the high rate of groundwater loss in these areas reported by Ahmadi et al. (2015). Future flow manipulation in the Harirud river in Afghanistan would add to the negative trend in annual discharge in north-east Iran. In north-west Iran (Lake Urmia and Aras basins), all hydrometric stations investigated showed a negative or no trend in annual river discharge. According to a study by Ashraf et al. (2019), the Lake Urmia basin has experienced the greatest rate of decline in water storage in Iran, an effect dominated by anthropogenic activities. Therefore, impacts of human interventions and climate change are contributing to severe negative trends in river discharge in these regions (Hassanzadeh et al., 2012; Fazel et al., 2017). In the west of the country, where the rivers mostly flow out of Iran (Figure 6), a negative trend in river annual discharge was observed probably due to river flow regulation, as the region has the highest concentration of dams in Iran. This can pose a threat to downstream terrestrial and aquatic ecosystems and exacerbate the small-scale storms caused by drying up of small wetlands in Iran and Iraq (Richardson et al., 2005).

Projected change and variability in climate indices using climate models indicate a general decreasing trend in precipitation in Iran in the future (Ashraf et al., 2019; Vaghefi et al., 2019). Anthropogenic impacts of development and policies to achieve national independence in crop production via harnessing water resources would exacerbate the impacts of climate change on Iran’s rivers (Abghari et al., 2013). Therefore, the increased pressures on Iran’s water resources can be expected to continue or even accelerate in the future. Milly et al. (2005) found a decrease in projected runoff in the Middle East region, including Iran, by 2050. Therefore, a shift toward sustainable water and land management is required to mitigate the negative effects of water shortages across Iran.

Figure 7 shows the results of seasonal trend analysis performed on river flows at all 139 hydrometric stations across Iran. Also, the RoCs of streamflow averaged in each primary hydrological basins are given in Table 1. The results revealed that the seasonal mean river flow change in summer was −3.11 ± −0.72 mm/season/yr. The highest RoCs in summer season was observed at station #42009 located in Gavkhouni basin and the most negative at station #17041 located in the Sefid-rud River basin. Negative trend of river discharge in summer were observed at 60 stations (43%), positive rates at 10 stations (7%), and no significant change at 69 stations (50%) (Figure 7). All primary hydrological basins showed a negative RoC in streamflow, except the Eastern Boundary basin where no significant RoC was observed. At the same time, no trend was observed in seasonal precipitation in the primary hydrological basins (Table 1). In autumn, positive trends were seen at 17 stations (12.2%) and negative trends at 52 stations (37.4%) (Figure 7). Similar to summer, primary hydrological basins showed a negative RoC in streamflow (except the Eastern Boundary primary hydrological basin) while no trend was observed in precipitation in these basins (Table 1). The maximum, minimum, and mean trend in river discharge in autumn were 2.2 (station #21237), −0.6 (station #17043), and 0.001 ± 0.29 mm/season/yr, respectively. Regarding spring season, the maximum and minimum trend in streamflow observed at station #21237 located in the Great Karoon River basin and station #21191 located in Karkheh basin, respectively. The negative and positive trend of river discharge were observed at 53.9% (75 stations) and 4.3% (6 stations), respectively (Figure 7). The maximum decline in river’s discharge in this season took place at the southwest and north of the country, where the climate change has had a profound effect on snowmelt and river discharge capacity (Dodangeh et al., 2017). Primary hydrological basins showed a negative RoC in streamflow in spring (except the Eastern Boundary primary hydrological basin) while no trend was observed in precipitation in these basins (Table 1). In winter, the maximum, minimum, and mean trend of river flows were 4.9 (station #21237), −2.85 (station #22023), and −0.2 ± 0.9 mm/season/yr, respectively. In winter, 59 stations (42.4%) showed a significant decrease in river discharge and 12 stations (8.6%) showed an increase. On the contrary to other seasons, the negative trends observed in streamflow in 2 primary hydrological basins were consistent with those observed in precipitation (i.e., Caspian Sea, and Persian Gulf & Sea of Oman primary hydrological basins) (Table 1).

Based on the results, the highest number of stations with a significant positive trend and the lowest number with a significant negative trend in seasonal mean streamflow occurred in autumn. Water needs for agricultural crops and exploitation of surface water resources are low in autumn. Increasing river discharge in autumn at regional/country scale over multiple years has also been reported by Lins and Slack (1999) and Birsan et al. (2012). The greatest negative trend in river discharge was seen in spring. This is in line with previous findings for west regions of Iran, which show that rivers with snowmelt sources are experiencing a decreasing trend in spring discharge (Abghari et al., 2013). Other factors such as increasing water demand to support the doubled area in irrigation during the last decades (Maghrebi et al., 2020) can also contribute to the large decline in river discharge in spring, as a main cultivation season in Iran. In general, the RoC in seasonal mean streamflow calculated in primary hydrological basins were not consistent with the those corresponding to the seasonal mean precipitation (Table 1), highlighting few impacts of precipitation changes on the decline in Iran’s river flows. Therefore, the seasonal decline observed in streamflow in Iran is more impacted by anthropogenic activities than natural drivers. This is consistent with the anthropogenic decline in river discharge in the Middle East, with a reported median decrease in streamflow more than 15% (1971–2000) (Haddeland et al., 2014). With respect to winter season, the RoCs in streamflow were consistent with those corresponding to the precipitation in half of the primary hydrological basins, resulting the contribution of precipitation to decline in streamflow during winter.

Table 2 shows the RoCs in monthly streamflow averaged in each primary hydrological basin. A statistically significant negative RoC was observed in 5 (out of 6) primary hydrological basins from January to December, except Persian Gulf & Sea of Oman in December. Eastern Boundary primary hydrological basin showed no statistically significant trend in monthly mean streamflow in all months.

In general, almost all of RoCs observed in monthly mean precipitation were statistically nonsignificant (P value ≥ 0.1) (Table 2). The observed RoCs are not consistent with those corresponding to the monthly mean streamflow. Even, they were against each other in some of the basins. For example, our results showed a positive RoC in precipitation in Central Plateau primary hydrological basin during September and November, while the RoC in streamflow during these 2 months were negative. Same results were also observed in Qarehgom (in August) and Persian Gulf & Sea of Oman (in September) primary hydrological basins. These inconsistencies between the trends of precipitation and streamflow in the primary hydrological basins clearly indicate the dominant impacts of anthropogenic activates on the Iran’s river flows. More specifically, the adverse trend discovered in Central Plateau, Qarehghom, and Persian Gulf & Sea of Oman primary hydrological basins specifies the importance of anthropogenic drivers on streamflow in these regions. These 3 primary hydrological basins cover the areas with the most decline in groundwater table in Iran (Noori et al., 2021c). This condition can change the groundwater role from a “recharge” to a “discharger” of rivers, leading to reduce streamflow (Jasechko et al., 2021). With respect to Central Plateau primary hydrological basin, it accommodates more than 50% of the country’s population, lodges many of big industries, and is the country’s agricultural hub. In addition, this basin is a hotspot with respect to interbasin water transfer projects in Iran (Gohari et al., 2013; Zahraei et al., 2016; Roozbahani et al., 2020). All these anthropogenic stressors have limited surface water availability in these regions, leading to transformation of some perennial rivers to seasonal rivers (e.g., Zayandeh-Rud as one of the most water-rich river in Iran).

From the results of trend analysis, we found that areas adjacent to the south coastlines of the Caspian Sea had the largest number of stations with no significant trend in river discharge in all months. River flows at stations located in the Lake Urmia basin showed negative or no trend (except for Lighvan station, i.e., station #31019, in April, May, and September) as a result of dam construction and excessive agriculture development in this area. Hassanzadeh et al. (2012) found that declining inflow contributed about 65% to the drop in lake volume and water level. Considering the negligible share of groundwater feed to Lake Urmia (approximately 3%) (Vaheddoost and Aksoy, 2018), decreased streamflow would dramatically threaten the lake’s ecosystem. The decrease in river discharge in north-east Iran in the cropping season has increased utilization of groundwater resources, resulting in the greatest rate of drop in groundwater level in these months (Esmaeili-Vardanjani et al., 2015; Maghrebi et al., 2021; Noori et al., 2021c). The number of stations with positive, negative, or null trends, together with the rate of river discharge at these stations are given in Table 3. The greatest rate of decline in average discharge was seen in April and May. The highest number of stations with a positive (negative) significant trend in river discharge was seen in September and the lowest in May. The highest number of stations with a negative significant trend in river discharge was seen in June and the lowest in December. The number of stations with no significant trend varied from 53 to 79 stations in different months of the year.

Increasing demand for water to secure food and water for Iran’s rapidly growing population and extreme climate conditions have placed severe pressures on the availability of surface water resources in the country. This study investigated changes in river flow in Iran over time and space. On an annual scale, around 56% of the stations showed a decreasing trend in river discharge, with the proportion indicating a decline being 2.5 times the global average (about 22%). Such declines in Iran’s river flows pose serious threats to the country’s water and food security and its environmental sustainability. In addition, severe declines in river flows and shifts from perennial to intermittent rivers in some subbasins of Iran may indicate the development of new hydrological regimes, with significant implications for future surface water storage. In general, the declines in streamflow were not consistent with the changes obtained in precipitation, as the main natural driver of streamflow, concluding the dominant impact of anthropogenic pressures on Iranian rivers’ discharge. Considering the general decreasing trend in future precipitation projected for Iran by climate models and the country’s ambitious development plans, regardless of renewable water resources, the pressures on Iran’s water resources are expected to continue or even increase in the future. Therefore, prompt actions should be taken to mitigate the negative impacts of natural and more importantly man-made stressors on the country’s water resources.

The data used in this are publicly available via Data Archive of the Iran Water Resources Management Company stu.wrm.ir/login.asp.

The supplemental files for this article can be found as follows:

Table S1. Hydrometric stations used in the present study and their statistical characteristics (File Format: PDF).

Table S2. Detailed information about the 139 sub-basins corresponding to hydrometric stations.

Figure S1. Percentage of precipitation falling on Iran’s surface area (70% of surface area receives just 31% of total rainfall). (File Format: PDF).

Figure S2. Basins, selected hydrometric stations (orange triangles) and mean annual precipitation across Iran. (File Format: PDF).

Figure S3. The six primary hydrological basins and 30 secondary basins, and location of the selected hydrometric stations across Iran. (File Format: PDF).

Figure S4. Location of hydrometric stations (green dots) and selected stations (orange triangles) in Iran. (File Format: PDF).

Figure S5. Flowchart of the trend analysis’s methodology used in the present study. (File Format: PDF).

No funding was received for this paper.

The authors have declared that no competing interests exist.

Contributed to conception and design: MM, RN.

Contributed to acquisition of data: MM, RN, FD, HF, RR, HT, SMRAM.

Contributed to analysis and interpretation of data: MM, RN, ADM, RL.

Drafted the article: MM, RN, AA, BK.

How to cite this article: Maghrebi, M, Noori, R, Mehr, AD, Lak, R, Darougheh, F, Razmgir, R, Farnoush, H, Taherpour, H, Moghaddam, SMRA, Araghi, A, Kløve, B. 2023. Spatiotemporal changes in Iranian rivers’ discharge. Elementa: Science of the Anthropocene 11(1). DOI: https://doi.org/10.1525/elementa.2022.00002

Domain Editor-in-Chief: Steven Allison, University of California, Irvine, CA, USA

Associate Editor: Martha Sutula, Southern California Coastal Water Research Project, Costa Mesa, CA, USA

Knowledge Domain: Ecology and Earth Systems